Reference Manual: ADEPT Automatic Video Trackers Edition 4B

Reference Manual
ADEPT Automatic Video Trackers

Edition 4B
Document No: IN95-0005-01
Document History
Edition Date Description

2 January 2009
3 January 2010
4 June 2012
4A February 2014 Template and manual layout refreshed
4B January 2017 Rebranding only; no technical content change
2 ADEPT Automatic Video Trackers Reference Manual Document No: IN95-0005-01 Edition 4B
Table of Contents
1 • About This Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

1.1 Notices. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2 Further Information. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 Technical Support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4 Returns. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 • Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1 System Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 Hardware and Software Comparisons. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3 • Tracker Theory Of Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.1 Non-Operational State. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.2 Operational State. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.3 Acquisition Mode. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.4 Track Mode (Autotrack). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4 • Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.1 Centroid. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4.2 Edge. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.3 Multiple Target Track (MTT). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4.4 Phase Correlation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4.5 Combined. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.6 Scenelock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.7 Moving Target Detection (MTD). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4.8 Polarity Selection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.9 Track Algorithm Performance Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5 • Symbology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
5.1 Graphics. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
5.2 Text. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
6 • Track Error Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
7 • Sensor Compensation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
7.1 Sensor Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
7.2 Setting Up Boresight and FOV Compensation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
8 • Platform Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
8.1 Acquisition Mode Rate Demand. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
8.2 Track Mode Rate Demand. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
8.3 Platform Characteristics Compensation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
8.4 AVT Platform Tuning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
9 • Synthetic Test Target . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
9.1 Synthetic Target Configuration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
9.2 Example Trajectories. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
10 • Picture in Picture (PiP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
11 • Tracker Configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
12 • Tracker Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
13 • AVT Control Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
14 • WinMon Utility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
Document No: IN95-0005-01 Edition 4B Table of Contents 3

A • Data Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .72
A.1 Serial Interfaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .72
A.2 DPR Interface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .76
B • External Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .89
B.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .89
B.2 External Command Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .89
B.3 External Command Response Format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .90
B.4 Line of Sight Data Command . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .91
B.5 Component Interface Commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .92
C • User Symbology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .104
C.1 User Symbology Command Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
C.2 Sprite Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
C.3 Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
D • Command Reference Manuals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .129
E • Technical Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .130
E.1 Built In Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .134
1 • About This Manual
This manual describes procedures for using the ADEPT Automatic Video
Trackers, software versions, 4.6 (or higher).
1.1 Notices
This manual uses the following types of notice:
CAUTION
Cautions call attention to actions that may cause system damage or loss of data.
NOTE
Notes call attention to important features or instructions.
LINK
Links provide hyperlinks to documentation or other sites of interest. The purple link color may
also be used to indicate a link (or hyperlink) within a body of text or paragraph.
1.2 Further Information

1.2.1 Abaco Website
You can find information on Abaco products on the following website:
LINK
https://www.abaco.com
1.3 Technical Support

You can find technical assistance contact details on the website Embedded
Support page.
LINK
https://www.abaco.com/embedded-support
Abaco will log your query on the Technical Support database and allocate it a
unique Case number for use in any future correspondence.
Alternatively, you may also contact Abaco’s Technical Support via:
LINK
[email protected]
Document No: IN95-0005-01 Edition 4B About This Manual 5

1.4 Returns
If you need to return a product, there is a Return Materials Authorization (RMA)
request form available via the website Embedded Support page.
LINK
https://www.abaco.com/embedded-support
Do not return products without first contacting the Abaco Repairs facility.
2 • Introduction
2.1 System Overview

A complete electro‐optic tracking system consists of an imaging sensor (typically
TV or IR), which is mounted on a two‐axis servo platform and a tracker. The
Tracker controls the position of the platform based on the scene observed through
the imaging sensor.
In the case of a manual tracking system, an operator controls the position of the
platform (via a joystick, etc.) based on the image observed via a video monitor. As
a result, the operator attempts to drive the platform so that the object of interest is
fixed on the boresight.
When using an automatic tracking system, the operator is replaced by an
electronic system. This system processes the video images directly in real‐time to
establish the position of a designated object relative to the sensor boresight. This
error is then used to control the platform so that the platform and attached
sensors accurately follow the target.
2.1.1 Electro-Optical Tracking System

Figure 2-1 Typical Electro Optical Tracking System
CONTROL CONSOLE
SERVO PLATFORM
AND SENSORS
COMMANDS VIDEO
ANNOTATED VIDEO RATE DEMAND
AUTOMATIC VIDEO
TRACKER
Document No: IN95-0005-01 Edition 4B Introduction 7

2.1.2 AVT System Example
In this example, the AVT is designed for integration with a servo platform and
video sensor. A conceptual system is shown in the typical installation diagram
below.
Figure 2-2 Typical Installation
Video Camera
Video IN 1
PC running
AVT Control
Programme
Rate Demand Servo
Platform
Serial
COM Port
Channel A Video Out
POW E R
RE SET
Video Monitor
OCTEC ADEPT VIDEO TRACKER
PC
Electronic Chassis
Joystick
containing ADEPT Tracker
The AVT is used to automatically detect and track a single target in the video
picture from the sensor mounted on the servo platform and provide an output
demand. This controls the position of the servo platform so that the platform and
attached sensors follow the target.
IMP20 (image fusion add
ADEPT74
Hardware ADEPT60 ADEPT64(1) ADEPT74 ADEPT104 ADEPT105 AIM12 IMP16 on card for ADEPT104 &
+ mezz
AIM12)
Document No: IN95-0005-01 Edition 4B

Double Half Length Half Length PMC (requires
Physical Format Full length PCI PC-104 cPCI style 90 x 77mm 110mmx68mm
Eurocard PCIe PCIe A60 host)
PCIe x1 PCIe x1
Data Bus VME PCI PCI or ISA N/A N/A PCI N/A
(x4 connector) (x4 connector)
2 x RS232 2 x RS232 2 x RS232
RS232/422 Serial Links 4 4 4 4 8 8
2 x RS422 2 x RS422 2 x RS422
PMC Site 9 9 8 8 8 8 8 8 8
Analogue Video Inputs 4 2 0 1 2 2 2 2 3
Analogue Video Output (with
2 1 0 1 1 1 1 2 1 (from tracker)
symbology)
Raw Analogue Video Output
2 1 0 0 1 1 1 0 1
(no symbology)
Via Mezzanine Via Mezzanine
Digital Video Input 9(2) 8 DVI Camera Link (2) 8 (2) 8 8
2.2 Hardware and Software Comparisons
Digital Video Output 8 8 DVI 8 8 8 8 8 8

Analogue DC Input (control) 9 9 8 8 8 8 8 8 8
Analogue DC Output (control) 9 9 8 8 8 8 8 8 8
Discrete Outputs (control) 9 9 8 8 8 8 8 8 8
Conduction/convection cooling Both Convection Convection Convection Convection Convection Convection Convection Conduction
Power 15W 15W 17W 17W 5W 12W 5W 10W 16W (including tracker)
Software
Synthetic Target 9 9 9 9 9 9 9 8 8
Moving Target Detection 9 9 9 9 9 9 9 8 8
Picture-In-Picture 9 9 8 8 9 9 9 2 2
(3) (3) (3)
Image Stabilisation with IMP16 8 8 8 9 9 9 with A60 8
Image Fusion with IMP16 8 8 8 with IMP20 8 with IMP20 with A60 with A104/AIM12
Introduction 9
3 • Tracker Theory Of Operation
3.1 Non-Operational State

The AVT is considered to be in a Non‐Operational State when:
• Initializing
• Performing start‐up or requested Built In Test (BIT)
• Performing a configuration management activity involving a save to or
restore from non‐volatile memory
• A BIT failure has occurred
While in the Non Operational State:
• The AVT does not perform any image processing functions.
• The AVT does not accept any command messages
• The AVT does not update any on‐screen symbology except that required to
report the BIT result.
Initialize Mode AVT initialization is triggered after powering‐on the AVT. During initialization,

the AVT performs all actions required to prepare the AVT for operation. The
duration of initialization is less than 5 seconds. Following initialization, the AVT
enters BIT mode.
Built-in Test (BIT) The BIT mode is entered following initialization or by a BIT command message in

Operational Mode.
The BIT Failed mode is entered when a fault is detected during BIT mode or by
the Background Built‐in Test (BBIT). In this mode, no further action occurs other
than to report the fault and to continue to service the Serial and Bus Interfaces
(assuming the cause of the BIT fail does not prevent this).
Further details on BIT are described in Appendix E • Technical Notes.
3.2 Operational State

While in the Operational State:
• The AVT processes the currently selected video input to detect or track
objects of interest
• The AVT outputs status messages through the status interfaces
• The AVT accepts command messages through the command interfaces
• The AVT updates the on‐screen symbology
The AVT has two primary Operational modes:
• Acquisition mode
• Track mode (Autotrack)
3.3 Acquisition Mode
Introduction In Acquisition Mode, the AVT automatically detects and cues targets entering or
contained within a defined area (or window) of the video image. Once a target
has been detected, it is then possible to issue a Track command message to the
AVT, which will cause it to start automatic tracking of that target and to exit
Acquisition mode.
Acquisition Window The Acquisition Window is used to define the area of operation within the video

image. The Acquisition Window position and size are controlled by command.
The Acquisition Window can be nominally positioned to any location within the
Active Video Region. The minimum and maximum allowable sizes of the
acquisition window depend on the selected algorithm, the current video standard
and the ADEPT hardware being used.
NOTE
Refer to the Video Definitions section in the ADEPT AVT Hardware Reference Guides.
Acquisition The Acquisition Algorithm is selectable and described in more detail in Chapter 4.
Algorithm There are several types of algorithm; some can only identify a single target and
some may identify several possible targets.
Armed Mode It is possible to configure the AVT so that if a target is not present when the Track

command message is accepted, the AVT remains in the Acquisition Mode in an
‘Armed’ state so that when a target appears, Track Mode is automatically selected.
The ‘Armed’ state is indicated via the Acquisition Mode Status item and Status
Line symbology.
Reversionary Track The AVT can also be configured to attempt a reversionary track mode if an

acquisition target is not identified when the Track command message is accepted.
NOTE
Refer to Block 002 in the Command Reference.
3.4 Track Mode (Autotrack)

Introduction In Track Mode (or Autotrack), the AVT automatically tracks the object contained
within the tracking window. The AVT enters Track Mode following acceptance of
a Track command message during Acquisition Mode, providing a valid target has
been identified. Acquisition Mode provides target designation data, which allows
the track window position and size to be initialized to give a high probability of
the AVT successfully tracking the selected target. The tracking window position is
automatically controlled so that the object of interest remains in the tracking
window, and if possible the size is also automatically controlled (not all tracking
algorithms are able to self‐size the tracking window).
The AVT exits Track Mode and returns to Acquisition Mode:
• Following a command to stop tracking
• Following an automatically detected loss of lock (detected by the breaklock
function)
Document No: IN95-0005-01 Edition 4B Tracker Theory Of Operation 11

Track Window The track window defines the area of operation of the track algorithm in Track
Mode. The track window position is automatically controlled by the track
algorithm, and can be positioned to any location within the Active Video Region.
The minimum and maximum size of the track window depends on the selected
track algorithm, the current video standard and the ADEPT hardware being used.
The track window size can be controlled manually or, for some algorithms,
automatically.
NOTE
Initial Track When automatic track window size is selected, the track window size is initialized

Window Size based on the detected target size from the acquisition mode. The ratio of
acquisition target size to initial track window size is configurable.
Breaklock In Track Mode, the Breaklock function continually monitors tracking parameters
Detection (Coast) for consistency, since variations are good indications of target obscuration, clutter
or noise, which can cause the AVT to lose track. Detection of excessive variations
in the tracking parameters triggers coast actions within the track algorithm and
platform filters to improve the possibility of tracking through the difficulty. The
operation of the coast action depends on the selected track algorithm and may be
separately configured for each algorithm.
3.4.1 Target Tracking Modes

Track Mode has the following main sub‐modes:
Verify Mode Following the command to Track, a brief Verify mode is entered, during which
the track algorithm attempts to establish track on a target. If track has been
established on the target, then Lock mode is entered. If the AVT fails to establish a
track, then the AVT returns to Acquisition mode.
Lock Mode In Lock mode, the AVT automatically tracks the target contained within the track

window. The AVT remains in Lock mode as long as it can maintain track on the
target.
Breaklock Stage 1 Following track loss detected by the breaklock algorithm, the AVT initially enters

Breaklock Stage 1.
Breaklock Stage 2 Following a failure to re‐acquire the target in Breaklock Stage 1, the AVT enters

Breaklock Stage 2, where the track window is positioned on the boresight. If
target re‐acquisition is not successful in Breaklock Stage 2, the AVT returns to
Acquisition mode.
Forced Breaklock/ An operator can use the Force Coast command to force the tracker into breaklock.

Coast This can be useful when the operator observes an impending condition in the
image that may cause problems to the tracking.
NOTE
Refer to Block 003, 004, 005, 007, 013, 015, 016 in the Command Reference.
4 • Algorithms
4.1 Centroid
Application The Centroid algorithm can be used for tracking bounded objects such as an
aircraft, which can be fully contained within the track window.
Aimpoint The Centroid algorithm finds the centre of energy of the target within the current
processing window, which becomes the target Aimpoint.
Figure 4-1 Centroid Position Movement
Polarity Pre- Since the algorithm calculates the centre of energy of the target intensity, it ideally

Processor requires an image where the target intensity is high and the background intensity
is suppressed. To achieve this, an image pre‐processor is used based on the
commanded Polarity.
Target Size The Centroid algorithm is able to estimate target size. When tracking, this

information can be used to adjust the size of the track window, and can be
monitored within the Breaklock algorithm.
In Acquisition Mode When the acquisition algorithm is set to Centroid, the target position and size are

used as target designation data for the Track algorithm in Track Mode.
In Track Mode The Track Window is positioned according to the centre of energy of the target

and can be automatically sized to minimize the possibility of extraneous data
entering the track window.
Advantages • The Centroid algorithm always finds the centre of energy of the target image
(it is ‘self‐centring’). This attribute is particularly useful when the target is
first acquired by the tracker and also following a change in target shape
• The Centroid algorithm can follow highly dynamic targets. In this instance,
‘highly dynamic’ refers to targets whose acceleration or unpredictable
motion means the servo platform cannot follow the motion, which results in
target motion in the image
Document No: IN95-0005-01 Edition 4B Algorithms 13

• The dynamic ability of the Centroid algorithm comes from use of a track
window that is larger than target. This creates a margin around the target,
which should be large enough to ‘contain’ unpredictable target motion.
Track window size can be adjusted manually or automatically, but even in
the automatic mode, it is possible to specify the required target to window
size ratio
• Under normal circumstances, it is desirable to set the track window as small
as possible around the target (to minimize the risk of a decoy [target‐like
object] entering the track window), while keeping the window large enough
to contain the target motion. The adaptive window sizing algorithm
attempts to do this automatically
• The Centroid algorithm considers all objects within the track window to be
part of a single target. This gives robust tracking of fragmented targets
Disadvantages • The Centroid algorithm requires a pre‐processor that yields an enhanced
image with a light target on a black background. This requirement may not
be met if the target expands or accelerates rapidly, or if the target to
background contrast is insufficient, or if target‐like pixel intensities occur in
the background area of the image pre‐processor
NOTE
Refer to Block 004 the Command Reference.
4.2 Edge
Application The Edge algorithm can be used for tracking many types of object, but is most
effective when tracking objects that have one or more defined edges with the
remaining edges unbounded. Typically this would be a rocket, which has a well‐
defined leading edge, but an uncertain extended trailing edge generated by the
plume.
Aimpoint The Edge algorithm finds the specified edge of the target within the current
processing window, which becomes the target Aimpoint. The Edge algorithm
uses the Target Bias command to select which edge of the target to report as the
Aimpoint.
Figure 4-2 Left Edge Track
50% Energy
Target Centreline Y
50% Energy
Target Left Edge
Polarity Pre- See Centroid.
Processor
Target Size The Edge algorithm is not able to estimate target size in the axis perpendicular to

the selected edge, hence during track, the window size in that axis is set to the
manually commanded track window size. To summarize:
• Single edge is selected:
Top or Bottom => x‐axis may be automatically sized
Right or Left => y‐axis may be automatically sized
• Dual edge (corner) is selected:
Top Right, Bottom Right, Bottom Left, Top Left => no axis is
automatically sized
In Acquisition Mode The Edge acquisition algorithm identifies the selected edge (or corner) of the
image contained within the Acquisition window. It provides the edge position
and, where possible, size as designation data. The Edge Acquisition algorithm
should be selected for use with Edge Track algorithm.
In Track Mode The Track Window is positioned according to the selected edge of the target. The

track window size cannot be automatically sized in both axes for reasons
described earlier. Usually, the Edge Acquisition algorithm is selected for use with
the Edge Track algorithm, since it ensures the track window is initialized on the
appropriate target edge when entering Track Mode.
Advantages • The Edge algorithm tracks a defined point on an unbounded target
• The Edge algorithm is able to follow highly dynamic targets
Disadvantages • The Edge algorithm requires a pre‐processor that yields an enhanced image
with a light target on a black background. This requirement may not be met
if the target expands or accelerates rapidly, or if the target to background
contrast is insufficient
NOTE

4.3 Multiple Target Track (MTT)
Application The MTT algorithm is used for tracking bounded objects such as an aircraft or
ground vehicles that are fully contained within the track window. The MTT gives
good performance with small targets in high clutter.
Aimpoint The MTT algorithm discriminates between separate objects within the current
processing window and identifies up to five of the objects. The target Aimpoint is
the centre of energy of the primary target.
Figure 4-3 MTT Track
Polarity Pre- Same as Centroid.

Processor
Target Size The MTT algorithm can estimate target size.
In Acquisition Mode When the MTT is selected as the Acquisition algorithm, it can be used to

simultaneously identify up to five potential objects within the acquisition
window. In this case, temporal track files are established on each of the objects
that pass configurable target discrimination criteria including size, aspect and
speed. Internally, the AVT prioritizes the track files and presents the highest
priority target as the primary acquisition target, and the remainder as secondary
acquisition targets. The prioritization can either be configured to be automatic
based on a characteristic such as size, position, speed etc., or prioritization can be
disabled, allowing the user to identify the highest priority target by cycling
through the identified targets or selecting a particular target ID. Should a track
command message be received by the AVT, it enters Track Mode on the primary
acquisition target.
In Track Mode On entry to track mode, the MTT focuses on the primary acquisition target. The

MTT target size information can be used to automatically adjust the size of the
track window and can be monitored within the Breaklock algorithm. The ratio of
the automatically‐sized track window to target size is much larger with the MTT
than other algorithms; this is possible because the MTT discriminates between
separate objects within the track window.
Advantages • The primary advantage of the MTT is the performance in clutter. By
maintaining track files on distinct objects, the MTT can tolerate other targets
within the Track Window providing they are not too close to the tracked
target
NOTE
The Centroid algorithm considers all objects within the track window to be part of a single target, and so
keeps the size of the window as close as possible to the size of the target.
• Using the MTT algorithm for target acquisition generally provides better
designation of the target for tracking and also means that a comparatively
large acquisition window can be used, which makes it easier to acquire
targets
• The MTT algorithm is ‘self centring’. This attribute is particularly useful
when the target is first acquired by the tracker and also following a change
in target shape
Disadvantages • The MTT tracks all objects separately. If a target becomes fragmented, the
performance of the tracking can therefore degrade if the tracker starts
tracking a part of the target instead of the whole target
NOTE
Refer to Block 007 and 008 in the Command Reference.
4.4 Phase Correlation

Application The Phase Correlation algorithm is a general purpose algorithm that can be used
for tracking both bounded and un‐bounded moving objects within the image. The
algorithm can give good results when tracking bounded moving targets in high
clutter such as tracking ground vehicles from an airborne platform.
Aimpoint The Phase Correlation algorithm computes the best match position between a
stored reference image (the target) and the current image local to the track
window. The best match position is the aimpoint.
Polarity Pre- The Phase Correlation algorithm does not use an image pre‐processor, so polarity

Processor is irrelevant to this algorithm.
Target Size The Phase Correlation algorithm does not produce an estimate of target size, so

automatic track window sizing is unavailable during Phase Correlation tracking.
However, as the reference image is continually refreshed, changes in the target are
reflected in the reference image. In practice, if an area of interest within the
reference image grows in size such that it exceeds the track window, structure
within the area of interest will be resolvable to provide a new area to track. At any
time however, even during tracking, the track window can be re‐sized manually.
In Acquisition Mode The Phase Correlation acquisition algorithm simply estimates the optimal

window size that should be used for tracking based on the current Acquisition
Window image. The target aimpoint is defined as the centre of the Acquisition
window.

In Track Mode The Phase Correlation Track algorithm uses the designation data to position the
Track window and, if automatic window sizing is enabled, to set the initial track
window size. The Phase Correlation algorithm cannot determine target size, so
the track window size remains fixed unless an external event changes it, e.g.
changing the track window size with automatic window size disabled.
The reference image is initialized when tracking is initiated, by storing the central
portion of the track window. On subsequent TV fields, when the match position
has been found, the reference image is partially refreshed using image data from
around the match position in the track window. The update rate is configurable.
This updating of the reference allows the Phase Correlation algorithm to cope
with changes in target due to range or orientation changes.
In Track Mode, it is possible to modify the aimpoint using the system joystick (or
other control) to ‘nudge’ the current aimpoint to a new point in the scene.
Advantages • The Phase Correlation algorithm is a general purpose algorithm that can be
used for tracking both bounded and un‐bounded moving objects within the
image
• The Phase Correlation algorithm is fairly tolerant of clutter within the track
window when the clutter is transient (i.e. passing through)
• The Phase Correlation algorithm does not require a pre‐processor that will
yield an enhanced image with a light target on a black background. The
algorithm functions by matching image profiles, so structure is required
within the reference image
Disadvantages • The Phase Correlation algorithm cannot self‐centre. As the target image
changes due to, for example, the target orientation changing, the best match
position is found, but this may not coincide with the best Aimpoint as
determined by the operator. Over a period of time, particularly with a
continually changing target, the Aimpoint may drift away from that desired
and may require manual correction
• The Phase Correlation match process occurs over the whole track window.
This means that if the track window contains both target and background
structure, one partial match will probably be obtained from the target, plus a
second partial match from the background. If the target is moving relative to
the background, the overall match position will therefore be a combination
of target and background motion. If possible, the track window should
contain only target structure
NOTE
4.5 Combined
Application The Combined algorithm is a general purpose algorithm that can be used for
tracking both bounded and un‐bounded moving objects within the image. The
algorithm can give good results when tracking bounded moving targets in high
clutter such as tracking ground vehicles from an airborne platform.
Aimpoint The Combined algorithm uses the results from two complementary track
algorithms running in parallel to produce a combined tracking algorithm.
Outputs from a Centroid type algorithm (one of Centroid or MTT) are combined
with outputs from a Correlation type algorithm (Phase Correlation only).
The constituent algorithms position their own track windows. The output from
the Correlation type algorithm is used as the target aim‐point for calculation of
Target to Boresight Errors. The constituent algorithms are re‐designated with a
position, velocity and size determined by the Combined algorithm if their outputs
are determined to have diverged from the Combined algorithm outputs.
Polarity Pre- Since the Centroid type algorithm calculates the centre of energy of the target

Processor intensity, it ideally requires an image where the target intensity is high and the
background intensity is suppressed. To achieve this, an image pre‐processor is
used based on the commanded Polarity.
The Correlation type algorithm does not use an image pre‐processor so polarity is
irrelevant to the Correlation type algorithm.
Target Size The Combined algorithm obtains target size information from the Centroid type

algorithm. This can be used to adjust the size of the Combined track window.
In Acquisition Mode Combined is not available as an Acquisition algorithm.
In Track Mode The Combined track algorithm behaves in a similar fashion to the Phase

Correlation algorithm except that it may automatically resize or re‐centre the
track window (depending on the particular target).
Advantages • The Combined algorithm is the most general purpose algorithm that can be
used for tracking both un‐bounded objects and bounded moving objects
within the image
• The Combined algorithm is very tolerant to clutter since the two track
algorithms are often complementary (so when one fails due to clutter, the
other may still be able to track)
• The Combined algorithm may self‐centre and self‐size
Disadvantages • Increased latency due to running two algorithms
• Self‐centring and self‐sizing can only occur if the target presents
characteristics that can be tracked by the Centroid Type algorithm (at least
some of the time)
NOTE

4.6 Scenelock
Application The Scenelock algorithm tracks a point in the overall scene. It is particularly
useful for tracking a point on the ground when observed from a moving airborne
platform.
Aimpoint The Scenelock algorithm tracks several points in the scene and then combines the
motion estimates of each point to estimate the overall scene motion. The
allocation and selection of tracking points in the scene occurs automatically. When
existing tracking points move outside of the field of view, new tracking points are
automatically identified and assigned.
The aimpoint is initialized to the point in the scene that is located at the boresight
when tracking is initiated. The Scenelock algorithm then continues to track and
report the position of this point in the scene. The symbology uses a cross to
indicate the Scenelock aimpoint.
Polarity Pre- The Scenelock algorithm does not use an image pre‐processor, so polarity is

Processor irrelevant to this algorithm.
Area of Interest The Scenelock algorithm operates over an area of interest within the image that

Size has configurable position and size. This area of interest is defined separately from
the normal acquisition/track window.
In Acquisition Mode The Scenelock algorithm automatically identifies and assigns tracking points. The

symbology uses a small rectangle at the boresight to indicate when it has
identified suitable tracking points. When no tracking points are detected, the
small rectangle is removed.
In Track Mode On entering track mode, the aimpoint is initially set to the point in the scene at the

boresight. The allocation and selection of tracking points in the scene continues to
occur automatically.
In Track Mode, it is possible to modify the aimpoint using the system joystick (or
other control) to ‘nudge’ the current aimpoint to a new point in the scene.
Advantages • The Scenelock algorithm offers an easy to use, robust algorithm for tracking
a point in the scene
• The Scenelock algorithm is unaffected by individual moving objects within
the scene as long as the moving objects are small with respect to the overall
scene
• The selection of tracking points is automatic
• As the aimpoint is calculated from the combination of all tracking points, it
is not necessary to have an actual object at the aimpoint that can be tracked
Disadvantages • The Scenelock algorithm cannot be used to track a specific object in the scene
unless the object is large and fills the majority of the scene
• The Scenelock algorithm cannot be used if there is insufficient structure in
the background to track, e.g. sea or air backgrounds
NOTE
4.7 Moving Target Detection (MTD)
Application The MTD algorithm is used for detecting multiple moving targets such as vehicles
or people. The MTD algorithm gives good performance with small targets in high
static clutter.
Aimpoint The MTD acquisition algorithm extracts objects from the scene that are moving
independently in relation to the background. This is achieved by tracking the
background motion using a Scenelock process, and aligning the background with
a reference image. Any differences between the reference image and the current
aligned image are assumed to be moving targets. The detection process can be
configured to detect only targets of a certain size, travelling at a certain speed, or
only targets that have moved by a particular distance.
Polarity Pre- The MTD algorithm operates on the differenced raw image, so polarity is

Processor irrelevant to this algorithm.
Target Size The MTD algorithm is able to estimate target size.
In Acquisition Mode The MTD algorithm can simultaneously identify up to five moving targets and

performs very similarly to the MTT algorithm except that it operates on a
difference image. The target position and size are used as target designation for
the Track algorithm in Track Mode.
In Track Mode MTD is not available as a Track algorithm.
Advantages • The MTD algorithm performs well in cluttered scenes
Disadvantages • The MTD algorithm cannot be used if there is insufficient structure in the
background for the Scenelock algorithm to track, e.g. sea or air backgrounds
• The MTD algorithm can produce false alarms particularly if extraneous
motion is exhibited by other artifacts in the scene (trees blown by the wind,
for example)
NOTE
Refer to Blocks 053 and 054 in the Command Reference.

4.8 Polarity Selection
Introduction Algorithms that find the centre of energy such as MTT, Centroid and Edge use an
image pre‐processor that attempts to intensify the contrast of the target, usually
by suppressing the background pixels within the window. These algorithms use
the Polarity setting to indicate whether they are looking for a bright, dark or
bipolar target to select an appropriate image pre‐processor. These algorithms do
not necessarily use the same pre‐processors.
Correlation and image differencing based algorithms do not require a pre‐
processor and so ignore the Polarity setting.
Threshold Pre- The majority of the pre‐processors used are threshold pre‐processors. This means

Processors a threshold level is set where pixel intensities on one side of the threshold are
suppressed (set to intensity 0) and pixel intensities on the other side of the
threshold are left unchanged (white) or inverted (black). The AVT provides the
following mechanisms for setting the threshold level in these cases:
• Manual: the threshold is set to a User defined level
• Manual Offset: the threshold is set to a User defined offset above an
automatically calculated background level
• Automatic: the threshold is automatically set
For most cases, Automatic control is suitable.
When a threshold pre‐processor is in use, the Threshold Gauge symbol can be
displayed (see Section 5 • Symbology ).
Statistical Pre- The statistical pre‐processor uses a distribution analysis to optimize the image

Processor and does not set a single threshold. For this reason, only the Automatic threshold
mode is available; Manual and Manual Offset threshold modes are ignored.
Summary
Algorithm Polarity Sensitive Pre-processor Mechanism
MTT Yes Threshold for all polarities
Centroid, Yes Threshold for white or black
Edge Statistical for bipolar
Phase Correlation, No N/A
Scenelock,
MTD
4.9 Track Algorithm Performance Summary
Table 4-1 Track Algorithm Performance Summary
Criteria Centroid Edge MTT Centroid Edge MTT Phase Combined Scenelock
Positive/ Positive/ Positive/ Bipolar Bipolar Bipolar Correlation
Negative Negative Negative
Polarity Polarity Polarity
Bounded 5 4 5 5 4 5 4 5 3
Targets
Unbounded 1 3 1 1 3 1 5 4 5
Targets
Dynamic 5 5 4 5 5 4 4 4 3
Performance
Self Centring 5 3 5 5 3 5 1 4 1
Decoy 4 3 5 3 3 4 4 5 4
Resistant
Pattern 1 1 1 1 1 1 5 3 5
Tracking
Drift Resistant 5 5 5 5 5 5 4 4 4
General 3 3 3 5 4 4 5 5 1
Purpose
Constant Polarity 5 5 5 3 3 5 3 3 3
1. = Not Applicable.
2. = Below Average.
3. = Average.
4. = Good.
5. = Excellent.
Acquisition-Track The following table suggests suitable pairings for Acquisition‐Track mode
Pairings algorithms:
Acquisition Mode Algorithm Track Mode Algorithm
Centroid Centroid
Edge Edge
MTT Centroid, MTT, Edge, Phase Correlation, Combined
MTD Centroid, MTT, Edge, Phase Correlation, Combined
(Phase Correlation or Combined would be preference)
Phase Correlation Phase Correlation, Combined
Scenelock Scenelock

Window Sizes The minimum and maximum sizes of the acquisition/track window depend on
the selected algorithm, the current video standard and the ADEPT hardware that
is being used. The acquisition window size is controlled by the Command. The
track window size can be configured to be controlled manually or automatically
depending on the selected track algorithm. The following values are typical for
analog video formats:
Algorithm Size (% of Active Video or Pixels)
Centroid 4% to 85%
Edge 4% to 85%
MTT 4% to 90%
Phase Correlation 24 x 12 to 176 x 88 pixels
MTD (Acquisition only) 32 x 16 to 256 x 128 pixels
Combined (Track only) 24 x 12 to 176 x 88 pixels
Scenelock is not included, as it does not process every pixel contained in the
Scenelock window and so can process any window size up to the video
processing limits.
For digital formats, the variability in frame sizes and video processing rates
makes it impractical to define these limits.
NOTE
Refer to Block 002, 003 in the Command Reference.
5 • Symbology
Overview The AVT has a symbology generator facility, which inserts graphics and text
(alphanumeric characters) into the video output. This is user‐configurable to suit
the application.
The symbology is contained within the active graphic region.
Each symbology item can be selectively enabled and disabled, as can the entire
symbology overlay.
The display position of text symbology items can be configured.
NOTE
Refer to Block 001, 011, 042 in the Command Reference.
Figure 5-1 Symbology
Document No: IN95-0005-01 Edition 4B Symbology 25

5.1 Graphics
The intensity of the inserted graphics symbology is programmable to a level
between black and white, or can be white with black shadows. The symbology is
contained within the active video region.
5.1.1 Sensor Boresight

The sensor boresight symbology consists of a cross, and marks the position of the
boresight of the current sensor.
Figure 5-2 Sensor Boresight Symbology
5.1.2 Window
The window symbology consists of four right‐angle markers, which outline the
area of either the Track Window or Acquisition Window depending on the AVT
mode.
Figure 5-3 Window Symbology
5.1.3 Target Aimpoint
The target Aimpoint symbology consists of a small cross, and marks the position
of the target Aimpoint as calculated by the Track Algorithm.
When the AVT is in Autotrack Mode, the Aimpoint symbology is suppressed
when a Breaklock condition has occurred. It is restored once the Breaklock
condition has been removed. The target Aimpoint symbology will not represent
any Aimpoint offset vector in use.
Figure 5-4 Target Aimpoint Symbology
5.1.4 Acquisition Target Marker/Primary Detect Target

The acquisition target marker symbology marks the position of the ‘Acquisition
Target’ that is assigned for automatic tracking when Autotrack mode is selected.
In Scenelock acquisition mode, the acquisition target marker is displayed centred
on the boresight to indicate that the scenelock algorithm has detected suitable
tracker points. When there are no suitable tracking points, the acquisition target
marker is not displayed.
NOTE
In Centroid or edge acquisition, the acquisition target is marked by a cross.
Figure 5-5 Acquisition Target Marker Symbology

5.1.5 Detected Target Marker
The Detected Target symbology marks the position of secondary targets detected
by the MTT algorithm. It consists of a medium sized cross and an optional
numeric tag. There can be up to four detect markers displayed at any one time.
Figure 5-6 Detected Target Symbology
5.1.6 Manual Window Size Marker

The manual window size symbology marks the current manual window size
demanded when the automatic window size mode is selected during Track mode.
This enables a suitable manual window size to be set before switching to manual
window size mode, and so reduces the possibility of loss of track due to an
incorrectly sized window.
Figure 5-7 Manual Window Size Symbology
5.1.7 Edge Selection

When the Edge algorithm is in use, the Edge Marker symbology indicates the
currently selected edge. It consists of a medium sized marker that is placed next
to the appropriate edge/corner of the window. The edge symbology is
automatically enabled when the edge algorithm is selected.
Figure 5-8 Edge Marker Symbology
5.1.8 Threshold Gauge
The pre‐processor Threshold Gauge symbology provides an indication of the
current target and background peak intensities, and the current threshold level.
This allows the operator to determine how to set the ‘threshold mode’ and the
‘manual offset’ when viewing the current scene.
The Threshold Gauge consists of a horizontal axis, where the left of the axis is
equivalent to an intensity of 0 and the right of the axis is an intensity of 255. The
Threshold Gauge display depends on the polarity being used.
Positive Contrast The top cursor indicates the current setting of the ‘Threshold level’ with a line

drawn from intensity 0 to a pre‐processor‐derived background intensity. A
second cursor (bottom) indicates the current peak target intensity level.
Figure 5-9 Positive Contrast
Negative Contrast The top cursor indicates the current ‘Threshold level’ with a line drawn from

intensity 255 to a pre‐processor‐derived background intensity. A second cursor
(bottom) indicates the current peak target intensity level.
Figure 5-10 Negative Contrast

Bi-Polar • The top left cursor indicates the current setting of the Negative Contrast
‘Threshold level’
• The bottom left cursor indicates the current Negative Contrast peak target
intensity level
• The top right cursor indicates the current setting of the Positive Contrast
‘Threshold level’
• The bottom right cursor indicates the current Positive Contrast peak target
intensity level
• The line drawn between the Negative and Positive Contrast ‘Threshold level’
represents the pre‐processor derived background intensity range
Figure 5-11 Bi-Polar
5.2 Text
5.2.1 Status Line
The status line indicates the AVT status as detailed below.
Figure 5-12 Status Line
1 2 3 4 5 6 7
1. Target Priority.
2. Acquisition Mode.
3. Acquisition Algorithm.
4. Acquisition Polarity.
5. Track Mode.
6. Track Algorithm.
7. Track Polarity.
Acquisition Status
Field Label Description
1. Target Priority S Size
P Position
M Motion (highest speed)
I Intensity
Manual
2. Acquisition Mode ACQ Acquisition Mode
ARM Armed Acquisition Mode
REV Reversionary Acquisition Mode
3. Acquisition Algorithm CENT Centroid
EDG Edge
MTT Multiple Target Track
PCOR Phase Correlation
SNLK Scenelock
MTD Moving Target Detection
4. Acquisition Polarity Bp Bipolar
Wt White (Positive Contrast)
Bl Black (Negative Contrast)
Au Automatic

Track Status
Field Label Description
5. Track Mode TRK Track Mode
CST Coast Mode
REV Reversionary Track Mode
6. Track Algorithm CENT Centroid
EDG Edge
MTT Multiple Target Track
PCOR Phase Correlation
SNLK Scenelock
COMB Combined
7. Track Polarity Bp Bipolar
Wt White(Positive Contrast)
Bl Black (Negative Contrast)
Au Automatic
NOTE
The AVT is in Acquisition Mode when the “Target Priority” and “Acquisition Mode” fields are highlighted.
The AVT is in Track Mode when the “Track Mode” field is highlighted.
Polarity is only shown when using MTT, Centroid, Edge or Combined.
5.2.2 LOS
The LOS symbology indicates the current LOS used by the AVT.
5.2.3 Sensor FOV

The Sensor text Symbology displays the current FOV used by the AVT.
6 • Track Error Generation
Introduction The Track Error (Target to Boresight Error) is defined as the position of the target
Aimpoint with respect to the Boresight. The Track Error is used as the input to the
platform filters, and hence used to calculate the platform rate demand outputs.
Target Aimpoint The target Aimpoint is defined as follows:

Track Algorithm Aimpoint
Centroid and MTT Target centre of energy
Phase Correlation Match position
Edge Target edge
Combined Match position or target centre
Scenelock Point in the scene
Aimpoint Offset It is possible to modify the Aimpoint by adding an offset vector to the position

calculated by the track algorithm, before the calculation of the platform demands.
When the output of the AVT is connected to a servo platform, this results in the
target being tracked off‐Boresight. Aimpoint offset has no effect on the target
Aimpoint symbology.
There is a ‘settle time’ of 2 seconds following entry to Autotrack Mode before the
offset vector is applied. Following this settle time, Aimpoint offset vectors are
applied within two fields of receipt of command. The offset is set to zero in all
AVT modes other than Autotrack. Typically, the Aimpoint offset is controlled
from the system joystick.
NOTE
Figure 6-1 Track Error Generation

Demand
Error X
AIMPOINT
OFFSET
VECTOR
Demand
Error Y
BORESIGHT
AIMPOINT
Document No: IN95-0005-01 Edition 4B Track Error Generation 33

Aimpoint In Track Mode with either phase correlator or scenelock algorithm, it is possible to
Refinement modify the tracking aimpoint by choosing a new reference position.
There is a ‘settle time’ of 2 seconds following entry to Autotrack Mode before the
aimpoint refinement is applied. Following this settle time, aimpoint refinement
vectors are applied within two fields of receipt of command. Typically the
aimpoint refinement is controlled from the system joystick. It is recommended
that aimpoint offset is disabled when aimpoint refinement is being used.
NOTE
Boresight The AVT boresight defines the origin for the Target to Boresight Errors. It is
possible to locate the boresight at any position within the Active Video Region
with the position of the boresight held in the Sensor Configuration Block.
It is possible to adjust the AVT boresight position. The correction is applied within
two fields of receipt of command. The boresight position is independently set for
each of the sensor configurations. The boresight position therefore automatically
changes to match the selected sensor and FOV.
NOTE
Refer to Sensor Configuration Blocks in the Command Reference.
7 • Sensor Compensation
Overview Sensor Compensation enhances the performance of the AVT when operating with
multi FOV systems comprising:
1. Fixed FOV sensors.
2. Switched FOV sensors.
3. Zoom sensors.
4. Combinations of the above.
Sensor Compensation requires the sensors to be calibrated to determine the actual
Sensor FOV and Boresight position.
Sensor FOV AVT image processing is naturally performed in pixels, but with knowledge of

the current Sensor FOV, the AVT can convert measurements into angular
measurements. This has the following advantages:
1. AVT tracking and breaklock functions using target position and size can be
performed in angular coordinates, which are unaffected by FOV changes.
2. Target to boresight measurements can be input to the AVT Platform Filters in
angular coordinates and so are unaffected by FOV changes.
3. Target measurements can be reported to the User in angular coordinates.
Sensor Boresight AVT measurements are generally reported relative to the Sensor Boresight, so by

offering boresight configuration settings for each sensor, the AVT can handle
small boresight misalignments between:
1. Different sensors.
2. Switched FOVs on the same sensor.
3. A set of defined zoom points on a zoom sensor arising from mechanical
wander of the boresight during zoom (providing it is consistent).
Sensor LOS If the current LOS Rate or Position can be provided to the AVT, there are the

following advantages:
1. When used with a correctly configured Sensor FOV, it will improve target
acquisition and tracking performance, since tracking will be performed
using absolute angular information (inertial coordinates) rather than
boresight relative angles/pixels (set block 001 field 9 to 1).
2. It can be used to initialize platform rate demand filters.
3. It allows use of the Synthetic Target in closed loop situations.
4. It allows easier calibration of Sensor FOVs.
Document No: IN95-0005-01 Edition 4B Sensor Compensation 35

7.1 Sensor Configuration
The AVT provides Sensor Configuration blocks to support up to four input
sensors. This minimizes the amount of data that needs to be sent to the AVT when
switching between sensors. Once the appropriate Sensor configuration settings
have been established and set in the AVT, the User can switch between them via
the Select Video Sensor configuration item.
Sensors Each sensor has the following configurable parameters:
1. Video Input:
• Analog Video Input 1
2. Video Standard:
• CCIR
• RS170
• PAL
• NTSC
• Digital Video Standards (Application specific)
NOTE
You can limit the Video Processing area to a programmable rectangle within the field of view.
3. Boresight X position and Y position.
4. Field of View:
• Switched FOV
• Zoom FOV
• User Defined FOV
NOTE
Refer to the Sensor Configuration Blocks 001, 024-035 and 055-060 in the Command Reference.
Sensor Field of The sensor field of view input to the AVT defines the Horizontal Field of View

View corresponding to the Active Video Region.
NOTE
The AVT calculates the vertical FOV from the HFOV using the FOV Ratio which is
defined as follows:
FOV Ratio = FOV of Vertical Active Video Region
FOV of Horizontal Active Video Region
Example The AVT defines the Horizontal Active Video Region (AFH) as 512 Pixels which is
Calculation of FOV equivalent to 51.2μs.
Ratio for CCIR CCIR defines the sensor horizontal active video region (SFH) as 52μs which is
(ADEPT 60) equivalent to 520 pixels. Thus:
AFH = (512/520) x SFH
The AVT defines the Vertical Active Video Region (AFV) as 256 Pixels which is
equivalent to 512 TV Lines.
CCIR defines the vertical active video region (SFV) as 575 TV Lines which is
equivalent to 287.5 pixels. Thus:
AFV = (256/287.5) x SFV
The AVT FOV ratio is defined as the FOV of the Vertical Active Video Region over
the FOV of the Horizontal Active Video Region, which can be expressed in terms
of CCIR active video FOV as follows:
(AFV/AFH) = ((256/287.5)/(512/520)) x (SFV/SFH)
or
(AFV/AFH) = 0.904 x (SFV/SFH)
CCIR defines (SFV/SFH) = 3/4
Therefore, for CCIR:
(AFV/AFH) = 0.678
Table 7-1 Typical Values of FOV Ratio

Standard ADEPT60, ADEPT64 ADEPT104, AIM12,
(10 MHz) (13.5 MHz)
CCIR 0.678 0.732
RS170 0.764 0.732
Switched FOV The switched FOV compensation supports up to five different FOVs. Each FOV

Compensation has a dynamic Boresight position associated with it.
• Switched FOV 1
• Switched FOV 2
• Switched FOV 3
• Switched FOV 4
• Switched FOV 5
When a change of switched sensor FOV occurs in Track mode, the AVT enters a
forced Breaklock phase during which it attempts to re‐acquire the target in the
new FOV. The duration of the forced Breaklock phase is configured by the FOV
Change Time command. If the automatic window‐sizing mode is selected, then
the window is automatically re‐sized by the ratio of the FOV change.
If the Dynamic Boresight mode is enabled, then the Boresight position associated
with the new FOV is applied at the end of the forced Breaklock phase. At the end
of the forced Breaklock phase, the Track Window position is recalculated based on
its offset from the original Boresight position, the ratio of the FOV change and the
new Boresight position.

Zoomed FOV The horizontal FOV value for Zoomed FOV Sensors is calculated using a value,
Compensation the Zoom Feedback, input via the Command Interface. The Zoom Feedback is
used to estimate a value for the Horizontal Field of View (HFOV) using a Lookup
table that consists of 13 entries of Zoom Feedback with the corresponding values
of horizontal FOV.
If the Dynamic Boresight mode is selected, the Boresight positions for Zoomed
FOV Sensors are calculated using a value, the Zoom Feedback, input via the
Command Interface. The Zoom Feedback is used to estimate a value for the
Dynamic Boresight positions using a Lookup table that consist of 13 entries of
Zoom Feedback with the corresponding values of Boresight X and Y positions.
The Zoom FOV and Boresight tables are ordered with the lowest Zoom Feedback
at the first entry and ascending values in consecutive entries. The lookup tables
use a straight line interpolation to calculate the HFOV and Boresight positions for
Zoom Feedback values that occur between two consecutive entries.
Table 7-2 Zoomed Sensor FOV and Boresight
Entry Zoom Feedback FOV Boresight X Boresight Y
1 Lowest zoom value
2
..
13 Highest zoom value
Sensor Position The AVT accepts sensor pointing angle inputs. It is possible to configure the AVT

Compensation to combine these inputs with measured, pixel, target positions to reconstruct an
‘absolute’ position with respect to a static coordinate set. To utilize this function,
the units used for Sensor Position must be same as those used for the sensor field
of view.
The sensor pointing angles are input as either position or rate inputs, or both. If
only rate inputs are supplied, then these are integrated internally to derive a
position.
NOTE
Compensated The reconstructed position information is used in the following AVT functions.
Functions
Track Window The Track Window Position is controlled by the reconstructed ‘absolute’ target

Position position. This removes some of the AVT reliance on smooth platform control for
secure tracking. It is possible to enable prediction of the reconstructed window
position using the ‘Track Window Position Filter’ and ‘Coast Mode’ commands.
NOTE
Track Breaklock It is possible to use the reconstructed ‘absolute’ position as an input to the target

Breaklock monitoring. The Breaklock measures the variation of the target speed
and reports the target as ‘inconsistent’ during periods of high acceleration.
7.2 Setting Up Boresight and FOV Compensation
Introduction This section describes how to configure the AVT so that it automatically
compensates for the changes in Field of View (FOV) and Boresight position that
occur when the focal length of the input camera changes.
7.2.1 Initial Work

Assumptions The following procedures are aimed at determining the compensation settings for
two cameras. The procedures use the following assumptions:
Step Assumption
1 The platform position feedback is available
The azimuth platform position increases as the camera moves to the right (as seen
2
through the camera video images)
The elevation platform position increases as the camera moves up (as seen through the
3
camera video images)
4 For a zoom lens camera, a tell back of the zoom position is available
The first camera uses a switched FOV lens and is configured through Sensor 1 definition
5
within the AVT
The second camera uses a continuous zoom lens and is configured through Sensor 2
6
definition within the AVT
7 Sample points within the range of the zoom lens have been already chosen
Switched FOV The switched FOV camera compensation is designed to allow the AVT to track

Camera through changes in FOV. To do this, the AVT must be told how long the video
image is disturbed by the lens change. This duration can be estimated or
measured by observing the action of the lens. It is typically better to over‐estimate
the duration of the lens change. The value is set using the FOV Change Time
command (Block 024, Field 10 for Sensor 1).
The value is entered in video fields. For example, a 1 second disturbance would be
commanded by setting a value of 60 for a RS170 camera system.
Zoom Lens Camera The zoom lens camera compensation is based around a 13‐point look‐up table

which has one input (zoom tell back) and three outputs (FOV, Boresight X and
Boresight Y).
It is not strictly necessary to use all of the look up table points, however it is
important to determine where the points should be within the range of the zoom
tell back. It is important to have one point at the narrowest FOV and another point
at the widest FOV. The other points should be distributed through the range of tell
back, concentrating on areas where the relationship between the tell back and the
FOV (or Boresight position) are non‐linear.
The look‐up table function requires that the lowest value of zoom tell back is
entered as point 1 and that the value of the subsequent points are strictly
increasing.

7.2.2 Boresighting Procedure
Physical Boresight Follow the steps in the table below:
Alignment
Step Action
1 Drive both cameras to their narrowest FOV
Disable the dynamic Boresight calculation on Sensor 1 (Block 055, Field 1 set to 0) and
2
Sensor 2 (Block 058, Field 1 set to 0)
Set the Boresight to the centre of the active region (nominally 256, 128 for CCIR and 256,
3 120 for RS170 video), on Sensor 1 (Block 024, Fields 3 & 15) and Sensor 2
(Block 027, Fields 3 & 15)
4 Enable the Boresight symbology (Block 011, Field 7 set to 1)
Select the camera with the narrowest FOV. The following procedures assume that this is
5
the first, switched FOV camera. To select this camera (Sensor 1) set Block 001, Field 2 to 0
Drive the platform so that the Boresight symbology is lined up over a static scene feature.
6 The feature should be a FOV intransient point (such as the corner of a building), which is
visible in both cameras
7 Lock the platform position
8 Select the second camera (set Block 001, Field 2 to 1)
9 Mechanically move the camera until the Boresight symbology is over the scene feature
10 Select the first camera again (set Block 001, Field 2 to 0)
11 Verify that the Boresight symbology is still over the scene feature
Electronic This procedure assumes that the alignment follows on from above, so the AVT
Boresight configuration is the same.
Alignment
Step Action
Select the camera with the narrowest FOV (assuming this is the first, switched FOV
1
camera, (set Block 001, Field 2 to 0)
Drive the platform so that the Boresight symbology is lined up over a static scene feature.
2 The feature should be a FOV intransient point (such as the corner of a building), which is
visible in both cameras
Switched FOV Follow the steps in the table below:
Camera (Sensor 1)
Step Action
1 Note the current X position of the Boresight symbology (Block 024, Field 3, should be 256)
2 Write this value to Switched FOV 1 Boresight Position X (Block 055, Field 4)
Note the current Y position of the Boresight symbology
3
(Block 024, Field 15, should be 128 or 120)
4 Write this value to Switched FOV 1 Boresight Position Y (Block 055, Field 8)
5 Switch the camera to the next FOV
Move the Boresight symbology (Block 024, Fields 3 & 15) until the Boresight symbology is
6
over the scene feature
7 Note the current X position of the Boresight symbology (Block 024, Field 3)
8 Write this value to Switched FOV 2 Boresight Position X (Block 055, Field 5)
9 Note the current Y position of the Boresight symbology (Block 024, Field 15)
10 Write this value to Switched FOV 2 Boresight Position Y (Block 055, Field 9)
Repeat steps 5 through 10 for each subsequent switched FOV.
11
Set the sensor configuration ready for use
12 Enable the dynamic Boresight calculation on Sensor 1 (Block 055, Field 1 set to 1)
13 Set Sensor 1 to use switched FOV compensation (Block 024, Field 11 to 1)
14 Save the configuration
15 Use the sensor compensation

Zoom Lens Camera Follow the steps in the table below:
(Sensor 2)
Step Action
1 Select the second camera (set Block, 001 Field 2 to 1)
2 Ensure the camera is at its narrowest FOV (highest zoom)
3 Note the current value of the FOV tell back
4 Write this value to the Input Side Zoom Feedback 1 (Block 028, Field 1). See Note 1
Move the Boresight symbology using (Block 027, Fields 3 & 15) until the Boresight
5
symbology is over the scene feature
7 Write this value to Boresight X Output 1 (Block 059, Field 1)
9 Write this value to Boresight Y Output 1 (Block 060, Field 1)
10 Drive the zoom to the next tell back sample point
11 Note the current value of the FOV tell back
12 Write this value to the Input Side Zoom Feedback 2 (Block 028, Field 2). See Note 2
Move the Boresight symbology using (Block 027, Fields 3 & 15) until the Boresight
13
symbology is over the scene feature
15 Write this value to Boresight X Output 2 (Block 059, Field 2)
17 Write this value to Boresight Y Output 2 (Block 060, Field 2)
Repeat steps 10 through 17 for the subsequent tell back sample points. Set the sensor
18
configuration ready for use
19 Enable the dynamic Boresight calculation on Sensor 1 (Block 058, Field 1 set to 1)
20 Set Sensor 2 to use zoom lens compensation (Block 024, Field 11 to 0)
21 Save the configuration
NOTES
1. This assumes that the zoom tell back increases as the FOV increases. However if the tell back
reduces as the FOV increases, the tell back value should be written to (Block 028, Field 13 and the
boresight values to Blocks 59, 60, Field 13)
reduces as the FOV increases, the tell back value should be written to (Block 028, Field 12 and the
boresight values to Blocks 59, 60, Field 12).
7.2.3 FOV Calculation Procedure
Follow the steps in the table below:
Step Action
Configure the AVT algorithms for MTT acquisition (set Block 002, Field 1 to 2) and for
1
Centroid tracking set Block 003, Field 1 to 0
2 Enable the LOS on-screen symbology (set Block, 042 Field 1 to 1)
Configure the platform drive so that the position of the platform is under operator control
3
regardless of the tracking state of the AVT
Switched FOV Follow the steps in the table below:

Camera (Sensor 1)
Step Action
1 Select the switched FOV camera (set Block 001, Field 2 to 0)
2 Select the narrowest FOV of the camera
Locate a small, static, high contrast, bounded target, such as a light, window, sign etc. The
3
ideal size would be about 20 x 10 pixels
Configure the AVT acquisition target polarity to automatic (set Block 002, Field 3 to 3) and
4 set the tracking target polarity based on the contrast of the object against its local
background (set Block 003, Field 2 to 1 for Positive Contrast and 2 for Negative Contrast)
Move the platform or acquisition window so that the target is within the acquisition
5
window and engage the AVT track mode (set Block 001, Field 1 to 1)
6 Drive the platform, gently, to move the target to the bottom left of the screen
Monitor, average and record the platform azimuth and elevation positions, either directly
8 from the platform; from the on-screen symbology; or through Block 152, Fields 2 and 3.
Record these as PP1 X and Y
Monitor, average and record the tracker Target to Boresight Errors (Block 129, Fields 8 and
9
9). Record these as TP1 X and Y
10 Drive the platform, gently, to move the target to the top right of the screen
11 Ensure the AVT is still tracking the same target
Monitor, average and record the platform azimuth and elevation positions. Record these as
13
PP2 X and Y
14
15 Calculate the X axis FOV
16 Write this value to Switched FOV 1 (Block 025, Field 14)
17 Calculate the FOV ratio
18 Record the FOV ratio
19 Switch to the next FOV of the camera
20 Repeat steps 3 through 19 for each subsequent FOV of the camera
21 Calculate the average of the FOV ratios
22 Write this value to FOV Ratio (Block 024, Field 9)
23 Save the sensor configuration

Zoom Lens Camera Follow the steps in the table below:
(Sensor 2)
Step Action
1 Select the second camera (set Block 001, Field 2 to 1)
2 Ensure the camera is at its narrowest FOV (highest zoom)
Locate a small, static, high contrast, bounded target, such as a light, window, sign etc. The
3
ideal size would be about 20 x 10 pixels
Configure the AVT acquisition target polarity to automatic (set Block 002, Field 3 to 3) and
4 set the tracking target polarity based on the contrast of the object against its local
background (set Block 003, Field 2 to 1 for Positive Contrast and 2 for Negative Contrast)
Move the platform or acquisition window so that the target is within the acquisition
5
window and engage the AVT track mode (set Block 001, Field 1 to 1)
6 Drive the platform, gently, to move the target to the bottom left of the screen
Monitor, average and record the platform azimuth and elevation positions, either directly
8 from the platform; from the on-screen symbology; or through Block 152, Fields 2 and 3.
Record these as PP1 X and Y
9
10 Drive the platform, gently, to move the target to the top right of the screen
11 Ensure the AVT is still tracking the same target
Monitor, average and record the platform azimuth and elevation positions. Record these
13
as PP2 X and Y
14
15 Calculate the X axis FOV
16 Write this value to Output Side FOV 1 (Block 029, Field 1). See Note 1
17 Calculate the FOV ratio
18 Record the FOV ratio
19 Drive the zoom to the next tell back sample point
20 Repeat steps 3 through 19 for subsequent tell back sample points
21 Calculate the average of the FOV ratios
22 Write this value to FOV Ratio (Block 027, Field 9)
23 Save the sensor configuration
NOTE
reduces as the FOV increases, the FOV value should be written to Block 029 Field 13.
FOV Calculation X‐axis FOV = 512.0 x (PP1X – PP2X)/(TP2X – TP1X)

This equation assumes a 10 MHz pixel digitalization rate, which provides
512 active pixels in the X‐axis. Some AVTs offer alternative digitalization rates;
refer to the Technical Specification to find the number of active pixels in the X‐axis
of the chosen video coordinate system.
NOTE
Refer to the Video Definitions section in the ADEPT AVT Hardware Reference Guides
Aspect Ratio Y‐axis FOV = 256.0 x (PP1Y – PP2Y)/(TP2Y – TP1Y)
Calculation FOV Ratio equals Y‐axis FOV/X‐axis FOV
This equation assumes a CCIR video signal using 256 active video lines per field.
For an RS170 video signal use a value of 240. Additionally, some AVTs offer an
increased number of active video lines. Refer to the Technical Specification to find
the number of active lines in the Y‐axis of the chosen video coordinate system.
7.2.4 Saving FOV and Boresight Configuration Values

Once the Sensor Configuration settings have been established, they should be
saved either on the host system to initialize AVT on power up or into the AVT
non‐volatile memory.
LINK
Refer to Section 11 • Tracker Configuration .
7.2.5 Using the FOV and Boresight Values

Switched FOV When the FOV of the camera changes, set the Set Switched FOV (Block 024, Field
Camera (Sensor 1) 13) to the appropriate index. The above procedures assume that the narrowest
FOV is index 0 and that the subsequent indices refer to wider FOVs. It is
important that this command is issued as soon as possible after the FOV change
has been started; an alternative is to send the command to the AVT when the
command to the camera is issued.
Zoom Lens Camera Continually send the tell back from the zoom lens to the AVT, by setting the

(Sensor 2) ‘Zoom Feedback’ (Block 027, Field 12). It is important that the delay in the
transmission of the zoom tell back to the AVT is low. On some AVTs, it is possible
to send the analog signal directly into the AVT card.

8 • Platform Control
Overview The primary objective of the complete Electro‐Optical Tracking System (EOTS) is
to control the angular position of the platform and attached sensor such that the
object of interest viewed through the sensor appears at the centre of the field of
view (boresight) and is maintained as close as possible to this point as the object
moves.
Rate Demand In a rate demand control system, the tracker outputs a rate demand to the

platform servo, which is derived from the Target to Boresight Error using the
platform filters.
The ‘Platform Filters’ can be used to modify the response of the closed loop
control. The Platform Filters include a compensator that is located in the forward
path of the control loop. The compensator consists of a standard five‐term
difference equation with coefficients that can be configured. This can implement a
compensator network with a complexity equivalent to a Second Order Transfer
function. To allow the response to be separately optimized for Manual gain and
joystick shaping and Automatic Tracking, there is a separate compensation
network for Manual (gain and joystick shaping) and Automatic Tracking which
are switched depending on the mode.
The AVT provides rate demand outputs for a servo platform, which are output
through the Serial and VME/PCI/ISA bus Interface and the analog outputs.
Figure 8-1 Control Loop
ADEPT AUTOMATIC
Zero
VIDEO TRACKER
Analog (ADEPT60 only)
Joystick Raw Acquisition
ADC
Joystick Mode Rate
Demand
Joystick
Shaped
Joystick
Interface rate Interface

Rate
Joystick
Integral
joystick
Target to
Boresight error
PLATFORM ZERO ORDER HOLD SERVO

VIDEO SENSOR AUTOTRACK FOV PLATFORM
CHARACTERISTICS (dependent on PLATFORM
SAMPLER DELAY DELAY COMPENSATION FILTERS ACQ COMPENSATION servo platform) (digital)
TARGET PLATFORM
POSITION + POSITION
TRK
-
DAC SERVO
ZERO ORDER HOLD PLATFORM
(ADEPT60 only) (analog)
Linear or
Integrating
Filter Coeffs
Mechanical
8.1 Acquisition Mode Rate Demand
In Acquisition Mode, the rate demand can be configured to operate in one of the
following ways:
1. Zero Outputs
The rate demands are set to zero.
2. Raw Outputs
The rate demands are set to the voltages received through the analog inputs.
This is not supported on all AVTs (see the Hardware Reference Manual for
each product).
3. Shaped and Gained Joystick
The rate demands are calculated by applying gains and shaping to the
selected Joystick Demand X and Joystick Demand Y device inputs Figure 8‐2
and Figure 8‐3.
4. Interface Rate Demand
The rate demands are set to the rate demand supplied across the command
interface.
5. Shaped Gained and Integrated Joystick
The rate demands are calculated by applying gains and shaping to the
selected Joystick Demand X and Joystick Demand Y device inputs. In
addition, the output is integrated over subsequent video cycles.
6. Absolute Position
The rate demands are calculated to move the platform to the absolute
position as defined in (Block 88, Fields 11, 12).
7. LOS Rate with Zero Output
The rate demands are set to zero, but when track is entered, the platform
filter is initialized with the LOS rate calculated from platform feedback. This
is used when the platform is under external control in acquisition.
8. LOS Rate
As above, but rate demands are not zeroed – they reflect the LOS rate
calculated from platform feedback.
Joystic Shaping Figure 8-2 Joystick Gains and Shaping
Joystick Rate Demand

Input Output
Offset and Deadband Input Shaping Output Integrator
Compensation Gain Gain (Optional)
Document No: IN95-0005-01 Edition 4B Platform Control 47

Joystick Shaping The joystick shaping mode can be defined as:
Mode • Single cutpoint ‐ two stage joystick control with lower gain up to the
cutpoint and higher gain after
• Lookup table ‐ a predefined lookup table yielding increasing gain with
joystick deflection
The joystick compensation coefficients are contained in (Blocks 19, 20).
Figure 8-3 Joystick Shaping
Output
Input
Cutpoint
8.2 Track Mode Rate Demand

Autotrack Mode In Autotrack mode, the platform filters consist of the following elements:
Rate Demand
Figure 8-4 Autotrack Mode Rate Demand Filters
Filter 1 yc (k)
Target to (Linear)
Boresight Rate
Error Demand
Field of View
Compensation Select
Filter 1 or 2
X(k)
Output
Filter 2
(Integrating)
Xc (k)
yc (k)
FOV Field of View Compensation
Compensation
xc(k) = (x(k). KFOV)/NPIX
xc(k) = FOV compensated Track error
x(k) = Track error (including Aimpoint offset) in pixels
KFOV = Axial (azimuth or elevation) Field of View (mRad) of the selected video input
NPIX = Number of pixels in that axis (azimuth or elevation)
LINK
See “Sensor Compensation” on page 35.
Platform Filters There are two switchable platform filters for each axis:

Filter 1. Typically this would be configured with a gain‐only closed loop
controller.
Filter 2. Typically this would be configured with an integrating closed loop
controller that provides improved following performance when tracking a
moving target.
NOTE
Each platform filter consists of a digital filter with the following structure for each
axis.
yc(k) = G.[P0.xc(k) + P1. xc(k-1) + P2.xc(k-2)] + [I1.yc(k-1)+ I2.yc(k-2)]
yc(k) = Filter Output
yc(k-1) = Filter Output delayed by 1 sample
yc(k-2) = Filter Output delayed by 2 samples
xc(k) = Current Track error
xc(k-1) = Track error delayed by 1 sample
xc(k-2) = Track error delayed by 2 samples
G = Filter gain coefficient. Reference (Block 22, Block 23)
P0,P1,P2,I1,I2 = Filter coefficients. Reference (Block 22, Block 23)
Sample Rate = Input Video Field Rate
Coast The output of the platform filters is coasted at a constant rate when the tracker
enters Breaklock mode.
External Los Rate The following notes should be used to help configure the AVT to use the supplied

for PID Initialization LOS rate to initialize the closed loop tracking filters:
At the point of transition to Track mode, the AVT initializes the closed loop PID
rate term with the current rate demand output – i.e. the last rate calculated during
Acquisition mode. Hence, the way to initialize the closed loop PID with the
supplied LOS rate is to command the AVT to use this rate to form the Acquisition
mode rate demand. The relevant command is the Acquisition Mode Platform
Output Selection. This should be set to either ‘LOS rate with zero output’ (value
6), or ‘LOS rate’ (value 7).
NOTE
The effect of correct PID initialization will only be seen when a PID filter with an integrating term has
been configured

8.3 Platform Characteristics Compensation
Platform compensation converts the rate demands to platform demands. The
units of the platform demand outputs are nominally millivolts.
NOTE
The deadband and drift calculations are not performed when using the Zero Outputs and Raw Outputs in
Acquisition mode).
Figure 8-5 Platform Compensation
Rate Platform
Demand Demand
Platform Limit Platform Platform

Scalar Demand Deadband Drift
Platform Scalar The output scaling converts the rate output to a Platform Demand voltage as

follows:
YV(k) = Yc(k)/KV
Where:
KV = Platform output scalar (mRad/S to millivolts)
YC(k) = Compensation filter output (mRad/S)
YV(k) = Platform demand (millivolts)
Limit Demand The Platform Demand is limited to user and system specified limits as follows.

YL(k) = DMAX for YV(k) > DMAX
= DMIN for YV(k) < DMIN
Where:
DMAX = The minimum of the User and System Maximum Demands
DMIN = The maximum of the User and System Minimum Demands
YV(k) = Platform demand (millivolts)
YL(k) = Limited demand (millivolts)
Platform Deadband The Deadband compensation adds a fixed voltage to the Platform Demand to

compensate for any Deadband in the platform.
YD(k) = YL(k) + KDEAD for YL(k) > 0
= YL(k) - KDEAD for YL(k) < 0
Where:
KDEAD = Deadband compensation (millivolts)
YD(k) = Deadband compensated platform demand (millivolts)
YL(k) = Limited demand (millivolts)
Drift Compensation The drift compensation adds a fixed voltage to the Platform Demand to
compensate for any drift in the platform.
Y(k) = YD(k) + KDRIFT
Where:
KDRIFT = Drift Compensation value (millivolts)
YD(k) = Deadband compensated platform demand (millivolts)
Y(k) = Fully compensated platform demand (millivolts)
Notes on the The following notes should be used to configure the AVT so that the internally

Scaling of Platform calculated rate demands are the same ‘units’ as the supplied LOS positions and
rates:
The following diagram illustrates the scaling that the AVT performs, in Track
mode, between measuring the target position and generating a platform rate
demand voltage.
Figure 8-6 Scaling of Platform Outputs
Rate Demand
Target Position FOV Scaling Target Position Platform Rate Demand
PID Filter (Milliradians
(Pixels) (Milliradians) (Milliradians) Scalar (Millivolts)
per second)
When the platform is moving

at a steady speed this value
should match the LOS rate
fed to the AVT
(Not output by the AVT!)
NOTE
This process is separately applied in the X and Y axis.
FOV Scaling The FOV scalar used by the AVT can be set in several ways, but fortunately the

X‐axis FOV scalar currently being used by the AVT is output in (Field 3, Block
130).
The Y‐axis FOV scalar can be calculated by applying the FOV ratio. This value is
supplied to the AVT through the sensor configuration block (Field 9, Block 24).
NOTE
To successfully use the synthetic target, the FOV values should be entered as milliradians. If the FOV
values are not currently sent to the AVT in milliradians, then changing them to milliradians will mean that
the PID gain terms will also need to be adjusted. If the supplied FOV values have increased, then the PID
gain terms should be reduced by the same scale factor.
PID Filter The scaling applied by the PID filter should be adjusted using the gain terms of

the PID equation. It is important to check which set of PID filters is currently
being used by the AVT (Field 6, Block 21).
The gain terms of the PID filters are located in (Fields 6,12, Blocks 22, 23).
Platform Scalar The platform scalar is located in (Field 4, Block 121).
NOTE
One scalar is applied in both X and Y axes. The AVT uses the platform scalar as follows:
Rate (millivolts) = Rate (millradians)/Platform Scalar.

Calculating the The platform scalar may be a known characteristic of the platform. Alternatively,
Platform Scalar assuming that the AVT is currently configured to adequately closed‐loop‐track a
moving target, the following process can be used to calculate the platform scalar
value. If a different platform scalar is deduced, it will then be necessary to adjust
the PID gain values as described in step 3.
1. While closed‐loop‐tracking a moving target:
a. Command the tracker to enter a Forced Coast (Field 10, Block 3).
b. Note the current AVT Rate Demand (Fields 14, 15, Block 129).
c. Note the current supplied LOS rate (Fields 9, 13, Block 152).
2. Calculate the platform scalar as (LOS rate/AVT Rate demand). If the X and Y
axis are different, the X‐axis should be used where more target velocity is
seen.
3. Increase the PID gain terms by the same factor as the increase in the platform
scalar.
8.4 AVT Platform Tuning

Platform Gain The following is a method to improve the closed loop tracking performance of a
Tuning system that uses the AVT Platform demand outputs (Block 129, Fields 14 and 15),
by adjusting the X and Y gain values for filters 1 & 2 (defaults are Linear and
Integrating) settings, Block 022, Fields 6 & 12, and Block 023, Fields 6 & 12.
This is best performed after the platform has been (static and dynamically)
balanced. If this has not been done, then the gain values may need reducing to
avoid instability with closed loop tracking.
For tuning, an ideal static target is required, i.e. a high contrast single polarity
target and no obscuration in the FOV. If Sensor focus distance allows, then within
a Lab environment is ideal. If the system is to operate with several FOVs, whether
from single or multiple sensors, then the sensor compensation procedure must be
completed before the Platform tuning.
Platform Output Setting the platform output scaler (Block 121, Field 4) as per the platform is

Scaler optional, but if it is going to be set, it is best to do it first, otherwise various AVT
gain settings will need readjusting when its changed later.
The following changes to the standard default shipping configuration are
required,
1 Acquisition Algorithm (Block 002, Field 1) set to Centroid (0)
2 Acquisition Target Polarity (Block 002, Field 3) set to Automatic (3)
3 Track Algorithm (Block 003, Field 1) set to Centroid (0)
Track Polarity (Block 003, Field 2) set to either Positive Contrast (1) or Negative Contrast (2)
4
to match Target selected
5 TBE Bleed mode (Block 21, Field 7) set to Disabled (0)
Platform Filter Select (Block 21, Field 6) set to Filter 1 (0).
6
(Assumes Platform Filter 1 (Block 022) settings are the Linear Filter defaults)
7 (Centroid) Breaklock Algorithm (Block 004, Field 6) set to Disabled (0)
Tuning the Axis The easiest way to tune the axis gains is to do them separately. This is aided by
Gains disabling the Platform demand for the other axis in both Acquisition and
Tracking, and displacing the Target from the boresight only in the axis to be
tuned. The method for disabling the Acquisition platform demand will be system
configuration dependant, as the Platform could be driven directly by the host in
Acquisition mode, or be derived from a Joystick input to the AVT, which can be
configured via several methods.
8.4.1 Horizontal (Azimuth) Gain Tuning

1 Manually position target on Boresight
2 Disable Vertical (Elevation) Acquisition Platform output
3 Set (Linear) Filter 1 Y axis gain coefficient (Block 022, Field 12) to 0.0
4 Drive platform so target is horizontally displaced from boresight by approximately ¼ FOV
Set Acquisition gate position (Block 002, Fields 6 & 7) and size (Block 002, Fields 8 & 9) to
5
ensure target is designated
Enter AVT Track (set Block 001, Field 1 to 1) and observe platform response on AVT Video
6
output
7 Exit AVT Track (set Block 001, Field 1 to 0)
The observed platform response will fall into one of the following categories:
1. Satisfactory: the platform snapped promptly onto the target with no overshoot, in which
case the current gain setting is satisfactory and can move on to the next step.
2. Unsatisfactory: the gain will need to be adjusted depending on the response as follows
and then steps 4 to 7 repeated.
NOTES
1. If it oscillates, then the gain is too high.
2. If it disappears altogether, then the gain is either exceedingly high or reversed i.e. incorrect.
3. If It moves slowly, then gain is too low
Note down Filter 1 X axis Gain Coefficient (Block 022, Field 6) for use later. If there is a
significant static imbalance of the platform, then different results will be obtained for
9
displacements on opposite side of the boresight. In this case, the smallest absolute gain
should be used
10 Restore all configuration items to the start of the Horizontal Gain Tuning
8.4.2 Vertical (Elevation) Gain Tuning

Follow the horizontal procedure above, except that all the displacements are
vertical and that Filter 1 X axis gain coefficient (Block 022, Field 6) is set to 0.0 and
Filter 1 Y axis gain coefficient is adjusted until satisfactory.
Set both Filter 1 X & Y axis Gain Coefficients (Block 022, Field 6 & 12) using the
recorded values.
Carry the derived linear filter gains (Block 022, Field 6 & 12) over to the
integrating filter (Block 023, Field 6 & 12) and switch to using the integrating
filters for the AVT Platform demand outputs, by setting Platform Filter Select
(Block 021, Field 6) to Filter 2 (1). Using the integrating set will give better
following error performance, which is generally preferred.
Re‐enable the Centroid Breaklock Algorithm (Block 004, Field 6) by setting it to 1,
as it was disabled for the gain tuning above.

8.4.3 Platform Bleed Tuning
Usually Platform or TBE bleed is required, as otherwise the initial step demand
on entering tracking is too great for the system to hold lock.
Position the Target at the extreme of FOV displacement and enter tracking. If the
system loses lock or goes into Coast a lot while pulling the target to the boresight,
then enable Platform bleed by setting TBE bleed mode (Block 021, Field 7) to
Platform (1).
Repeat the above but change the bleed rate in the X and Y axis (Block 021, Fields 8
& 9) until the target is pulled to the boresight satisfactorily. These settings are a
compromise between the speed of response and maintaining tracking: too small a
bleed rate and the time to pull the target onto the boresight will be too long; too
large a bleed rate and the tracking will coast excessively or be lost.
These setting should now be transferred into the Host initialization of the AVT, as
they are platform configuration dependant (Servo gains). Hence, if Servo systems
within the platform are re‐calibrated, the above procedure will need repeating.
They may be set on the AVT and saved into Flash as part of the start‐up
configuration data set.
8.4.4 System Performance Test

Before actually testing under real world conditions with either the desired
Target(s) or Targets of opportunity, it is possible to perform tests in a Lab
environment using a rotating wheel (preferably with a small electric motor to give
a consistent rpm), though there will always be a following error with these due to
angular accelerations. Suitable targets to mount on the wheel would be:
• For TV Sensors: Solid filled circle for CENT/MTT and Cross‐Hatch filled
circle for PCORR
• For IR Sensors: An incandescent lamp (which will require slip rings) for
CENT/MTT. This usually works well with a blurred out of focus image
9 • Synthetic Test Target
Overview The AVT has the ability to overlay the image of two targets on the incoming video
image. The position of the targets is governed by internally calculated target
trajectories and the pointing angle of the video sensor.
The overlaid video is subsequently used as an input into the Tracking function of
the AVT. The AVT can then be commanded to track one of the overlaid target
images. The rate demand output by the AVT platform filters completes the closed
loop tracking of the overlaid target.
Synthetic target has the ability to be driven by real time user inputs as an
alternative option to the internal trajectories.
The Synthetic Test Target function is used to:
• Demonstrate and test the closed loop performance of video tracking system
• Provide inexpensive, highly realistic operator training
• Validate operational equipment configurations such as target range, attitude
and sensor FOV
NOTE
Refer to Block 011, 090-100 in the Command Reference.
Figure 9-1 Synthetic Video Target Generation
Sensor
Sensor
FOV
Servo Video Rate Demand

Platform
Platform
Position
Target
Target Platform
Position and Target - Position
Overlay on
Video
Demand
Size Position in FOV Tracking
Video Calculation
Calculations
Target Size Video Tracker Functions
The functional blocks required to allow synthetic test target generation are shown
in Figure 9‐1. The grey boxes indicate functionality that has been added to the
AVT for this purpose.
To operate the Synthetic Target in a Polar or Cartesian mode it is necessary to:
• Have properly calibrated sensor configurations
• Provide LOS to the AVT
Alternatively, the User can provide Dynamic (i.e. real‐time) pixel data, which the
AVT will use to insert the Synthetic Targets, but this would require the User to be
aware of Sensor FOV and LOS.
Document No: IN95-0005-01 Edition 4B Synthetic Test Target 55

9.1 Synthetic Target Configuration
The following steps should be followed to enable the Synthetic Target.
To select ‘Synthetic target’ source Peephole Source Select Set to Value 3
Display of each of the synthetic targets is controlled individually Enable Target Display Set to Value 1 to Enable
To configure each of the targets and their associated trajectory it Target Mode Value 0 (Disabled)
is necessary to disable the synthetic target trajectory, beforehand
Now set up the required Synthetic Target Trajectory
To apply the configuration changes it is necessary to enable Target Mode 1= Constant Polar Velocity
target trajectory 2= Polar Acceleration
3= Cartesian Velocity Dynamic
4= Position Mode
9.1.1 Polar - Constant Velocity Trajectory Configuration

Constant Velocity Trajectory Mode will produce a target that moves around the
screen at constant velocity. When the defined border limits are hit, the target will
change direction.
Size The azimuth and elevation size of the target is controlled using:
• Target Azimuth Size
• Target Elevation Size
Start Position The start azimuth and elevation position of the target is controlled using:

• Target Azimuth Start Position
• Target Elevation Start Position
Speed The azimuth and elevation velocity of the target in milliradians per second is
controlled using:
• Target Azimuth Velocity
• Target Elevation Velocity
Bounding Box The bounding box that limits the target to a defined area is set using:

• Target Minimum Azimuth
• Target Minimum Elevation
• Target Maximum Azimuth
• Target Maximum Elevation
9.1.2 Polar - Sinusoidal Trajectory Configuration
Sinusoidal trajectory mode uses cosine and sine functions to produce smooth
synthetic target acceleration on the screen. There are four types of sinusoidal
trajectory motions. The azimuth and elevation amplitude multipliers are set to
determine the targets position on the screen.
Size The azimuth and elevation size of the target is set using:
• Target Azimuth Size
• Target Elevation Size
Origin Position The origin position of the targets azimuth and elevation position is set using:

• Target Azimuth Start Position
• Target Elevation Start Position
Amplitude The amplitude of cosine and sine wave positioning and acceleration is set using:
• Target Azimuth Amplitude
• Target Elevation Amplitude
Frequency The frequency of cosine and sine wave positioning and acceleration is set using:
• Target Azimuth Frequency
• Target Elevation Frequency
For example, 500 milli‐Hertz = 0.5 Hertz = 2 seconds to carry out 1 rotation.
Target Mode The synthetic target is set to follow one of four target modes:

• Azimuth sine/Elevation sine
• Azimuth sine/Elevation cosine
• Azimuth cosine/Elevation sine
• Azimuth cosine/Elevation cosine
9.1.3 Cartesian Configuration

In the Cartesian mode, target trajectories are specified in an East, North and
Altitude coordinate set. The trajectory is defined as a set of initial conditions
(position, speed and acceleration) followed by a series of acceleration segments.
There are two targets available in Cartesian mode. Target 2 is designed to simulate
an object detaching from target 1. Target 2 can be configured to appear some time
into the trajectory of target 1. Once enabled, its trajectory is initialized to the
position and speed of target 1 and then modified by its own set of acceleration
segments.
The configuration of each Cartesian target is broken into two sections: the Base
configuration and the Segment configuration. The Base configuration sets the
initial trajectory values as well as the target size, shape and intensity. The Segment
configuration sets the accelerations that will be applied in each time segment.
Each segment has a time delay. This time effectively sets the time duration of the
previous segment.

The configuration of target 2 is very similar to that of target 1, except that it has an
additional time delay in the Base configuration. This time sets the point in the
trajectory of target 1 that target 2’s independent trajectory starts.
9.1.4 Synthetic Target Dynamic Position Mode

Dynamic Position This enables a user to control the trajectory of the synthetic target from an external
Mode source. The target position is supplied to the AVT over any of the supported data
interfaces.
Polar Modes - The East and Altitude values are interpreted as Azimuth and Elevation positions

Const. and Accel in milliradians.
The North value is interpreted as a range, allowing control over the Target
drawing order, with the lowest value considered closest and hence drawn last.
Target 1 and 2 Enable, Intensity and Sizes are taken from the normal Target Polar
Blocks 91 and 93 respectively (all other value in these blocks and 92 and 94 are
ignored in this mode).
Cartesian Mode The East, North and Altitude values are interpreted as distances from the

Boresight. Target 1 and 2 Enable, Intensity and Sizes are taken from the normal
Target Cartesian Blocks 95 and 96 respectively (All other values in these blocks
and 97 to 100 are ignored in this mode).
NOTE
The Intensity and Sizes for a Target are locked when both the Synthetic Target mode is active and the
Target is enabled (this is unchanged from the existing non-dynamic synthetic mode).
9.2 Example Trajectories
Example 1 The trajectory for this example synthetic target is produced with these
Polar Sinusoidal configuration settings:
Trajectory
Value 1 (Az Sin/El Cos) Sinusoidal Trajectory Type
Value 50 (50 Milliradians) Azimuth Amplitude
Value 50 (50 Milliradians) Elevation Amplitude
Value 500 (500 Milli-hertz) Azimuth Frequency
Value 500 (500 Milli-hertz) Elevation Frequency
Figure 9-2 Target Mode = Az sin/El cos (Sinusoidal Trajectory Type, Value 1)
A z s in / E l c o s T ra je c t o ry
A z im u t h
Az
p o s it io n
E le v a t io n
El
p o s it io n
A z im u t h &
e le v a t io n Az / El
p o s it io n
c o m b in e d
( h o w t h e t a rg e t
a p p e a rs o n
t ra c k e r o u t p u t )
5 0 m illira d ia n s

Example 2 This is an example of a sinusoidal trajectory configuration.
Polar Sinusoidal The azimuth frequency is different to the elevation amplitude.
Trajectory
Value 2 (Az Cos/El Sin) Sinusoidal Trajectory Type
Value 50 (50 Milliradians) Azimuth Amplitude
Value 50 (50 Milliradians) Elevation Amplitude
Value 500 (500 Milli-hertz) Azimuth Frequency
Value 250 (250 Milli-hertz) Elevation Frequency
Value 1 Enable Synthetic Target 1
Figure 9-3 Target Mode = Az cos/El sin (Sinusoidal Trajectory Type, Value 2)
Az cos / El sin Trajectory
Azimuth
Az
position
Elevation
El
position
Azimuth & Az / El
elevation
position
combined
(how the target

appears on
tracker output )
50 milliradians
Example 3 This configuration shows a target moving from left to right. After 5 seconds a
Cartesian Base second target will appear from target 1 moving in a downward direction. These
Configuration configuration settings simulate a target dropping a second object.
This example is for illustrative purposes. As a result, a platform or sensor does not
need to be attached to the AVT.
To see the entire trajectory paths of the targets, the sensor FOV settings need to be
set to simulate a fish‐eye camera lens. Assuming sensor 1 is used, the
configuration will be as follows:
Sensor 1 FOV Configuration Target 1 Base Configuration
FOV Ratio 1.000 Acceleration North 0
FOV Type Used Defined FOV (2) Acceleration Altitude 0
User Defined FOV 1570 Size X 100
Synthetic Target Configuration Size Y 100
Force Video Enable (1) Size Z 100
Background Intensity 200 Target 1 Segments
Target Mode Cartesian Mode (3) Segments 1-6, all elements 0
Target 1 Base Configuration Target 2 Base Configuration
Target Intensity Black (1) Time Delay 5
Position East -10000 Target Intensity Grey (100)
Position North 5000 Offset Position East 0
Position Altitude 1000 Offset Position North 0
Velocity East 1250 Offset Position Altitude 0
Velocity North 0 Acceleration East 0
Velocity Altitude 0 Acceleration North 0
Acceleration East 0 Acceleration Altitude 0
Size X 100
Size Y 100
Size Z 100
Target 2 Segments
Segments 1-5, all elements 0
Seg 6 Time Delay 0
Seg 6 East 0
Seg 6 North 0
Seg 6 Altitude -100
To trigger the Cartesian Trajectory set,
Target 1 Base Configuration
Target Enables (2) Target 1&2
This produces displays as shown in Figure 9‐4 and Figure 9‐5.

Figure 9-4 Cartesian Example 1, Base Configuration, Target 1 Only
Figure 9-5 Cartesian Example 1, Base Configuration, Target 1 and Target 2
Example 4 The Cartesian mode provides the facility to change the accelerations of the targets
Cartesian Segment during their flight. For example, to simulate a target dropping an object there
Configurations needs to be at least one change in acceleration of target 2 to simulate the object
falling.
In this example, the common configuration, FOV configuration and Synthetic
target configuration use the settings described in Example 3.
Target 1 Moves from west to east, so use the settings for target 1 as described in Example 3.
Target 2 Is dropped after 5 seconds and will fall using the Earth’s gravity of 9.8m/s2.
Offset East 0
Offset North 0
Offset Altitude 0
Acceleration East 0
Acceleration North 0
Altitude Acceleration -100
Size X 100
Size Y 100
Size Z 100
Segment 1 Time Delay 3
(3 seconds At –100m/s2 in the Altitude Axis)

Segment 1 Acceleration East 0
Segment 1 Acceleration North 0
Segment 1 Altitude Acceleration 0
Set all other target 2 segment time delays to 0.
NOTE
The segment 1 accelerations will continue to apply to target 2, until the target disappears because all of
the other segment times have been set to 0.
This produces the following display:
Figure 9-6 Cartesian Example 2 Segment Configurations

10 • Picture in Picture (PiP)
Overview Picture in Picture is provided in the AVT to give users an enhanced display of
targets that the system acquires or is tracking.
A small window is overlaid on the output video display showing the target of
interest whenever it is available. The details of size and position of the window,
and the zooming and image enhancements performed on the target are fully
configurable, along with persistence and frequency of updating of the picture.
NOTE
Operating The Picture in Picture function uses the scan converter and the display memory
Restrictions hardware to output its picture in the output video, and so is only available on the
ADEPT60.
The use of Picture in Picture is not possible when other facilities that require these
resources are in use. In particular Picture in Picture is disabled when either
‘Synthetic Target’ or ‘Digital Video’ is being used.
Some non‐standard builds of AVT software may require use of the display
memory and/or the scan converter, and care will be required when this is the case
to avoid a conflict. It is suggested that when such builds are required, the Picture
in Picture function should be disabled.
Enabling/Disabling The Picture in Picture function can be switched on or off using the enable
configuration item.
Zoom Modes The zoom configuration item defines whether the picture will be magnified in the

display window. The zooming can be set to a predefined level (x1, x2, x3, x4) or to
a ‘best fit’ mode (this is the default) that tries to maintain the target display size to
fully fit the window.
Enhancement The Enhancement Mode configuration item determines what, if any, image
Modes enhancement is applied when the picture is displayed:
• Nearest Neighbor. This mode provides no image enhancement. On zoomed
images it tends to give stepped edges (especially on higher zoom values)
• Bi‐linear Interpolation. This mode uses bi‐linear interpolation to smooth the
edges on zoomed images
• Contrast Stretch. This mode can be useful where there is little contrast
between the target and background of the image. Contrast is stretched by
taking the average image intensity for the picture and then ‘stretching’ the
intensity value at each pixel from that average value by an Enhancement
Factor.
For example, an enhancement factor of 1.0 gives no stretch ; values above 1.0
stretch by multiplying the offset from the average value by that factor. This
mode can result in some image saturation
• Histogram Equalization. This mode generates a histogram of the picture
and then applies a LUT generated from an ‘equalized’ version of the
histogram
Operating The maximum display update rate achievable with PiP depends on the size of
Performance window and the loading on the CPU. The implementation is designed only to use
spare CPU capacity for generating the PiP display and will automatically
‘throttle’ to the best update rate possible in the given scenario.
A typical CPU loading achieves an update rate of 25 Hz (on the ADEPT60) using a
default configuration window of 100 * 75 pixels (with bilinear interpolation).
Document No: IN95-0005-01 Edition 4B Picture in Picture (PiP) 65

11 • Tracker Configuration
Overview When the AVT configuration is changed via the maintenance interface, the
changes are only retained until the AVT is power cycled or reset. Following a
power cycle or reset, the AVT will load initial values for every Configuration Item
(except real‐time command items). The AVT contains an in‐built set of initial
values known as the Default Configuration; this set cannot be deleted. However,
an area of non‐volatile memory within the AVT has been reserved where Users
may save up to two further Configuration sets. Once saved, these Configuration
sets may be restored on command or be selected for loading during AVT
initialization, overriding the Default System Configuration. This is a useful short‐
term method of saving and using the configuration, and is frequently used in one‐
off AVT integrations.
When the AVT is to be installed in a system that uses an intelligent controlling
computer, it is recommended that changes from the AVT Default configuration be
initialized by the Host System at system power‐up.
NOTE
Saved configurations may be lost if the AVT Application Software is updated.
There are four commands related to saving and restoring configuration from non‐
volatile memory:
NOTE
Select Startup Saves an identifier to non‐volatile memory of the Configuration to be used at

System Config startup. This can either identify the Default configuration, or a User
Configuration providing one has previously been saved. When the AVT is power
cycled, it will load the selected configuration.
Save System Saves the current AVT configuration to non‐volatile memory as either Config 1 or

Config to Flash 2.
Save Startup Saves the current AVT configuration to non‐volatile memory as either Config 1 or

System Config To 2 and makes it the Startup configuration (combines the previous two commands
Flash but only requires one write to non‐volatile memory).
Load System Loads either the Default configuration or a saved User configuration from non‐

Config From Flash volatile memory.
12 • Tracker Performance
Target Tracking The AVT has the following target tracking performance:

Target To Boresight Error Update Rate 50 or 60 Hz
(For digital video this will be at the video frame rate)
Target to Boresight Error Latency <15ms
(Measured with respect to the target centre in video image)
Minimum Target Contrast 5%
Minimum Signal to Noise Ratio 4
Minimum Target Size 6 x 3 pixels
(This is the minimum size to meet the stated performance e.g. tracking rate,
minimum contrast etc. For a target with higher contrast and lower crossing
rates the AVT will be capable of tracking a single pixel target)
RMS Noise on Target Position Output <0.5 pixel
Maximum Target Size 200 x 100 pixels
Tracking Rate 4 pixels/field
(This is the tracking rate with respect to the sensor field of view ‘open loop
video’ that can be obtained for all targets. For larger targets and track
windows the AVT will be capable of tracking higher rates)
Target Detection The AVT has the following target detection performance:

Minimum Target Contrast 5%
Minimum Signal to Noise Ratio 4
Minimum Target Size Variable Down to 1x1 Pixel
Maximum Target Size Variable Up to 200 x 100 Pixels
RMS Noise on Target Position < 1.0 Pixel
Maximum Number of Targets 5
Minimum Target Speed Variable
Maximum Target Speed Variable
Definitions The following definitions are used:
Contrast
(Average Target Intensity ‐ Average Background Intensity) x 100
Peak White Intensity
Signal to Noise
Ratio
(Average Target Intensity ‐ Average Background Intensity)
RMS Noise
Document No: IN95-0005-01 Edition 4B Tracker Performance 67

13 • AVT Control Software
Overview The AVT Controller for Windows (avt.exe) is a stand‐alone Microsoft Windows
x32 utility that provides an easy to use interface for accessing the AVT
configuration and status data. The access is structured in a series of User Screen
menus that group Command and Status data by function. The content of these
User Screens is defined by a Script File provided with the installation. The Script
File is an editable text file that can be modified by Users.
Installation The AVT executable file (avt.exe) is provided on the CD‐ROM.
1. Copy the “CDROM:\Tools” directory and all of its contents to a directory on
the PC Hard Disk Drive. As well as copying the executable file “avt.exe”,
this will also copy the product script files and the Compiled HTML Help File
(AVT.chm).
2. From the “CDROM:\Tools\scripts” directory, choose the AVT HW product
script file required. Copy this into the “AVT_control” directory on the PC
Hard Disk Drive.
The first time the program is executed it will:
1. Display a Window titled “Select Product Type”. The User should select the
appropriate AVT product type.
2. Display a Window titled “Select AVT Script File”. The User should select the
script file copied earlier (which has a *.txt extension).
3. Default to using COM1 at 115200 baud rate (comms port settings can be
changed by selecting menu option <Configure><Comms Port…>).
Operation The main window of the program provides the primary control functions such as
enter/exit track, algorithm selection, window sizes.
Function Keys A Programmable Function Key (PFK) file can be selected for use with the

program. This file allows one or more AVT configuration items to be allocated to
the function keys F1 ‐ F12 on the keyboard for easier use. The PFK file is a text file
that can be edited by Users. An example is included in the installation directory.
Joystick or The PFK file can be configured to enable a joystick or gamepad to be used with
Gamepad the program. Joystick or gamepad buttons may then be mapped to one or more
AVT configuration items, and the joystick axis can be allocated to control
functions such as Platform demands and Acquisition Window position.
Help Selecting the <Help><Contents…> Menu option provides access to the Compiled
HTML Help File, which describes further functionality.
Typical User Figure 13-1 Typical ADEPT AVT Control User Interface
Interface
Document No: IN95-0005-01 Edition 4B AVT Control Software 69

14 • WinMon Utility
Overview WinMon (WinMon.exe) is a stand‐alone Microsoft Windows x32 utility that
provides a general purpose communications interface to the ADEPT family of
trackers serially or via PCI. It can:
1. Provide a software download utility for reprogramming AVT software.
2. Send/receive low level communications to and from the tracker.
Installation The WinMon executable file (WinMon.exe) is provided on the CD‐ROM.
From the “CDROM:\Tools” directory, copy the executable file “WinMon.exe”
and the Compiled HTML Help File “WinMon.chm” to a directory on the PC
Hard Disk Drive.
The first time the program is executed, it will default to using COM1 at 38400
baud rate. Comms port settings can be changed by selecting menu option
<Configure><Comms Port…>)
Help Selecting the <Help><Contents…> Menu option provides access to the Compiled
HTML Help File, which describes further functionality.
AVT Software 1. Start the “WinMon.exe” PC application.

Update Via Serial 2. Connect the AVT to the PC using a serial cable.
Interface
3. Start WinMon and make sure <Configure><Comms Port…> is set for:
– 115200 baud (or 38400 baud for older software sets)
– No parity
– 8 data bits
– 1 stop bit
4. Select <Download><Connect to AVT...> (or press <ALT><S>) and then reset
the tracker.
5. After displaying version numbers, a prompt will appear to press a key to
download application.
6. Press a key.
7. A prompt will appear to select the download file.
8. Select <Download><Application File to AVT...> (or press <PageUp>) and then
select the file. It will be a “***.run” file.
9. The new application should download, and automatically program into
Flash.
10. After a successful download and program, the new application will be
launched.
11. The new application software has now been installed.
Figure 14-1 Typical WinMon Interface
Document No: IN95-0005-01 Edition 4B WinMon Utility 71

A • Data Interfaces
Overview The ADEPT AVT range of products have the following physical data interfaces:
ADEPTxx Serial VME PCI
ADEPT64 4 
ADEPT60 4 
ADEPT104 4 PCI or ISA
AIM12 4
ADEPT74 4 PCIe
Each of these interfaces can be used for data communication between the AVT
and host system. The communication allows for commands to be issued to the
AVT and status data to be returned by the AVT to the host system.
The ADEPT AVT supports several different communication protocols, which are
supported on each of the physical data interfaces:
Simple Protocols:
1. COMMAND and STATUS Block Protocol.
2. Datalog Messages (not via DPR).
Structured Protocols:
1. External Command; Symbology Interface.
2. External Command; Component Interface.
A.1 Serial Interfaces

The ADEPT AVT has up to four full duplex serial interfaces with the following
characteristics:
Format Asynchronous
Data Rate 115 200 Baud (typical)
Parity None
Data Bits 8
Start Bits 1
Stop Bits 1
A.1.1 COMMAND and STATUS Block Protocol
Serial Channel 1 is typically configured as the Maintenance Port, providing full
access to the COMMAND and STATUS data.
The COMMAND and STATUS blocks are used both to input commands to the
AVT and also to query AVT status. They are accessed using the following data
packet, which is used to transmit both commands and status requests to the AVT,
and also to receive status data from the AVT.
Byte Content
1 Sync Flag (0x55)
2 COMMAND Block BLOCK
3 COMMAND Block FIELD
4 Data (most significant byte)
5 Data
6 Data
7 Data (least significant byte)
8 Checksum (modulo 256 sum of bytes 2 to 7)
Command Cycle To issue a command, the Sync Flag is set to 0x55 and the corresponding Block

number, Field number and Data of the Command is input to the AVT in the above
format.
LINK
For legal values of BLOCK, FIELD and DATA See “Command Reference Manuals” on page 129. The
commands are actioned within two video fields of their receipt.
NOTE
When issuing a command, the serial data packet is ordered so that the most significant byte of the
command data is placed in byte 4 Data (most significant byte) of the serial data packet. This format is
followed regardless of the number of bytes of data required for the command.
Status Cycle To read‐back the status of any one of the COMMAND or STATUS Block variables,

the following sequence is used:
1. Input a Status request to the AVT:
Byte Content
1 Sync Flag (0x55)
2 Block IDENT = 255
3 Block FIELD = 0
4 Data = 0
5 Data = 0
6 Data = Required Block IDENT
7 Data = Required Block FIELD
Document No: IN95-0005-01 Edition 4B Data Interfaces 73

2. The AVT responds to the Status request, within two video fields, with the
following data:
Byte Content
1 Sync Flag (0x55)
2 Required Block IDENT
3 Required Block FIELD
4 Data (most significant byte)
5 Data
6 Data
7 Data (least significant byte)
A.1.2 Datalog Messages

It is possible to output a Datalog message with the following format through any
one of the serial links:
Byte Content
1 Sync Flag (0x12)
2 Field Count (modulo 256)
3 Sync Flag (0x34)
4 AVT Status
5 X Output (most significant byte)
6 X Output
7 X Output
8 X Output (least significant byte)
9 Y Output (most significant byte)
10 Y Output
11 Y Output
12 Y Output (least significant byte)
Table A-1 Output Datalog Packet

Byte Data Value
1 0x12 Sync Flag
2 0 to 255 Field Count
3 0x34 Sync Flag
4 Bit 7 = 1 AVT Mode = Autotrack
Bit 7 = 0 AVT Mode = Acquisition (Manual or Target Detect)
Bit 6 = 1 Window Size Mode = Automatic
Bit 6 = 0 Window Size Mode = Manual
Bit 4 = 1 AVT Mode = Breaklock (in Autotrack Mode)
Bit 4 = 0 AVT Mode = Not Breaklock (in Autotrack Mode)
5 to 8 X Output. The format and content of the X Output data is selectable
9 to 12 Y Output. The format and content of the Y Output data is selectable
Datalog Enable The Serial Datalogging Enable command (Block 9, Field 2) is used to
enable/disable the datalog output.
Datalog Port The Serial Datalogging Channel command (Block 9, Field 3) is used to specify the

Serial Port used for data logging. Only one serial port can be used for data
logging.
Datalog Output The content of the X and Y output data is set via the Select Datalogging Message

Format Format Command (Block 9, Field 4).
Value Data Content
0 32-bit IEEE floating point format representation of X and Y axis platform demand
(includes FOV compensation and platform filters)
1 Fixed point (32,12) format representation of X and Y axis platform demand
(includes FOV compensation and platform filters)
2 32-bit IEEE floating point format representation of X and Y axis target to Boresight errors
3 Fixed point (32,12) format representation of X and Y axis target to Boresight errors
Datalog Timing The Serial Datalogging Output synch mode command is used to synchronize the

(Block 9, Field 1) datalog output to one of the following events:
1. Video flyback.
NOTE
The data output may have been calculated more than one video field previously, depending on the
position and size of the track window.
2. Completion of the image processing function and calculation of the status
data. This can occur at any point within the video field.
A.1.3 External Command Protocol

The Maintenance Port can also be used to update AVT configuration items
through the External Commands protocol. External Commands sent over the
serial interface to the AVT are sent as a packet of consecutive bytes. Each External
Command Packet has the following format:
Byte Content
0 Sync Flag (0xAA)
1 to n External Command i.e. ID, data length, data
n+1 Packet Checksum (modulo 256 sum of bytes 1 to n)
Following receipt of each External Command the AVT returns an External
Command Response. This is a packet of consecutive bytes. Each External
Command Response Packet has the following format:
Byte Content
0 Sync Flag (0xAA)
1 to n External Command Response i.e. ID, data length, data
n+1 Packet Checksum (modulo 256 sum of bytes 1 to n)
A.1.4 Remote Control Interface

One of the serial links can be used to interface to a Remote Control Unit (RCU).
NOTE
Refer to the Remote Control Units section in the ADEPT60 or ADEPT64 AVT Hardware Reference Guide.

A.2 DPR Interface
Introduction The AVT DPR Interface allows access to an area of the on‐board SRAM
designated the Dual Port RAM. The DPR Interface provides full access to:
1. COMMAND and STATUS blocks.
2. External Commands.
3. Control Registers.
Memory The Dual Port RAM data is organized on a byte‐addressable basis, where lower
addresses correspond to higher bytes (big endian). A 32‐bit data element that is
stored in 4 bytes of memory will therefore have the most significant byte (MSB)
located at address N and the least significant byte (LSB) located at address N+3 as
shown below:
Address Byte Number N N+1 N+2 N+3
3 2 1 0
MSB LSB MSB LSB MSB LSB MSB LSB
D31 D24 D23 D16 D15 D8 D7 D0
Interrupts Two interrupts are used to facilitate flow control:
• The AVT Interrupt is a single interrupt raised by the host to initiate transfer
of Command data to the AVT
• The DPR Controller Interrupt is a single interrupt raised by the AVT to the
host through the DPR interface to initiate transfer of Output data to the host
Interface Refer to the HW Reference Manual for specific details of the DPR Interface.
A.2.1 COMMAND and STATUS Block Protocol
The COMMAND and STATUS Block Protocol uses an area of DPR that is
structured as shown in the table below. This uses an organization that partitions
the DPR into 32 pages, each of 64 bytes. The pages are designated 0 to 31.
Data transfer through the DPR is organized in 64‐byte Data Transfer Blocks that
are loaded into the 64‐byte pages of the DPR. Each Data Transfer Block is
designated as either a COMMAND (Input) or STATUS (Output) data block and
each block is allocated a unique identifier or Block ID, which defines its format.
Table A-2 Dual Port RAM Data Structure
Address Offset Size (Bytes) Content
0x04 0000 64 Page 0 (DPR Status)
0x04 0040 64 Page 1
0x04 0080 64 Page 2
0x04 00C0 64 Page 3
0x04 0100 64 Page 4
0x04 0140 64 Page 5
0x04 0180 64 Page 6
0x04 01C0 64 Page 7
0x04 0200 64 Page 8
0x04 0240 64 Page 9
0x04 0280 64 Page 10
0x04 02C0 64 Page 11
0x04 0300 64 Page 12
0x04 0340 64 Page 13
0x04 0380 64 Page 14
0x04 03C0 64 Page 15
0x04 0400 64 Page 16
0x04 0440 64 Page 17
0x04 0480 64 Page 18
0x04 04C0 64 Page 19
0x04 0500 64 Page 20
0x04 0540 64 Page 21
0x04 0580 64 Page 22
0x04 05C0 64 Page 23
0x04 0600 64 Page 24
0x04 0640 64 Page 25
0x04 0680 64 Page 26
0x04 06C0 64 Page 27
0x04 0700 64 Page 28
0x04 0740 64 Page 29
0x04 0780 64 Page 30
0x04 07C0 64 Page 31 (DPR Command)

NOTE
The ADEPT104 and ADEPT74 do not have 04 address offset.
The operation of the DPR is controlled by one special STATUS block designated
DPR STATUS Block and one special COMMAND block designated DPR
Command Block, which are loaded in Page 0 and Page 31 of the DPR respectively.
The location of these two blocks remains fixed.
The remaining thirty Data Transfer Blocks, pages 1 to 30, are used for the transfer
of command and status data between the AVT and the DPR and may contain any
of the COMMAND or STATUS data blocks.
The configuration of the thirty pages is controlled by the DPR Controller, which
defines the configuration of the DPR through the Page Allocation Request fields
of the DPR Command Block.
STATUS Blocks STATUS data blocks output data from the AVT. Each STATUS block is partitioned

into sixteen 4‐byte fields as shown below.
Table A-3 STATUS Data Block Format
Block Ident Range: 129 To 255 Block Name:
Field Size (Bytes) Contents
0 4 Tags
1 4 Field 1 Data
2 4 Field 2 Data
3 4 Field 3 Data
4 4 Field 4 Data
5 4 Field 5 Data
6 4 Field 6 Data
7 4 Field 7 Data
8 4 Field 8 Data
9 4 Field 9 Data
10 4 Field 10 Data
11 4 Field 11 Data
12 4 Field 12 Data
13 4 Field 13 Data
14 4 Field 14 Data
15 4 Field 15 Data
A single flag in the DPR Command Block is used to control access to the STATUS
data blocks. This is set by the AVT to transfer control of the STATUS blocks to the
DPR Controller and also indicate that all the STATUS blocks contain valid data.
The flag is cleared by the DPR Controller to transfer control of the STATUS blocks
back to the AVT. The AVT raises the DPR Controller Interrupt when the STATUS
data block data becomes valid.
DPR STATUS Block The DPR STATUS block is partitioned as shown below:
(Page 0) Table A-4 Block ID 128, DPR STATUS
Block Ident: 128 Block Name: DPR Status
Byte Command
0 Field Counter
1 to 3 Reserved
4 to 35 Current Page Allocations
36 to 39 Reserved
40 to 43 BIT Result
44 to 63 Reserved
Field Counter The field counter provides a modulo 256 count that is incremented at the

incoming video field sync rate (50/60Hz). The field counter starts counting when
the AVT initialization is complete. The contents of the DPR is not valid and must
not be modified until the field counter has started counting.
Current Page The Current Page Allocation block reports the contents of the 32 pages in the

Allocations DPR. The page allocation consists of a 32‐byte block where each of the bytes is
used to identify the COMMAND or STATUS block, which occupies each of the 32
pages.
The Page 0 identifier is held in the first byte of the block, and consecutive page
identifiers are held at consecutive locations in the block up to 32 bytes. The
Current Page Allocations are updated during each Status cycle.
The first and last page allocations are reserved for the DPR status block and DPR
Command Block respectively.
BIT Result The BIT Result indicates the result of the Built In Test. The BIT Result should be

read after the Field Counter has started incrementing. Any non‐zero value in
these four bytes indicates a BIT failure. The format of the BIT Result is as follows:
Byte 3 2 1 0
BIT Test Area BIT Test Number BIT Test Status Code
BIT Test Area This item identifies the resource that failed the BIT test as follows:

0x00 Background
0x01 Histogram Memory
0x02 LUT Memory
0x03 Graphics Memory
BIT Test Number This item identifies the test number that failed for the resource.
BIT Test Status This item identifies the Status Code of the test that failed as follows:

Code
0x0000 No Failure
0x0001 Memory Read/Write Verification Failure

COMMAND Blocks COMMAND data blocks input commands to the AVT. Each COMMAND block is
partitioned into sixteen 4‐byte fields as shown below.
Table A-5 Command Block Data Structure
Block Ident Range: 1 to 127 Block Name:
Field Size (Bytes) Contents
0 4 Tags
1 4 Field 1 Data
2 4 Field 2 Data
3 4 Field 3 Data
4 4 Field 4 Data
5 4 Field 5 Data
6 4 Field 6 Data
7 4 Field 7 Data
8 4 Field 8 Data
9 4 Field 9 Data
10 4 Field 10 Data
11 4 Field 11 Data
12 4 Field 12 Data
13 4 Field 13 Data
14 4 Field 14 Data
15 4 Field 15 Data
A system of ‘tags’ that are located in the first field, field 0, of each COMMAND
block are used to indicate the fields that contain a command to be executed by the
AVT. The field 0 tag is held in the least significant bit of the 32‐bit field 0, and
consecutive tags are held in the consecutive bits of field 0 up to bit 15.
Where a command occupies more than one field, any one of the tags
corresponding to the fields within the command may be used. The tag is set to 1
by the DPR Controller to indicate that the command in the respective field is valid
and should be processed by the AVT. The tag is cleared to 0 by the AVT when it
has accepted the command data.
A single flag in the DPR Command Block is used to control access to the
COMMAND data blocks. This is set by the DPR Controller to transfer control of
the COMMAND blocks to the AVT and also indicate that all the COMMAND
blocks contain valid data. The DPR Controller raises the AVT Interrupt to request
a Command Cycle. The flag is cleared by the AVT to transfer control of the
COMMAND blocks back to the DPR Controller.
DPR Command The DPR Command Block is partitioned as shown below:

Block (Page 31) Table A-6 Block ID 0, DPR COMMAND
Block Ident: 0 Block Name: DPR Command
Field Command
0 Tags
1 to 8 Page Allocation Requests
9 to 14 Reserved
15 Control Flags
Page Allocation The Page Allocation Requests command defines the contents of the 32 pages in
Requests the DPR. It consists of a 32‐byte block where each of the bytes is used to identify
the COMMAND or STATUS block, which occupies each of the 32 pages.
The Page 0 identifier is held in the first byte of the block and consecutive page
identifiers are held at consecutive locations in the block up to 32 bytes. Page 0 and
Page 31 contents are fixed and any change requests for these pages will be
ignored. Page allocation requests, indicated by the appropriate tag fields, are
processed following receipt of the AVT Interrupt.
Control Flags The control flags field contains the following two flags:

STATUS Block Flag (offset address 0407FC)
The STATUS Block Flag is located in the first byte of the Control Flags field and
indicates the state of the STATUS blocks. This flag is set to a non‐zero value by the
AVT to indicate that the data in all the STATUS Blocks is valid.
The flag should be reset to 0 by the DPR Controller when all the status data
required has been read from the DPR. The AVT will not update the status data
blocks until this flag has been cleared to 0.
Command Block Flag (offset address 0407FD)
The Command Block Flag is located in the second byte of the Control Flags field
and indicates the status of the COMMAND blocks. This flag is set to a non‐zero
value by the DPR Controller to indicate that the data in all the COMMAND blocks
is valid. The AVT then reads in the data fields marked by valid tags.
The Command Block Flag is reset to 0 by the AVT when it has completed access of
the COMMAND blocks (this does not indicate that all commands have been
accepted). The DPR Controller must not update the command data blocks until
this has been cleared to 0.
NOTE
The ADEPT104 and ADEPT74 do not have a 04 address offset.
COMMAND Data The AVT reads data in the COMMAND blocks on receipt of an ADEPT Interrupt.

Blocks (Timing) Processing of the command data read in from the COMMAND blocks is
configurable to be synchronized to one of the following events:
• Video flyback
• Completion of the image processing function and calculation of the status
data
NOTE
This can occur at any point within the video field.
• Receipt of the AVT Interrupt raised by the DPR Controller
The AVT is capable of accepting an average of 30 commands per video field. If the
number of commands exceeds this, then the AVT may delay accepting the
commands. This is indicated by the respective tags, which will remain set after
the COMMAND cycle has been executed.
There is no pre‐defined sequence of command execution, so to guarantee the
update rate of a command, the rate at which the commands are issued must be
controlled.

STATUS Data The execution of an AVT STATUS cycle is configurable to be synchronized to one
Blocks (Timing) of the following events:
• Video flyback
NOTE
The data output may have been calculated more than one video field previously depending on the
position and size of the track window.
• Completion of the image processing function and calculation of the status
data
NOTE
This can occur at any point within the video field.
The STATUS cycle is only executed if the STATUS Block Flag has been cleared
from the previous cycle. If it is set, then the STATUS cycle is abandoned.
To ensure that STATUS blocks are updated every field, the STATUS Block Flag
must be cleared to indicate the status data has been accepted no longer than 10ms
after it has been set by the AVT.
A.2.2 Example of Command and Status Protocol
Overview This section explains how the DPR interface is used to access COMMAND and
STATUS blocks.
Address Generation
NOTE
The ADEPT104 and ADEPT74 do not have an 04 address offset.
Figure A-1 Address Generation

Offset from
Page Address
Offset from
Page allocations Tag field x00
Base Address
Command / Status Field 1 x04
x040000 Page 0 Block 128 Command / Status Field 2 x08
x040040 Page 1 Block a Command / Status Field 3 x0C
x040080 Page 2 Block b Command / Status Field 4 x10
x0400C0 Page 3 Block c Command / Status Field 5 x14
Command / Status Field 7 x1C
x040700 Page 28 Block x Command / Status Field 8 x20
x040740 Page 29 Block y Command / Status Field 9 x24
x040780 Page 30 Block z Command / Status Field 10 x28
x0407C0 Page 31 Block 0 Command / Status Field 11 x2C
Command and
Interface Pages Command / Status Field 13 x34
Status Blocks
Command / Status Field 15 x3C
Data Fields within

a Block
Writing Commands 1. Check that the AVT has completed the last command cycle. This is indicated

to the AVT by the COMMAND Block Flag being cleared (to zero) by the AVT.
2. Confirm that the Blocks containing the required commands are mapped into
memory by reading the Current Page Allocation locations
(See Reading Status Data from the AVT).
3. If the Blocks are not mapped in, then update the COMMAND Block Page
Mapping.
4. Send the command to the AVT by writing the command data into the
appropriate field of the mapped Command Page.
5. Indicate to the AVT that new data is in that command field by setting the
appropriate tag bit in field 0 of the same Command Page.
6. Indicate to the AVT that new commands are available by setting the
COMMAND Block Flag and the AVT Interrupt.

Figure A-2 Command Cycle Actions
Issuing
Isusing Commands
Commands
Check that last

Command Cycle is
complete
Are the Blocks

Update the Command
mapped in to the No
Block page mapping
interface?
Yes
Start For Each

Command
Set Command Data into

appropriate field of the
mapped page
Set the associated tag bit

in field 0 of the same
page
End For Each Command
Set the Command Block

Flag
Set the Interrupt
Updating the 1. Write the Page Allocation Requests to page 31. This is achieved by writing
COMMAND Block the numbers of the required COMMAND Blocks into the locations
Page Mapping appropriate for the Interface Pages that you wish them to be mapped to. For
example to map Block 7 to page 5, write the value 0x07 to byte 1 of field 2 of
page 31 (hex offset xx07CA).
2. Set the appropriate tag bits for the command fields containing the new Page
Allocations. For example, when mapping Block 7 to page 5, command field 2
of page 31 is used, so it is necessary to set bit 2 (0x04) of the tag field (field 0)
of page 31 (hex offset xx07C0).
NOTE
When setting tag bits for Page Allocation Requests, it is important to realise that the Command Field
associated with each tag bit contains four page allocations. It is therefore important to ensure that all 4
Page Allocation Requests in a modified Command Field are set to the users required Block numbers.
3. Indicate to the AVT that new commands are available by setting the
COMMAND Block Flag and then the AVT Interrupt.
4. Wait for the AVT to complete the current command cycle. This is indicated
by the COMMAND Block Flag being cleared by the AVT.
5. Wait for the AVT to complete the next status cycle and thus update the
Current Page Allocations. Completion of the status cycle is indicated by the
STATUS Block Flag being set by the AVT.
6. Confirm that the required block now appears in the Current Page
Allocations.
Figure A-3 Update Interface Block Mapping
Update Block to Page

Mapping
Write Page Allocation

Requests to Page 31
Set the appropriate tag

bits in Field 0 of Page 31
Set the Command Block

Flag
No
Wait for AVT to
complete the Command
Cycle
Wait for a Status Cycle

to complete
Are the Blocks

mapped in to the
interface?

Verifying Check that the AVT has completed the command cycle. This is indicated by the
Commands Have COMMAND Block Flag being cleared (to zero) by the AVT. Check that the tag bits
Been Read By the have been cleared by the AVT.
AVT
Reading Status 1. The setting of the DPR Controller Interrupt by the AVT indicates that new

Data from the AVT Status data is available.
2. Check that the AVT has completed writing Status Data to the DPR. This is
indicated by the STATUS Block Flag being set by the AVT.
3. Read the Current Page Allocations from page 0.
4. Confirm that the Blocks containing the required status items are mapped to
the DPR interface.
5. If the Blocks are not mapped in then:
a. Complete the Status Cycle by clearing (to zero) the STATUS Block Flag.
b. Update the COMMAND Block page mapping.
c. Wait for the AVT to indicate that fresh status data has been written to the
DPR interface.
6. Read the required Status Data from the mapped Status Page.
7. Complete the Status Cycle by clearing (to zero) the Status Block Flag.
Figure A-4 Status Cycle Actions
Reading Status Data
Check that the AVT has

completed writing Status
Data
Are the Blocks

Update the Command
mapped in to the No
Block Page mapping
interface?
Yes
Read the Status Data

from the appropriate
Interface Pages
Clear the Staus Block

Flag to indicate the data
has been read
A.2.3 External Commands Protocol
The External Commands uses an area of DPR which is structured as shown in the
table below:
Table A-7 External Command Input/Output Area
Address Offset Size (Bytes) Content
0x04 0800 512 External Command Input Area
0x04 0A00 512 External Command Output Area
NOTE
The ADEPT104 and ADEPT74 do not have an 04 address offset.
When using the DPR interface to send an External Command to the AVT, the
following sequence of actions is taken by the DPR Controller:
• The first byte of the DPR External Command Input Area is read. If this does
not contain zero, then the AVT has not yet read the previous External
Command and another cannot be sent; otherwise the DPR Controller may
proceed
• The External Command, i.e. ID, Data Length and Data, is written to
consecutive bytes of memory, with the External Command ID being written
to the second byte of the DPR External Command Input Area
• A non‐zero value is written to the first byte of the DPR External Command
Input Area
• Any data value is written to the AVT Interrupt Register to generate an
interrupt to the AVT
The AVT generates a DPR Controller Interrupt to inform the DPR Controller
when an External Command Response should be read. The AVT will generate a
response to each command received. However, since the AVT may also generate
automatic notifications, it is quite possible for several responses to be queued in
the External Command Output Area.
The organization of the External Command Output Area is as follows:
Description Offset Length Value Comment
Response Flag 0 1 0 No Response available
(AVT has control of Output Area Controller).
N N responses are available to DPR Controller
(DPR has control of Output Area)
External Response 1 Variable CMD_ID to byte before CKSUM
Additional Responses 1 N-1 Count of further available responses
External Response Variable CMD_ID to byte before CKSUM
Additional Responses 1 N-2 Count of further available responses
:
Additional Responses 1 1 Last available response follows
External Response Variable CMD_ID to byte before CKSUM
Additional Responses 0 No more responses
Unused Remainder of fixed length Output Area
The responses will be queued in chronological order and the DPR Controller
should attempt to consume them and restore memory access to the AVT as soon
as possible.

NOTE
Although each response will be preceded by a count of remaining responses, only the Response Flag in
the first byte of the External Command Output Area will be used for memory access handshaking.
Once the Command Response has been copied, or processed, i.e. the DPR
External Command Output Area is again available for use by the AVT. Zero must
be written to the first byte of the DPR External Command Output Area.
B • External Commands
B.1 Introduction
The AVT supports a set of External Commands that may be issued to the AVT via
any of the supported external physical interfaces (currently Serial and DPR).
A User issues External Commands to the AVT via a supported interface and each
of these External Commands may cause the generation of one or more External
Command Responses by the AVT. The External Command Response(s) to an
External Command is passed to the User using the same physical interface used
to send the External Command.
B.2 External Command Format

Serial Link Header 8‐bit value which identifies the serial link header.
Not applicable for DPR.
Serial Link Tail 8‐bit checksum, modulo 256, of serial link data excluding the Serial Link Header.

Not applicable for DPR.
Command ID This 8‐bit value identifies which External Command is being issued
Command Data This 16‐bit value defines the number of bytes of Command Data that follows.

Length Some commands have no data, some have a fixed amount of data, and some have
a variable amount. The first byte contains the most significant 8 bits of the value;
the second byte contains the least significant 8 bits.
Command Data These bytes contain the data required to fully define the actions required by the

command. Its format and length is dictated by the specified Command ID. Data
values contained within the External Command Definition that are more than one
byte in length are ordered most significant byte(s) first.
Serial Interface
Byte Content
0 Serial Link Header (Sync flag)
1 Command ID
2 and 3 Command Data Length
4 to n Command Data
n+1 Serial Link Tail (Packet checksum)
DPR Interface
Byte Content
0 DPR Command Flag (DPR access control)
1 Command ID
2 and 3 Command Data Length
4 to n Command Data
Document No: IN95-0005-01 Edition 4B External Commands 89

B.3 External Command Response Format
Command ID This 8‐bit value identifies the External Command for which this is a response.
Response Data This 16‐bit value defines the number of bytes of Command Response Data that

Length follows. The first byte contains the most significant 8 bits of the value; the second
byte contains the least significant 8 bits.
Response Data These bytes contain the data defining the response. Its format and length is

dictated by the specified Command ID. Data values contained within the External
Command Response Definition that are more than one byte in length are ordered
most significant byte(s) first.
Serial Interface
Byte Content
0 Serial Link Header (Sync flag)
1 Command ID
2 and 3 Command Response Length
4 to n Command Response
n+1 Serial Link Tail (Packet checksum)
DPR Interface
Byte Content
0 DPR Response Flag (count of number of responses)
1 Command ID
2 and 3 Command Response Length
4 to n Command Response
n+1 Additional Responses etc
B.4 Line of Sight Data Command
Summary This command allows a Platform to send pointing position and rate of change to
the AVT.
Command ID 0x11, 17 decimal.
Name Offset Length Value Description
Serial Link Header (see Note 1)
SYNC 0 1 0xAA Synchronization code
External Command Header
CMD_ID 1 1 0x11 Identify LOS command
LENGTH 2 2 0x0015
Specific Mandatory Command Data
4 1 Mask bit set to 0 = Data not valid
bit set to 1 = Data valid
bit 0 = Latency Time Component Validity
bit 1 = Azimuth Rate Component Validity
bit 2 = Elevation Rate Component Validity
bit 3 = Azimuth Position Component Validity
bit 4 = Elevation Position Component Validity
5 4 Fix 12 LOS Latency Time
9 4 Fix 12 LOS Azimuth Rate (milliradians per second)
13 4 Fix 12 LOS Azimuth Rate (milliradians per second)
17 4 Fix 12 LOS Azimuth Position (milliradians)
21 4 Fix 12 LOS Azimuth Position (milliradians)
Serial Link Command Tail (See Note 1)
CKSUM 25 1 Checksum modulo 256 (excluding SYNC, CKSUM)
NOTES
1. Serial Interface only.
2. This definition is for the Serial Interface message and hence contains the additional bytes for Serial
Link Header and Tail.

B.5 Component Interface Commands
NOTE
The definitions provided in the following sections are for Serial Interface messages and hence contain
the additional bytes for Serial Link Header and Tail.
Functional Access to configuration, real time data and status data is only possible through
Overview the External Command Interface by first registering an interest. Each registration
may contain a set of up to 16 data items, which may be picked and mixed
according to requirements. However, since all subsequent data transfers will
always involve the entire set, care should be taken to achieve an appropriate
selection.
Components may be grouped to serve several requirements. For frequent
transfers, data redundancy should be avoided by only including those likely to
experience change at a similar frequency. To guarantee the integrity of integral
units, their constituent components should be included in the same registration.
Alternatively, for infrequent use, it may be sufficient to merely group by logical
association.
Once registered, a ‘component set transfer’ can be achieved on demand or, in the
case of status data, automatically every time a constituent component experiences
a change in value. For simplicity, one off or occasional transfers may be achieved
by transient registrations that will automatically generate a single transfer and
then expire.
The data order and format type of each ‘component set transfer’ will always be in
accordance with the effective mask defined by the parent registration.
In the interest of ordered operation and efficiency, new commands should not be
initiated until previous ones have received an acknowledgement response.
Likewise, registrations that are no longer needed should be explicitly cancelled.
Time‐outs may sensibly be applied to all acknowledgements as protection against
possible communication failures.
B.5.1 Component Interface Command Definitions
Command and The following table summarizes the list of available external commands for the
Response Summary component interface.
Name Communal ID Description AVT Response
ECID_CSREG 0x30 Register a new Component Set General Response Packet or
Read Response Packet
ECID_CSDREG 0x31 Cancel a Component Set General Response Packet
Registration
ECID_CSRD 0x32 Single Shot Read General Response Packet or
(previously registered Read Response Packet
Component Set)
ECID_CSUD 0x33 Single Shot Update General Response Packet
(previously registered
Component Set)
ECID_CSTRA 0x34 Transient Registration for General Response Packet or
immediate one time access Read Response Packet
Logical Protocol The external command interface operates with a ‘Logical Half Duplex Protocol’.
Commands must be submitted to the AVT singularly, and subsequent commands
must not be issued until the appropriate response message has been received. The
format of the response is dependant on the command issued.
AVT responses are issued on completion of the message processing, and occur
asynchronously to the video time base.
As a protection against communication failures, a response time‐out of 40 msec
can be applied.
Component Set Following a power cycle or system reset, the communications interface must be

Registration re‐initialized before use. This process is called ‘component set registration’, and
must be carried out on each interface that uses the external command protocol.
Additional registrations can be carried out at any time, and existing components
may be de‐registered if required.
Component Sets A ‘Component Set’ is a grouping of command and/or data fields that are related

by either context or frequency of use. This grouping is application specific, such
that only relevant items for a particular application are communicated across the
interface.
Constituent items for a component set must be chosen carefully to avoid data
redundancy, as all subsequent transfers will involve the entire set.
Registration Each Component Set is registered with the AVT with a specific registration
number that uniquely defines a grouping of commands and data. In addition,
several types of registration are available to the user, which may be used
dependant on application specific requirements.

AVT Data Each item of data communicated between the AVT and host system is referred to
Components as a ‘component’. The available data ‘components’ are listed in the Command
Reference Manuals, where each item can be uniquely identified by its ‘Block and
Field’ number.
The AVT uses four different component types as classified in the table below:
Component Type Description
Configuration Item AVT configuration data, available through Blocks 1 to 127
System Status AVT output data, available through Blocks 129 to 255
RTC Real Time Command, e.g. joystick input
RTC and System Status A real time command with an associated system status, e.g. Enter
Track (Block1 Field 1)
NOTE
A small subset of items in blocks 1 to 127 are defined as either RTC or RTC and System Status items.
Data Formats The Block/Field descriptions define the AVT native format of individual data

items. However, the external command protocol provides the capability to carry
out data transfers using non‐native number formats. i.e. using fixed rather than
floating point.
Users can specify the external interface (input and output) format for data items
using the CR_TYPE field of the component set registration. The following options
are available:
INTEGER 8‐bit, 16‐bit, 32‐bit
FLOATING 32‐bit,064‐bit
FIXED 20 integer bits,012 fractional bits
Where conversions between native and external formats are not exact, they will be
performed as accurately as possible.
Unless specified otherwise, data items larger than a single byte will always be
according to big‐endian conventions, e.g. most significant byte first, least
significant byte last.
There is no distinction, within the external command protocol, between signed
and unsigned integers, the user must refer to the block/field description and treat
the data accordingly. For example, the 8‐bit representation 0xFF can take the
values:
255 if unsigned
‐1 if signed
AVT Data Access Access to AVT data using the ‘external command interface’ is subject to
Rules read/write control based on the data type in question.
Any component set registration must adhere to the access limitations described in
the table below. Registrations that break the access rules of any of the data items
will be rejected.
Two levels of access are permitted:
• Read Only (RO) can be applied to any component (other than a Real Time
Command that has no associated system status item). It is intended to offer a
Write Protect function for otherwise updateable items (i.e. Configuration
and RTC Status components)
• Updateable (RW) allows both Read And Write access to the component set.
RW registrations that contain a system status item will be rejected
Table B-1 Access Limitation
Access Type
Component Type
RO RW
Configuration Item Accepted Accepted
System Status Accepted Rejected
RTC Rejected Accepted
RTC and System Status Accepted Accepted
Data Update Rates Writes:

External commands may be sent to the AVT at any time and frequency required,
subject to observing the logical half duplex structure of the command interface. It
is recommended however, that the maximum frequency of updates to individual
data items is restricted to the video field rate.
Reads:
AVT data items may be read at any time and frequency required, subject to
observing the logical half duplex structure of the command interface.
In addition, the ‘external command interface’, supports the use of automatic
notification. This function enables the AVT to notify the host when selected AVT
data items change. Notifications can be synchronized to the video field rate if
required.

B.5.2 Registering a New Component Set
The following command is used to register a new Component Set, and must be
sent before any read or write requests. The AVT will issue a response on receipt of
this command. Subsequent registration commands should only be submitted
once the AVT response has been received by the host system.
Command Name ECID_CSREG
Command ID 0x30
Command Type Register a Component Set for future access
Command Successful (CR_NOTIF = 0x00): General Response Packet
Command Successful (CR_NOTIF = 0xX1): Read Response Packet
AVT Response
NOTE
Only applies if the component set registration contains at least one readable
AVT ‘block and Field.
Command Successful (CR_NOTIF = 0xX2): General Response Packet
Command Failed: General Response Packet

SYNC 0 1 0xAA Synchronization code for message to Master gate
0xAB Synchronization code for message to Slave gate. See Note 1
CMD_ID 1 1 0x30 Command Type
LENGTH 2 2 6 + 3N Number of bytes (CR_ID to CR_TYPE (N))
Specific Command Data
CR_ID 4 1 1 – 254 Identify the registration (unique for specific interface)
CR_USE 5 1 0 Indicates Read Only (RO). See Note 2
1 Indicates Updateable (RW). See Note 3
CR_NOTIF 6 1 0x00 Indicates no automatic notification is required
0xX1 Indicates automatic notification of change is required. See Note 5.
0xX2 Indicates synchronized notification is required. See Note 5
CR_RSV 7 2 0x0000 Abaco internal use only – reserved for future use
CR_NBR 9 1 N (1–16) Number of component definitions (CR_CID to CR_TYPE)
Component Definitions (n= 1 to N)
CR_CID_1(n) 7+3n 1 Component ID upper byte.
For existing data this is the field of the ‘block and field’ address
CR_CID_29(n) 8+3n 1 Component ID lower byte.
For existing data this is the block of the ‘block and field’ address
CR_TYPE(n) 9+3n 1 0x00 Reserved – do not use
0x10 8-bit integer data type
0x30 32-bit fixed-point value, 20 integer bits, 12 fractional bits
0x40 64-bit floating point data type
0x90 16-bit integer data type. See Note 4
Serial Link Command Tail (see Note 6)
CKSUM 10 + 3N 1 Checksum modulo 256 (excluding SYNC, CKSUM)
NOTES
1. Only for use in Dual gate systems.
2. Not applicable to AVT ‘Real Time Command’ components.
3. Not applicable to AVT ‘System Status’ components.
4. The data types used should be appropriate for the AVT component in question.
5. See below for a description of this data field.
Automatic AVT data items except for RTC types that have been assigned to a component set
Notification registration, can be read by the DPR Controller at any time. In addition,
component set registrations can be configured to automatically report data
according to the value of the CR_NOTIF field (See AVT Data Access Rules).
Value of CR_NOTIF Notification Type Transmission Rate
0x00 None Polled
0xX1 On change On change
Periodic Every X+1 fields where
0xX2
X = upper nibble of CR_NOTIF
Automatic A change (within the AVT internal configuration space) to any constituent
Notification component will automatically trigger the transmission of a ‘Read Response’
Enabled message. The message will be structured in accordance with its Component Set
(CR_NOTIF = 0xX1): Registration ID.
This mode should only be used for data components that change on an irregular
or occasional basis.
Synchronized A ‘Read Response’ message will be transmitted by the AVT at a fixed time
Notification interval, regardless of the internal data state, i.e. no change is required. The
Enabled transmission of the message is either synchronized to the completion of video
(CR_NOTIF = 0xX2): processing task or video flyback.
Synchronization method is set by Block 123, Field 3.
The upper nibble of the CR_NOTIF value sets the maximum transmission rate of
the ‘Read Response’ message.
If X=0 AVT transmits every field
If X>0 AVT transmits every X+1 field

B.5.3 Cancelling a Component Set Registration
This command can be used to cancel a registration for a component set that is no
longer required. Cancelling a registration releases the registration number for
reuse.
Subsequent commands should only be submitted after a response from the AVT
has been received
Command Name ECID_CSDREG
Command ID 0x31
Command Type Cancel a Component Set registration
AVT Response General Response Packet

0 1 0xAA Synchronization code for message to Master gate
SYNC
CMD_ID 1 1 0x31 Identify a de-registration command
LENGTH 2 2 0x0001
4 1 1 to 254 Identify the existing single registration to be cancelled
CR_ID
255 Cancel all existing registrations for this user
NOTES
1. Only for use in dual gate systems.
LINK
See “AVT Data Access Rules” on page 95. and See “Automatic Notification” on page 97.
B.5.4 Single Shot Read Command
This command can be used to read the current state of a previously registered
component set. All items, except real time command items, can be accessed by use
of this command. Each response embodies all status items registered in the
component set.
has been received.
Command Name ECID_CSRD
Command ID 0x32
Command Type Read a tracker Status Component Set
Command Successful = Read Response Packet
AVT Response
Command Failure = General Response Packet

SYNC
CMD_ID 1 1 0x32 Identify a Component Set Read command
LENGTH 2 2 0x0001
CR_ID 4 1 1–254 Identify the existing single registration for the read. See Note 2
NOTES
1. Only for use in dual gate systems .
2. Component set must have been previously registered as a readable type. See Link below.
LINK
See “Registering a New Component Set” on page 96.

B.5.5 Single Shot Write Command
This command is used to update AVT command or data components that have
been previously registered as a component set.
has been received.
Command Name ECID_CSUD
Command ID 0x33
Command Type Update AVT Command or data Components
AVT Response General Response Packet

SYNC
0xAB Synchronization code for message to Slave gate. See Note 1.
CMD_ID 1 1 0x33 Identify a Component Set Write command
LENGTH 2 2 Variable Number of bytes (CR_ID to CR_DATN)
CR_ID 4 1 1-254 Identify the existing registration for the update. See Note 2.
CR_RSV 5 1 0 Reserved for future use
CR_DAT1 6 Variable New data per type for 1st registered component. See Note 3.
CR_DAT2 New data per type for 2nd registered component
: :
CR_DATN New data per type for Nth registered component
CKSUM 1 Checksum modulo 256 (excluding SYNC, CKSUM)
NOTES
2. Component set must have been previously registered as an updateable type. See Link below.
3. Data must be formatted in accordance with the original component registration.
LINK
See “Registering a New Component Set” on page 96.
B.5.6 Transient Component Set Registrations

To reduce the number of registrations required following a system reset, AVT
components (commands or data fields) that are infrequently used may be
accessed using a transient registration. Using this method, the registration and
data transfer occur in the same message cycle.
Following receipt of the command, an AVT response is formed and the
registration is immediately cancelled.
Command Name ECID_CSTRA
Command ID 0x34
Command Type Single shot access to command or data components
AVT Response Command Successful (type TR) = Read Response Packet
Command Successful (type TW) = General Response Packet
Command Failure (TR or TW) = General Response Packet

SYNC 0 1 0xAA Synchronization code for message to Master gate.
CMD_ID 1 1 0x34 Identify a Transient Registration command
LENGTH 2 2 Number of bytes (CR_ID to CR_TYPE(N) or CRDATN)
Specific Mandatory Command Data
CR_ID 4 1 1-254 Identify the registration. See Note 2
CR_USE 5 1 2 Indicates Transient read (TR). See Note 3.
3 Indicates Transient write (TW)
CR_NOTIF 6 1 0 N/A to transient registration – reserved for future use
CR_RSV 7 2 0x0000 Abaco internal use only – reserved for future use
CR_NBR 9 1 N (1-16) Number of component definitions (CR_CID to CR_TYPE)
Component Definitions (n= 1 to N)
CR_CID_1(n) 7+3n 1 Component ID upper byte.
For existing data this is the field of the ‘block and field’ address
CR_CID_2(n) 8+3n 1 Component ID lower byte.
For existing data this is the block of the ‘block and field’ address
CR_TYPE(n) 9+3n 1 0x00 Reserved – do not use
0x30 32-bit fixed-point value, 20 integer bits, 12 fractional bits
Specific Conditional (CR_USE = TW) COMPONENT DATA (I-N)
CR_DAT1 10+3N Variable New data per CR_TYPE for first CR_CID component
CR_DAT2 New data per CR_TYPE for second CR_CID component
: :
CR_DATN New data per CR_TYPE for nth CR_CID component
NOTE
1. Only for use in dual gate systems
2. Must be a unique registration number for future use.
3. Use ‘TR’ for AVT status components.

B.5.7 AVT Response Format
The AVT responds to all external commands using a ‘General Response’ or a
‘Read Response’, as indicated in the command descriptions.
Both the ‘General Response’ and the ‘Read Response’ share a common structure,
with the exception that the latter contains data associated with the components in
question.
Any command that generates a failure code (except 0x00), returns a ‘General
Response’ containing the appropriate failure code.
Name General Response or Read Response (dependant on data structure)
ID N/A
Type AVT response to received external command
AVT Response N/A

SYNC
0xAB Synchronization code for response from Slave gate. See Note 1
External Command Response Header
CMD_ID 1 1 Identifies the external command being responded to
LENGTH 2 2 Number of bytes (STATUS to CR_ID or CRDATN)
Specific Response Data
4 1 0x00 Success
STATUS
else Failure (See Failure Codes on next page)
CR_ID 5 1 Identifies the user’s component set registration number. See Note 2
Specific Response Data (Read Response Only) (see Note 3)
CR-RSV 6 1 N/s Reserved for future use
CR_DAT1 7 Variable Read data per CR_TYPE for first CR_CID component of CR_ID
CR-DAT2 Read data per CR_TYPE for second CR_CID component of CR_ID
: :
CR_DATN Read data per CR_TYPE for Nth CR_CID component of CR_ID
Serial Link Command Response Tail (see Note 4)
NOTES
2. Response will always reference the originating ID CR_ID which dictates the possible formats.
3. ‘Specific response data’ is only included in a Read Response. A General response Packet will not
include the bytes CR_RSV to CR_DATN.
Failure Codes Failure status codes are returned as part of the ‘General Response’, and are set
according to the nature of the failure.
Mnemonic Value Description
- 0 Successful command
ECST_BAD_CMD_ID 1 Unrecognized external command
ECST_BAD_DATA_LEN 2 Data length specification error
ECST_ALLOC_FAIL 3 Allocation failure
ECST_BAD_DATA 4 Generally incorrect data
ECST_BAD_ID 5 Generally invalid ID
ECST_BAD_TYPE 6 Generally invalid type
ECST_UNREG_ID 7 Unregistered ID
ECST_INCORR_TYPE 8 Inappropriate type
ECST_BAD_COMP_MASK 9 Invalid component mask
ECST_BAD_FLAG 10 Invalid flag value
ECST_BIT_FAIL 11 n/a
ECST_GEN_FAIL 12 General (possibly system) failure
ECST_PB_BAD_PROGDEF 13 Invalid RCP program – Program header definition
ECST_PB_BAD_PRESSDEF 14 Invalid RCP program – Button Press group definition
ECST_PB_MAX_BUTTS 15 Invalid RCP program – Too many button definitions
ECST_PB_BAD_LAMP_STATDEF 16 Invalid RCP program – Lamp status definition
ECST_PB_BAD_BUTT_STATDEF 17 Invalid RCP program – Button status definition
18 Invalid RCP program – Too many combinations/
ECST_PB_MAX_CMBCMP
comparisons
ECST_PB_BAD_MAP 19 Invalid RCP program – Map group definition
ECST_PB_MAX_GRPS 20 Invalid RCP program – Too many group definitions
ECST_PB_DUP_GRP 21 Invalid RCP program – Duplicate group
ECST_PB_BAD_VAL 22 Invalid RCP program – Value group definition
ECST_PB_BAD_GRP 23 Invalid RCP program – General group definition
24-29 Reserved for RCP program expansion
ECST_CS_BAD_REG 30 Invalid CS registration – General format
ECST_CS_BAD_COMP 31 Invalid CS registration – Component unavailable
ECST_CS_BAD_DATA_TYPE 32 Invalid CS registration – Data type unrecognized
ECST_CS_DUP_REG 33 Invalid CS registration – Duplicate ID
ECST_CS_BAD_UD 34 CS update – invalid format
ECST_CS_FAIL 35 General CS (not internal) failure
ECST_CS_DENY 36 CS request (temporarily) denied
ECST_CS_FAIL_UD_ITEM 37 CS update – invalid data for 1 or more components
ECST_CS_UD_RO 38 CS update - cannot update a read only registration

C • User Symbology
Introduction The User may generate textual and graphical displays which are overlaid on the
Video Output signal produced by the AVT. This is achieved by sending the
required External Commands to the AVT using one of the supported external
interfaces, i.e. Serial or DPR Interface.
The following External Commands are involved with the generation of User
Symbology:
• User Symbol Registration
• User Symbol De‐registration
• User Symbol Update
• User Symbol Disable
The User Symbology is specified in units of User Symbols. Each User Symbol is
either a User Text Symbol (consisting of a string of characters) which all reside on
the same line, or a User Sprite Symbol which consists of a collection of pixels
defined by the User (which produces a shape on the screen).
The User is able to control each User Symbol independently of any other User
Symbol.
C.1 User Symbology Command Definitions
C.1.1 User Symbol Registration Command
Summary This command registers a User Symbol with the AVT, the User specifies a User
Symbol Type, either Text or Sprite, and the AVT responds with a Status Code and
a numerical User Symbol ID that the User can then use in subsequent commands
to identify this User Symbol. The AVT will allow up to 50 user symbols to be
registered (0 to 49).
Command Data
Format
Offset Length Type Description
0 1 Unsigned User Symbol Type
0 = User Text
1 = User Sprite
1 1 Unsigned Symbol ID Request
0 to 49 (0x00 to 0x31) For user specified request 255 (0xFF) to
allow system to specify an ID and return the ID in the response
message
Response Data
Format
0 1 Unsigned Status Code
1 3 None Reserved (padded with 0x00)
4 1 Unsigned User Symbol ID (0 to 49)
User Symbol This command registers a sprite with the ID as 0.

Registration
S ync Flag AA
Example
E xternal Comm and M sg Id 18
Length 0002
User sprite type (1 = Sprite type) 01
Symbol ID 00
Checksum 1B
External Com mand Msg AA 18000201001B
Document No: IN95-0005-01 Edition 4B User Symbology 105

C.1.2 User Symbol De-Registration Command
Summary This command de‐registers a previously registered User Symbol. Following this,
the specified User Symbol may no longer be displayed, and the specified User
Symbol ID may be re‐used for a subsequently registered User Symbol.
Command Data
Format
0 1 Unsigned User Symbol ID
Response Data
Format
User Symbol This command de‐registers a sprite with the ID as 0.

De-registration
Sync Flag AA
Example
Msg Id 19
Len 0001
Symbol ID 00
Checksum 1A
External command message AA190001001A
C.1.3 User Symbol Update Command

This command displays, or updates, a previously registered User Symbol. If the
specified User Symbol is of type User Sprite and is on display when this
command is issued, then its current appearance is removed before the supplied
update is processed.
Command lD 0x1A, 26 decimal.
Command Data
Format
1 Unsigned User Symbol type
1 0 = User Text
1 = User Sprite
2 Var User Symbol update data
User Symbol The format of the User Symbol Update Data is dictated by whether the User

Update Data Symbol Type is User Text or User Sprite, however, they both use the following
common types of data:
Colors These are 8‐bit values that specify the required colors as follows:
0 = Transparent
1 = Black
2 = White
3 = Variable as currently configured
All other values are invalid.
Screen These are 16‐bit values that identify a pixel on the screen. The top left‐hand corner
Coordinates of the screen has coordinates (0,0). The dimensions of the screen are dictated by
the Video Standard of the currently selected Sensor.
NOTE
Refer to the Video Definitions section in the relevant ADEPT AVT Hardware Manual.
User Text
2 2 Unsigned Initial X coordinate
4 2 Unsigned Initial Y coordinate
6 1 Unsigned Foreground Color
7 1 Unsigned Background Color
8 1 Unsigned Font ID
9 Variable Unsigned Character string to display
The specified initial coordinates are for the top left‐hand pixel of the first
character of the specified string, the size of each character is determined by the
specified Font.
The Font ID will be a value in the range 0 to 1 as there are two fonts available. The
first is the Adept Base Font (0) and the second is the Adept Thin Base Font (1).
These fonts are fixed and separate from the downloadable fonts used for Sprite
text.
The Character String contains the ASCII Codes of the characters to be displayed,
all printable characters are supported, i.e. ASCII Code 0x20 (space) to 0x7E (~).
If the position or Font changes from the previous update of a particular User Text
Symbol, then the existing text displayed for the Symbol is removed from the
screen before the new text is displayed.
If the position or Font does not change, then the new text is written over the top of
the previous text. In this case, if the new text has fewer characters than the
previous text, the tail characters from the previous text will remain on the screen.
If this not what is required, then the new text should be appended with space
characters to overwrite the tail characters.

External Command This command uses the text item with ID 10 to write ‘Octec Ltd. 1’ to the screen.
Example The background is black and the text will be white.
Sync Flag AA
External Command Msg Id 1A
Length 0015
Symbol Id 0A
Sprite Type 00
Intial X 0064
Initial Y 0064
Foreground Colour (2 W hite) 02
Background Colour (1 Black) 01
Font Id 00
O 4F
c 63
t 74
e 65
c 63
20
L 4C
t 74
d 64
. 2E
20
1 31
Checksum B5
External Command Msg AA1A00150A00006400640201004F63746563204C74642E2031B5
User Sprite
2 2 Unsigned Initial X coordinate
4 2 Unsigned Initial Y coordinate
6 1 Unsigned Initial Color
7 1 Unsigned Number of Sprite Elements
8 Variable Sequence of Sprite Element Definitions
The Initial Color identifies the initial setting of the “Current Color” which
controls the color of each pixel set during the processing of this Update
Command.
The coordinates identify a pixel on the screen from where the plotting of the
Sprite begins. It sets the “Current Location” before plotting the first Sprite
Element.
The Number of Sprite Elements defines the number of Sprite Element Definitions
which follow. If zero is specified, then no Sprite Elements are specified in this
command and the most recently specified sequence of Sprite Elements for this
User Symbol ID will be used again. This means that Sprite Element Definitions do
not have to be re‐specified if it is only required to move and/or change the Initial
Color of a User Symbol.
Each Sprite may be drawn as a sequence of Sprite Elements. The Sprite Elements
are drawn in the specified sequence, starting at the specified Initial Screen
Location. The drawing of each Sprite Element may result in the setting of one or
more pixels in the Current Color, and/or a change to the Current Location or
Color.
Response Data
Format
C.1.4 User Symbol Disable Command

Summary This command removes the specified User Symbol from the display. To enable the
symbol after disabling, the update symbol command needs to be resent.
Command ID 0x1B, 27 decimal.
Command Data
Format
Response Data
Format
User Symbol This command will disable symbol ID ‘0’. This will have the effect of removing it

Disable Command from the display.
Example
Sync Flag AA
Msg Id 1B
Len 0001
Symbol ID 00
Checksum 1C
External command msg AA1B0001001C

C.2 Sprite Elements
C.2.1 Sprite Element Definition
0 1 Unsigned Sprite Element Type:
0 = Move
1 = Pixel
2 = Horizontal Line
3 = Vertical Line
4 = Filled Block
5 = Color Change
6 = Change Plane
7 = Change Line Style
8 = Outlined rectangle
9 = Outline
10 (0x0A) = Filled polygon
11 (0x0B) = Outline ellipse
12 (0x0C) = Filled ellipse
13 (0x0D) = Vector
14 (0x0E) = Arc
15 (0x0F) = Download font
16 (0x10) = Text sprite
1 Variable Sprite Element Data
C.2.2 Sprite Element Data for Move

0 2 Signed Offset in X direction
2 2 Signed Offset in Y direction
The processing of a Move Sprite Element Type does not result in any pixels being
set. The X and Y coordinates of the Current Location are adjusted by having the
specified 16‐bit signed offset values added to them.
C.2.3 Sprite Element Data for Pixel
0 2 Signed Offset from current X position
2 2 Signed Offset from current Y position
The specified 8‐bit signed offset values are added to the coordinates of the
Current Location, the pixel at the resulting Location is then set to the Current
Color. This then becomes the new Current Location.
Example External This command draws a white, four dot square at position 100, 100.

Command
Sync Flag AA
Length 0014
Symbol Id 00
Sprite Type 01
Intial X 0064
Initial Y 0064
Initial Colour (2 W hite) 02
No. of Sprite Elements 04
Sprite Type = Pixel 01
X (10) 0A
Y (0) 00
X (0) 00
Y (-10) F6
X (-10) F6
Y (0) 00
X (0) 00
Y (10) 0A
Checksum 01
External Command Msg A A 1A 00140001006400640204010A 000100F601F60001000A 01

C.2.4 Sprite Element Data for Horizontal Line
0 2 Signed Line length
Positive means to right of current location
Negative means to left of current location
The number of pixels given by the absolute value of the specified Line Length are
set to the Current Color. The pixels involved will:
• Start at and include the Current Location
• All have the same Y coordinate as the Current Location
• Have incrementing or decrementing X coordinates depending on whether
the specified Line Length is positive or negative respectively
The X coordinate of the current location is adjusted by having the 16‐bit signed
line length added to it.
Example External This command draws a horizontal line from 100, 100 to 150, 100.

Command
Sync Flag AA
Length 000B
Symbol Id 00
Sprite Type 01
Intial X 0064
Initial Y 0064
Initial Colour (2 White) 02
Sprite Type = Horizontal Line 02
Length 0032
Checksum 25
External Command Msg AA1A000B000100640064020102003225
C.2.5 Sprite Element Data for Vertical Line
0 2 Signed Line length
Positive means down from current location
Negative means up from current location
The number of pixels given by the absolute value of the specified Line Length are
set to the Current Color. The pixels involved will:
• Start at and include the Current Location
• All have the same X coordinate as the Current Location
• Have incrementing or decrementing Y coordinates depending on whether
the specified Line Length is positive or negative respectively
The Y coordinate of the current location is adjusted by having the 16‐bit signed
line length added to it.
Example External This command draws a vertical line from 100, 100 to 100, 150.

Command
Sync Flag AA
Length 000B
Symbol Id 00
Sprite Type 01
Intial X 0064
Initial Y 0064
Initial Colour (2 White) 02
Sprite Type = Vertical Line 03
Length 0032
Checksum 26
External Command Msg AA1A000B000100640064020103003226

C.2.6 Sprite Element Data for Filled Block
2 Signed Block Width
0 Positive means right from current location
Negative means left from current location
2 Signed Block Height
2 Positive means down from current location
Negative means up from current location
If either the specified width or height are zero, then no pixels are set; otherwise,
all pixels whose X and Y coordinates satisfy the following conditions are set to the
current color:
If Block Width is positive:
X coordinate lies in range (current X) to (current X + (Width‐1))
If Block Width is negative:
X coordinate lies in range (current X ‐ (Width+1)) to (current X)
If Block Height is positive:
Y coordinate lies in range (current Y) to (current Y + (Height‐1))
If Block Height is negative:
Y coordinate lies in range (current Y ‐ (Height+1)) to (current Y)
The Current Location remains unchanged.
Example External This command draws a white filled, 50 by 50 rectangle at position 100, 100.

Command
Sync Flag AA
Length 000D
Symbol Id 00
Sprite Type 01
Intial X 0064
Initial Y 0064
Sprite Type = Filled rectangle 04
W idth 0032
Height 0032
Checksum 5B
External Command Msg AA1A000D000100640064020104003200325B
C.2.7 Sprite Element Data for Color Change
0 1 Unsigned New color
The processing of a Color Change Sprite Element Type does not result in any
pixels being set. The Current Color is set to that specified so that subsequent
pixels set when drawing this Symbol will be in this color.
Example External This command draws 2 dots in white, 2 dots in black next to each other at position

Command 100, 100.
Sync Flag AA
Length 0016
Symbol Id 00
Sprite Type 01
Intial X 0064
Initial Y 0064
X (10) 0A
Y (0) 00
X (0) 00
Y (-10) F6
Sprite Type = Change colour 05
Colour (1 = Black) 01
X (-10) F6
Y (0) 00
X (0) 00
Y (-10) 0A
Checksum 0A
External Command Msg AA1A00160001006400640205010A000100F6050101F60001000A0A

C.2.8 Sprite Element Data for Change Plane
0 2 Unsigned Plane number (1 or 2)
To create shapes on top of other shapes, planes are required.
There are two planes for use with user sprites. These are labeled plane 1 and plane
2. Plane 1 sprites will be drawn on top of plane 2 sprites.
External Command This draws an outline box on the graphics plane 2.

Example S y nc F lag AA
M s g Id 1A
Length 0010
S y m bol Id 00
S prite Ty pe 01
Intial X 0064
Initial Y 0064
Initial Colour 02
No. of S prite E lem ents 02
S prite Ty pe = C hange plane 06
P lane No. 0002
S prite Ty pe = O utlined B ox 08
W idth 00C8
Height 0064
C hec k s um 33
E x ternal C om m and A A 1A 001000010064006402020600020800C 8006433
C.2.9 Sprite Element Data for Change Line Style

0 2 Unsigned Line Style
Changing line style will affect the following commands:
Outlined Rectangle (Sprite element type = 8)
Outlined Polygon (Sprite element type = 9)
Outlined Ellipse (Sprite element type = 11/0x1A)
Vector (Sprite element type = 13/0x0D)
Arc (Sprite element type = 14/0x0E)
The line style definitions are as follows:
0 = Continuous Line
1 = Dashed 1‐1 (1 pixel on, 1 pixel off)
2 = Dashed 5‐1‐1‐1 (5 pixels on, 1 off, 1 on, 1 off)
3 = Dashed 5‐3 (5 pixels on, 3 off)
4 = Dashed 12‐4 (12 pixels, 4 off)
5 = Dashed 9‐alt (9 pixels on, 7 alternating)
6 = Dashed 10‐2‐2‐2 (10 pixels on, 2 off, 2 on, 2 off)
7 = Dashed 8‐8 (8 pixels on, 8 pixels off)
Code Pattern
-----------------------------------------------
MSB LSB
0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0
2 1 1 1 1 1 0 1 0 1 1 1 1 1 0 1 0
3 1 1 1 1 1 0 0 0 1 1 1 1 1 0 0 0
4 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0
5 1 1 1 1 1 1 1 1 1 0 1 0 1 0 1 0
6 1 1 1 1 1 1 1 1 1 1 0 0 1 1 0 0
7 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0
1 = foreground colour, 0 = background colour
External Command This command draws an outlined rectangle starting at 100, 100 with width 200

Example and height 100. The line style for the rectangle is Dashed 5‐3.
Sync Flag AA
Length 0010
Symbol Id 00
Sprite Type 01
Intial X 0064
Initial Y 0064
Initial Colour = White 02
Sprite Type = Change line style 07
Line Style 0003
Sprite Type = Outlined Box 08
Width 00C8
Height 0064
Checksum 35
External Command Msg AA1A001000010064006402020700030800C8006435

C.2.10 Sprite Element Data for Outlined Rectangle
0 2 Signed Block Width
Positive means right from current location
Negative means left from current location
2 2 Signed Block Height
Positive means down from current position
Negative means up from current position
This will draw an outlined rectangle from the current cursor position to the co‐
ordinates given.
The outline will use the color defined by initial color or any change color
command.
The outline will use the line style defined by change line style. Line style is
initially continuous (value 0).
External Command This command draws an outlined rectangle starting at 100, 100 with width 200

Example and height 100.
Sync Flag AA
Length 000D
Symbol Id 00
Sprite Type 01
Intial X 0064
Initial Y 0064
Initial Colour 02
Sprite Type = Outlined Box 08
Width 00C8
Height 0064
Checksum 27
External Command Msg AA1A000D00010064006402010800C8006427
C.2.11 Sprite Element Data for Outlined Polygon
0 2 Unsigned Number of vertices N (maximum is 8)
For Number of Vertices (1 to N)
4N-2 2 Unsigned Vertex(X)
4N 2 Unsigned Vertex(Y)
command.
External Command This command draws an outlined polygon with 3 vertices.

Example Sync Flag AA
Length 0017
Symbol Id 00
Sprite Type 01
Intial X 0064
Initial Y 0064
Initial Colour 01
Sprite Type = Outlined Polygon 09
Vertices Count 0003
X1 0032
Y1 0032
X2 003C
Y2 0050
X3 005A
Y3 0046
Checksum 98
External Command Msg AA1A0017000100640064010109000300320032003C0050005A004698

C.2.12 Sprite Element Data for Filled Polygon
0 2 Unsigned Number of vertices N (maximum is 8)
For Number of Vertices (1 to N)
4N-2 2 Unsigned Vertex(X)
4N 2 Unsigned Vertex(Y)
The polygon will be drawn with the color defined by initial color or any change
color command.
The maximum height for filled polygons is 150 pixels.
The drawing method for filled polygons only uses a scan line algorithm. For each
scan line of the polygon, there cannot be more than 4 edges to the polygon..
External Command This command draws a filled polygon with 4 vertices.

Length 001B
Symbol Id 00
Sprite Type 01
Intial X 0064
Initial Y 0064
Initial Colour 01
Sprite Type = Polygon 0A
Vertices Count 0004
X1 0032
Y1 0032
X2 0037
Y2 0037
X3 0041
Y3 0037
X4 0046
Y4 0032
Checksum D0
External Command Msg AA1A001B00010064006401010A000400320032003700370041003700460032D0
C.2.13 Sprite Element Data for Outlined Ellipse
0 2 Unsigned Radius in X
2 2 Unsigned Radius in Y
command.
External Command This command draws an outlined ellipse with origin 256, 128, X‐radius 100 and

Example Y‐radius 50.
Sync Flag AA
Length 000D
Symbol Id 00
Sprite Type 01
Intial X 0100
Initial Y 0080
Initial Colour 01
Sprite Type = Outlined Ellipse 0B
Radius X 0064
Radius Y 0032
Checksum 4C
External Command Msg AA1A000D00010100008001010B006400324C

C.2.14 Sprite Element Data for Filled Ellipse
0 2 Unsigned Radius in X
2 2 Unsigned Radius in Y
The ellipse will be drawn with the color defined by initial color or any change
color command.
External Command This command draws a filled ellipse with origin at 256, 128, X‐radius 200 and Y‐

Example radius 10.
Sync Flag AA
Length 000D
Symbol Id 00
Sprite Type 01
Intial X 0100
Initial Y 0080
Initial Colour 01
Sprite Type = Filled Ellipse 0C
Radius X 00C8
Radius Y 000A
Checksum 89
External Command Msg AA1A000D00010100008001010C00C8000A89
C.2.15 Sprite Element Data for Vector

0 2 Unsigned Start point X
2 2 Unsigned Start point Y
4 2 Unsigned End point X
6 2 Unsigned End point Y
This will draw a line from the start point to the end point.
The line will use the line style defined by change line style. Line style is initially
continuous (value 0).
External Command This command draws a white line from 100, 100 to 200, 200.

Length 0011
Symbol Id 00
Sprite Type 01
Intial X 0064
Initial Y 0064
Initial Colour 02
Sprite Type = Vector 0D
X1 0064
Y1 0064
X2 00C8
Y2 00C8
Checksum 5C
External Command Msg AA1A001100010064006402010D0064006400C800C85C
C.2.16 Sprite Element Data for Arc
The function for drawing arcs will take a radius in X and Y to give the ellipse.
The ellipse is divided into 8 octants as shown below. The function also takes an
octant value and will only draw this octant.
The origin of the ellipse is at the current pointer.

0 2 Unsigned Radius X
2 2 Unsigned Radius Y
4 2 Unsigned Octant ID (1 to 8)
External Command This command draws an arc with origin at 256, 128, X‐radius 100, Y‐radius 100

Example and drawing the North‐>North East octant (octant 1).
Sync Flag AA
Length 0012
Symbol Id 00
Sprite Type 01
Intial X 0100
Initial Y 0080
Initial Colour 01
Sprite Type = Change Style 07
Radius X 0007
Sprite Type = Arc 0E
Radius X 0064
Radius Y 0064
Octant 0001
Checksum 96
External Command Msg AA1A001200010100008001020700070E00640064000196

C.2.17 Sprite Element Data for Download Font
There are four font sets available for the user. These range from 0 to 3. In each font
set, there are 255 characters available. The mapping character is a value from 0 to
255 inclusive.
You can create font characters with variable width and height. Maximum width of
a character is 24 pixels, and maximum height of a character is 18 pixels.
Characters are created using bits. A ‘0’ or off signifies the background color and a
‘1’ or on signifies the foreground color.
The ‘Data Byte Count’ must contain enough data bytes to fill the character.
If there is an overflow of data for a particular line, then that data is ignored.
0 1 Unsigned Font set number (0 to 3)
1 1 Unsigned Mapping character (0 to 255)
2 1 Unsigned Width of character (1 to 24)
3 1 Unsigned Height of character (1 to 18)
4 2 Unsigned Data byte count – count of bytes of data that make the symbol
6 Data byte count Unsigned Data
External Command
Example 1
Font Set Number 0
Mapping Character =0
Width Of Character =9
Height Of Character =9
Data Byte Count = 18
Data 00 00 3A 00 7B 00 03 00 63 00 60 00 6F 00 2E 00 00 00
Sync Flag AA
Length 0021
Symbol Id 00
Sprite Type 01
Intial X 0064
Initial Y 0064
Initial Colour 01
Sprite Type = Download font character 0F
Font set (0..3) 00
Mapping character (0..255) 00
W idth of character 09
Height of character 09
Data byte count 0012
Byte 1 00
Byte 2 00
Byte 3 3A
Byte 4 00
Byte 5 7B
Byte 6 00
Byte 7 03
Byte 8 00
Byte 9 63
Byte 10 00
Byte 11 60
Byte 12 00
Byte 13 6F
Byte 14 00
Byte 15 2E
Byte 16 00
Byte 17 00
Byte 18 00
Checksum 51
External Command Msg AA1A002100010064006401010F00000909001200003A007B000300630060006F002E00000051
NOTE
A symbol ID must be registered before any symbols can be downloaded.
External Command
Example 2
Font Set Number =0
Mapping Character =1
Width Of Character = 15
Height Of Character = 17
Data Byte Count = 34
Data 00 00 03 C0 0F F0 1F FC 3F F8 3F F0 7F E0 7F C0 7F
80 7F C0 7F E0 3F F0 3F F8 1F FC 0F F0 03 C0 00 00

Sync Flag AA
External Com m and Ms g Id 1A
Length 0031
Sym bol Id 00
Sprite Type 01
Intial X 0064
Initial Y 0064
Initial Colour 01
No. of Sprite Elem ents 01
Sprite Type = Download font character 0F
Font s et (0..3) 00
Mapping character (0..255) 00
Width of character 0F
Height of character 11
Data byte count 0022
Byte 1 00 Byte 18 80
Byte 2 00 Byte 19 7F
Byte 3 03 Byte 20 C0
Byte 4 C0 Byte 21 7F
Byte 5 0F Byte 22 E0
Byte 6 F0 Byte 23 3F
Byte 7 1F Byte 24 F0
Byte 8 FC Byte 25 3F
Byte 11 3F Byte 28 FC
Byte 14 E0 Byte 31 03
Byte 15 7F Byte 32 C0
Byte 16 C0 Byte 33 00
Byte 17 7F Byte 34 00
Checks um 28
External Com m and Ms g AA1A0021000100640064010131000009090012000003C00FF01FFC \
3FF83FF07FE07FC07F807FC07FE03FF03FF81FFC0FF003C000001C
C.2.18 Sprite Element Data for Text Sprite
0 1 Unsigned Foreground color
1 1 Unsigned Background color
2 2 Unsigned String length (N)
For String Length (n=1 to N)
4+2 (n-1) 1 Unsigned Font set number n (0-3)
5+2 (n-1) 1 Unsigned Mapping character n (0-255)
• New current location:
The X coordinate of the Current location is incremented by the sum of the
font width of each character in the text string.
External Command This command draws three symbols on the screen at position x=100, y=100. The
Example symbols in this example are Font 0‐Char 1, Font 0‐Char 0 and Font 0‐Char 1.
NOTE
The characters must have been downloaded before writing the text string.
Sync Flag AA
Length 0013
Symbol Id 01
Sprite Type 01
Intial X 0064
Initial Y 0064
Initial Colour 01
Sprite Type = Text 10
Foreground colour (2=W hite) 02
Background colour (1=Black) 01
Length 0003
Font number 00
Mapping character 01
Font number 00
Font number 00
Checksum 11
External Command Msg AA1A00130101006400640101100201000300010000000111

C.3 Troubleshooting
C.3.1 Symbology Clashes
Within the tracker there are four graphic planes, these are shown in descending
order of display priority:
1. Tracker text.
2. Tracker sprites.
3. User symbology plane 1 (user text, user sprites).
4. User symbology plane 2 (user sprites).
The user can only draw on user planes 1 and 2.
If symbols drawn on different planes overlap one another, then the overlapping
pixels will be displayed according to the pixel in the highest priority graphics
plane.
If symbology is drawn on the same plane and the pixels overlap one another, then
the symbology data will clash. The symbology of each sprite with have an XOR
operation applied over the four colors that are used.
For example, with black sprite 1 drawn on the same plane as black sprite 2, the
clashing pixels of black (color value 1) on black will produce a transparent color
(color value 0). Conversely, white (color value 2) clashing with black (color value
1) will produce variable intensity (color value 3).
Table C-1 Color Clash Table
Transparent (0) Black (1) White (2) Variable Intensity (3)
Transparent (0) Transparent (0) Black (1) White (2) Variable Intensity (3)
Black (1) Transparent (0) Variable White (2)
Black (1)
Intensity (3)
White (2) Variable Transparent (0) Black (1)
White (2)
Intensity (3)
Variable Intensity (3) Variable Intensity (3) White (2) Black (1) Transparent (0)
To avoid the clashing of sprites, use the Change Plane command.
D • Command Reference Manuals
Refer to the following publications:
• ADEPT104 Family Command Reference (IN82‐0302‐050‐01)
• ADEPT60 & 64 Command Reference (IN82‐0301‐050‐01)
• ADEPT74 Command Reference (IN82‐0303‐050‐01)
The format of the data located in each field of the COMMAND and STATUS
blocks is specified using the following notations:
ENUM The data is a 32‐bit unsigned integer consisting of a list of defined
values
BOOLEAN The data is a 32‐bit unsigned integer consisting of 2 defined
values:
0 = FALSE
1 = TRUE
INTEGER The command is a 32 bit 2ʹs complement number integer
FIXED (32,x) The command is a 32 bit 2ʹs complement number where the ‘x’
least significant bits of the 32 bit number are fractional
The following notations are used when specifying the contents of each word:
UNITS Specifies the value of the LSB of INTEGER data and the least
significant integer bit of FIXED data
RANGE Specifies the range of the value
INCREMENT Specifies the resolution of the command
All undefined fields are reserved and must not be used.
Document No: IN95-0005-01 Edition 4B Command Reference Manuals 129

E • Technical Notes
E.1 Built In Tests

This appendix describes the built‐in self test functions of the AVT. Information
has been consolidated from various internal specification documents as well as
the application source code.
Specification The AVT operates in two distinct modes: non‐operational and operational (see
System Modes). The operational mode of the AVT is either in the target
Acquisition state or one of the target Tracking states. Built in Test (BIT) operation
in each of these modes/states is specified as follows.
E.1.1 Non-Operational Mode

BIT State 1. BIT is performed following initialization and on demand.
2. The BIT result is reported and, if successful, the system moves to
Acquisition; if unsuccessful, the system moves to the BIT Failed state.
3. During BIT, the AVT continues to service the command and status interfaces,
reporting the fact it is performing BIT.
4. The AVT returns to the configuration that it was in when BIT was demanded.
NOTE
The AVT always returns to Acquisition mode.
5. The AVT provides background BIT.
BIT Failed State 1. The system reports the BIT fail and does nothing further.
E.1.2 Operational Mode

Acquisition State 1. The AVT accepts commands to run BIT and also change the Saved
Configuration.
2. If Background BIT detects a failure, the AVT terminates the Acquisition State
and enters the BIT Failed State.
3. If Requested BIT is commanded, the AVT terminates the Acquisition State
and enters the BIT Test State.
Track States 4. The AVT terminates the Track State and enters BIT Failed following
Background BIT fail.
E.1.3 Reporting
BIT progress and result information are reported via three different mechanisms:
symbology on video output, serial message and, where available, reading status
over the bus interface.
Video Output The AVT reports the overall BIT pass or fail result as a text message on the video

output symbology overlay.
Serial Message The real time output ‘System Mode Operational’ item returns ‘false’ while BIT is

in progress. If BIT subsequently fails, the mode continues to be reported as non‐
operational.
Bus The AVT reports BIT failure information via the DPR status block when read over
Communication the bus interface.
System Modes .
INITIALISE
BUILT CONFIGURATION
BIT FAIL IN MANAGEMENT
TEST
NON-
OPERATIONAL
MODE
BACKGROUND
BIT FAIL
TARGET
ACQUISITION
AUTOTRACK
OFF
AUTOTRACK
ON
TIME > STAGE 2 TIME
NO
ACQUIRE TARGET
TARGET PRESENT
PRESENT
OPERATIONAL
MODE TARGET
RE-ACQUIRED BREAKLOCK
STAGE 2
LOCK
TARGET
RE-ACQUIRED
BREAKLOCK
TIME > STAGE 1 TIME
BREAKLOCK
STAGE 1
AUTO
TRACK
Document No: IN95-0005-01 Edition 4B Technical Notes 131

E.1.4 Implementation
The Tests The self‐test functions that are performed are the same, whether initiated through
the Startup Bit or through the Requested BIT mechanism.
The suite of tests consists of memory checks on two distinct areas of read/write
memory:
1. LUT memory.
2. Graphics memory.
The tests are executed in the order given. If a test fails, then the BIT fail state is
entered immediately (i.e. any remaining memory tests are not run).
For each of the three memory tests, the memory is not read from or written to
directly by the test. Instead, requests are sent to the appropriate hardware driver
to perform the read or write operation. There are separate hardware drivers for
LUT and graphics operations.
NOTE
On earlier AVT hardware, separate physical hardware devices were used for each of the memory areas.
Newer designs use a unified memory architecture. Consequently, on newer hardware the self test
function comprises a single memory test.
LUT Memory Test The LUT memory read/write test operates on each LUT in turn.
The current LUT data is read and saved into a temporary buffer. Pseudo random
data is then written into the LUT. The LUT contents are immediately read back
and then compared with what had been written. If there is any data mismatch,
then a BIT fail is reported and the test stops.
Graphics Memory The graphics memory is partitioned into 1024 equal sized blocks. The read/write

Test test operates on each of these blocks in turn.
The current graphics block data is read and saved into a temporary buffer. Pseudo
random data is then written into the graphics block. The graphics block contents
are immediately read back and then compared with what had been written. If
there is any data mismatch, then a BIT fail is reported and the test stops.
Startup BIT (PBIT) The startup BIT function executes as part of the Tracker initialization process (i.e.

while in non‐operational mode following a power‐on event or system reset).
Running the self tests may be disabled through a configuration item.
Requested BIT Task In operational mode, this task responds to user‐initiated command messages to

(IBIT) run the full BIT suite. While the tests are running, all other tracker processing is
disabled.
Background BIT The background BIT task runs continuously while the Tracker is in the

Task (CBIT) operational mode. It is the lowest priority of all the tasks in the system (i.e. it only
gets to run when all other tasks have nothing to do). This task continually checks
the area of memory containing User Mode (i.e. task) code and read‐only data for
corruption. This Task also calculates and displays (if so configured) the CPU
Loading, and pulses the Background Activity LED.
The read‐only memory to be checked is divided into 256 blocks. The sizes of all of
all except the last block are equal. The size of the last block may be smaller than
that of the other blocks. A 32‐bit check code for each block is generated by reading
each 32‐bit word in turn and exclusive‐ORing it with a left‐rotated version of the
previous check code value. This operation is described in the following code
fragment.
U32 CalculateCheckCode( U32 *pData, U32 size )
{
U32 checkCode = 0;
for ( U32 i = 0; i < size; ++i, ++pData )

{
checkCode = RotateLeft( checkCode ) ^ *pData;
}
return checkCode;}
The check codes for all 256 blocks are calculated once on initialization and the
values are stored. The background BIT task then continuously re‐calculates the
check code for each block in turn and compares it with the previously stored
value. If a mismatch is detected then Tracker operation is stopped and the non‐
operational BIT fail state is entered.
E.1.5 Reporting Failure

BIT checks for different types and location of memory error. This presents a
problem when it comes to reporting such errors if that reporting itself has to make
use of memory. Since code execution of any kind will make use of ROM and most
code execution, including that for error reporting, will make use of RAM too it is
simply not possible to reliably report BIT fail results using any mechanism that
requires code execution on the part of the AVT CPU. BIT pass results may, of
course, be reported correctly.
BIT result information read over a bus interface (with that read initiated and
controlled) from a separate host processor has no such issues, since this
information may be read independently of correctly functioning AVT code
execution.
A host system connected to the AVT has two choices to detect AVT failure:
1. Read the BIT result information in the DPR status block (requires bus
connection to AVT).
2. Read the System Mode Operational status item (block 129, field 1). If this
continues to report “false” after a requested BIT or a request to read this
status item simply times out (which could happen on catastrophic failures)
then the AVT has failed (either due to the BIT itself or some other failure has
prevented correct reporting).
If the only connection between AVT and host system is a serial connection, then it
is strongly recommended that the host operates a time‐out mechanism on
communication with the AVT. This is needed to handle catastrophic AVT failures
as described above. It is also a sensible strategy to handle serial communication
failures (e.g. cable disconnected).
Document No: IN95-0005-01 Edition 4B Technical Notes 133

Glossary
AVT Automatic Video Tracker.
AVT Control A menu driven programme to set up parameters in the tracker and interface to a joystick
Software and Remote Control Unit.
BIT Built In Test.
CBIT Continuous BIT. Continuously running self‐check tests. Referred to in AVT
documentation as Background BIT.
CCIR Comité Consultatif International Des Radio Communications.
Command Interface Commands are input to the AVT via the Serial Interface, the Bus Interface and the Remote
Controller Interface. All these interfaces run in parallel.
Configuration Block A section of ROM containing the power‐on default constants.
CPU Central Processing Unit.
DPR Dual Port RAM. Mechanism by which AVT status may be interrogated over a bus
interface.
EOTS Electro‐Optical Tracking System.
FOV Field of View.
HFOV Horizontal Field Of View.
HTML HyperText Markup Language.
IBIT Interruptive BIT or Initiated BIT. User‐initiated tests. Referred to in AVT documentation as
Requested BIT.
ISA Industry Standard Architecture.
LOS Line Of Sight.
LUT Look Up Table.
MTD Moving Target Detection.
MTT Multiple Target Track.
NTSC National Television System Committee.
PAL Phase Alternation Line.
PBIT Power‐On BIT. Tests performed at power‐on or system reset. Referred to in AVT
documentation as Startup BIT.
PCI Peripheral Component Interconnect.
PFK Programmable Function Key.
PID Proportional–Integral–Derivative controller.
PiP Picture in Picture.
RAM Random Access Memory.
RCU Remote Control Unit.
RMS Root Mean Square.
ROM Read Only Memory.
Status Interface The AVT outputs Track Error values and status information via the available interfaces.
TBE Target to Boresight Error.
VME VERSAmodule Eurocard bus.
Document No: IN95-0005-01 Edition 4B Glossary 135

© 2017 Abaco Systems, Inc. Abaco Systems Additional Resources
All rights reserved.
Information Centers
* indicates a trademark of Abaco Systems, For more information, please visit the
Inc. and/or its affiliates. All other Americas: Abaco Systems web site at:
trademarks are the property of their
1-866-652-2226 (866-OK-ABACO)
respective owners.
www.abaco.com
Confidential Information - This document Europe, Middle East and Africa:
contains Confidential/Proprietary +44 (0)1327 359444
Information of Abaco Systems, Inc. and/or
its suppliers or vendors. Distribution or
reproduction prohibited without permission.
THIS DOCUMENT AND ITS CONTENTS ARE
PROVIDED "AS IS", WITH NO
REPRESENTATIONS OR WARRANTIES OF
ANY KIND, WHETHER EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO
WARRANTIES OF DESIGN,
MERCHANTABILITY, OR FITNESS FOR A
PARTICULAR PURPOSE. ALL OTHER
LIABILITY ARISING FROM RELIANCE ON
ANY INFORMATION CONTAINED HEREIN IS
EXPRESSLY DISCLAIMED.
Document No: IN95-0005-01 Edition 4B

Reference Manual: ADEPT Automatic Video Trackers Edition 4B

Uploaded by

Copyright:

Available Formats

Reference Manual: ADEPT Automatic Video Trackers Edition 4B

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Reference Manual: ADEPT Automatic Video Trackers Edition 4B

Uploaded by

Copyright:

Available Formats

What does the document discuss?

What does the document discuss?

What are the different states of the tracker?

What are the different states of the tracker?

Reference Manual

ADEPT Automatic Video Trackers

Edition Date Description

1 • About This Manual . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

Document No: IN95-0005-01 Edition 4B Table of Contents 3

1.2 Further Information

1.3 Technical Support

Document No: IN95-0005-01 Edition 4B About This Manual 5

2.1 System Overview

2.1.1 Electro-Optical Tracking System

ANNOTATED VIDEO RATE DEMAND

Document No: IN95-0005-01 Edition 4B Introduction 7

Document No: IN95-0005-01 Edition 4B

Digital Video Output 8 8 DVI 8 8 8 8 8 8

3.1 Non-Operational State

Initialize Mode AVT initialization is triggered after powering‐on the AVT. During initialization,

Built-in Test (BIT) The BIT mode is entered following initialization or by a BIT command message in

3.2 Operational State

Acquisition Window The Acquisition Window is used to define the area of operation within the video

Armed Mode It is possible to configure the AVT so that if a target is not present when the Track

Reversionary Track The AVT can also be configured to attempt a reversionary track mode if an

3.4 Track Mode (Autotrack)

Document No: IN95-0005-01 Edition 4B Tracker Theory Of Operation 11

Initial Track When automatic track window size is selected, the track window size is initialized

3.4.1 Target Tracking Modes

Lock Mode In Lock mode, the AVT automatically tracks the target contained within the track

Breaklock Stage 1 Following track loss detected by the breaklock algorithm, the AVT initially enters

Breaklock Stage 2 Following a failure to re‐acquire the target in Breaklock Stage 1, the AVT enters

Forced Breaklock/ An operator can use the Force Coast command to force the tracker into breaklock.

Polarity Pre- Since the algorithm calculates the centre of energy of the target intensity, it ideally

Target Size The Centroid algorithm is able to estimate target size. When tracking, this

In Acquisition Mode When the acquisition algorithm is set to Centroid, the target position and size are

In Track Mode The Track Window is positioned according to the centre of energy of the target

Document No: IN95-0005-01 Edition 4B Algorithms 13

Target Left Edge

Target Size The Edge algorithm is not able to estimate target size in the axis perpendicular to

In Track Mode The Track Window is positioned according to the selected edge of the target. The

Document No: IN95-0005-01 Edition 4B Algorithms 15

Polarity Pre- Same as Centroid.

Target Size The MTT algorithm can estimate target size.

In Acquisition Mode When the MTT is selected as the Acquisition algorithm, it can be used to

In Track Mode On entry to track mode, the MTT focuses on the primary acquisition target. The

4.4 Phase Correlation

Polarity Pre- The Phase Correlation algorithm does not use an image pre‐processor, so polarity

Target Size The Phase Correlation algorithm does not produce an estimate of target size, so

In Acquisition Mode The Phase Correlation acquisition algorithm simply estimates the optimal

Document No: IN95-0005-01 Edition 4B Algorithms 17

Polarity Pre- Since the Centroid type algorithm calculates the centre of energy of the target

Target Size The Combined algorithm obtains target size information from the Centroid type

In Acquisition Mode Combined is not available as an Acquisition algorithm.

In Track Mode The Combined track algorithm behaves in a similar fashion to the Phase

Document No: IN95-0005-01 Edition 4B Algorithms 19

Polarity Pre- The Scenelock algorithm does not use an image pre‐processor, so polarity is

Area of Interest The Scenelock algorithm operates over an area of interest within the image that

In Acquisition Mode The Scenelock algorithm automatically identifies and assigns tracking points. The