Scribd
Upload
11 views14 pages

Basics of Robotics l27

Hi

Uploaded by

25kusum08
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
11 views14 pages

Basics of Robotics l27

Hi

Uploaded by

25kusum08
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 14

Basics of Robotics

4.1: Control Considerations


4.1.1: Control modules
4.1.2: Modules for advance robots
CONTROL CONSIDERATIONS
• The control system of a robot is to supervise the activities of a robot.
It should have the following characteristics:
• Manipulation ability: Ability to move objects within the workspace
• Sensory ability: Able to obtain information on the state of the system
and working environment
• Intelligence: Ability to modify the system behavior based on the
sensor information
• Data-processing ability: Capability of storing, elaborating, and
providing data on system activity
CONTROL CONSIDERATIONS
• The above objectives can be achieved through an effective
implementation of a functional architecture, which is nothing but the
supervision of several activities arranged in a hierarchical structure.
• The term hierarchical is equivalent to the command structure of the
military, where the generals pass orders to the officers who pass
orders to the sergeants, and in turn orders are passed to the troops.
• In such hierarchy, control is supposed to flow from top to bottom, and
reporting is intended to flow up from the bottom.
CONTROL CONSIDERATIONS
• The lower levels are meant for physical motion execution, whereas
the higher levels are responsible for logical action planning.
• The levels are interconnected with the flow of measurement and/or
results of action data from bottom to top, whereas the directions are
transmitted from top to bottom.
Function Modules
• In order to achieve the above control architecture, three functional
modules at each level are assigned. They are as follows:
• The first sensory module is devoted to sensory data management.
These modules acquire, elaborate, correlate, and integrate sensory
data in time and space so that the system data and environment
characteristics are recognized.
• The second modeling module is to provide knowledge of the relevant
world. These modules contain models based on prior knowledge of
the system and environment, which are updated with the information
coming from the sensory modules.
Function Modules
• The third decision module is to decide the action policy, i.e., to
decompose high level tasks into low-level actions.
• Such task decomposition is in terms of both the time of sequential
actions and the space of concurrent actions.
• Each decision module is responsible to the functions concerning
management of elementary action assignments, task planning, and
execution.
• A decision module determines the level of hierarchy and the
associated functions of the modeling and sensory modules operating
at that level.
Modules for Advanced Robots
• An advanced robot system can be partitioned into three distinct
modules, as shown in fig. (a). These modules are as follows:
• The program module which accepts inputs from the user interactively
or in the form of a robot program. The inputs are translated into a set
of lower-level commands, perhaps including conditional branches,
much as a compiler might convert from FORTRAN to machine
language.
• The location module then serves to relate symbolic names of
locations, for example, clamp, to Cartesian descriptions of points in
the workspace.
Modules for Advanced Robots
• The control system module is finally responsible to control the motion
of the robot so that it can achieve the tasks described by it at the
locations described by the location module.
• This module has a real-time requirement which needs to be further
decomposed. Fig (b) shows the decomposition of the real time
control functions.
• The exact number of levels in this decomposition, as well as the exact
responsibilities of each level are those that are the concern of the
system designer.
Modules for Advanced Robots
• Note that at higher levels, the control functions are independent of
the arm geometry.
• That is, the determination of the path is performed in Cartesian
coordinates relative to constraints, for it may be necessary to specify
locations in a moving frame such as on a conveyor (which means that
the location module must be capable of real-time updates).
• Since the task description does not require the knowledge of the
robot geometry at this stage, the same algorithms and software may
be used for many different robots. At the lower levels of the control
hierarchy, however, the details of the algorithms become closely
related to the arm geometry.
Modules for Advanced Robots
• Once the kinematic transformations have been performed, the
control algorithms may vary from one robot to another.
• This is observed at the two lower levels of Fig. (b). The lowest level
compares set points (provided by the next level) with joint positions
(provided by sensors) and moves the joint to reduce the error.
• The previous level computes the kinematic transformations and
provides those set points to the bottom level. One may observe from
Fig. (b) that there are really two hierarchies involved.
Modules for Advanced Robots
• A top-down hierarchy of control, and a bottom-up hierarchy of status.
Fig. (c) shows this clearly by giving the details of a single level in the
structure. Since each level in the hierarchy is well defined functionally,
the robot control task can be divided into processing modules in
which each module has well-defined inputs and outputs and,
therefore, can be associated with an independent processing
element.

You might also like