Development of Simulation Software under Latest RT-

Linux Environment for Hardware-In-Loop Simulation

Test bed

Rajesh Shankar Karvande 1[0000-1111-2222-3333], Srinivas Enjapuri1 and Tatineni Madhavi2

1
Research Center Imarat, Hyderabad, Telangana-500069, INDIA
2
GITAM University, Hyderabad, Telangana-500265, INDAI

[email protected]
[email protected]
[email protected]

Abstract. Validation of Avionics Software is Critical and mandatory phase in field of

aerospace and defense. The Avionics software tested in real time environment
popularly known as Hardware-in-loop simulation. In this, simulation computer that is
used for plant software has been developed using real time kernel. Real time RT-Linux
is used for scheduling algorithms and interface with data acquisition system. Widely
Linux based real time is used for cost effectiveness and good technical support by the
developer. Traditionally in HILS environment, RT-Linux has been used with Intel
Pentium-4 (P#4) architecture that was based on decade old technology. Now as the
resources are obsolete and RT-kernel is technically outdated, it is mandatory to upgrade
HILS hardware architecture with latest RT-kernel for enhancement of performance and
speed of computation. This also offers more kernel's services accessibility. In this
paper, the process of development of software based on on new real time kernel patch,
Xenomai is elaborated. The Intel i7 system is developed on this RT kernel with all
drivers development of I/O interfaces has been discussed. Methodology, testing, results
and conclusion is explained.

Keywords: Avionics Subsystem, Hardware In Loop Simulation, Simulation Computer,

Real Time Operating System, RT-Linux, 6Dof Simulation software.
2 F. Author and S. Author

1 Introduction

In the field of aerospace and defense, the validation of avionics software stands as a
critical and mandatory phase. Avionics software, pivotal for ensuring the safe and
efficient operation of aircraft and defense systems, undergoes rigorous testing to meet
stringent industry standards and regulatory requirements. The main architecture of
this testing process is the real-time environments, commonly known as Hardware-in-
Loop Simulation (HILS), where avionics software is evaluated in different conditions
mirroring the real-world flight scenarios[1].

Fig. 1. HILS process block diagram..

Almost more than a decade, the of HILS Test-setup are running on legacy hardware
architectures, often based on GPU-Intel P-IV technology. However, with the rapid
evolution of technology and the obsolescence of resources, there is a need to upgrade
these architectures with the latest real-time kernels. The integration of cutting-edge
hardware and software solutions not only enhances the performance and
computational speed of HILS but also expands the accessibility of many kernel
services like watchdog timer, dual Ethernet ports etc. thereby augmenting the overall
effectiveness of avionics software validation[3]. In this paper, we delve into the
process of developing simulation software within the latest real-time kernel
environment, specifically focusing on the Xenomai real-time kernel patch. By
transitioning from outdated hardware architectures to modern Intel i7 systems
equipped with the Xenomai kernel, the aim is to elucidate the methodologies, testing
procedures, and outcomes of this crucial upgrade. Through this exploration, we shed
light on the significance of embracing advanced technologies in advancing the field of
avionics software validation[2].
 Contribution Title (shortened if too long) 3

1.1 Overview of Hardware-in-Loop Simulation (HILS) and Its Role in

Avionics Software Testing: HILS involves the integration of physical hardware
components, such as sensors, actuators, and control systems, with computer
simulations to emulate real-world operating conditions within a controlled
laboratory environment. In the context of avionics software testing, HILS serves
as a bridge between software development and real-world deployment, enabling
engineers to validate software functionality, system interactions, and performance
under a variety of simulated scenarios. By interfacing avionics software with
physical hardware components, HILS facilitates comprehensive testing of system
responses to diverse inputs, environmental conditions, and potential failure modes.
Six Degrees of freedom is three Rotational axis and three translation axis in three
dimensional space[4]. Also the body is change the position as forward/backward,
up/down, left/right freely. In our case the 6DOF is using for Aerospace and
defence purpose only. For these using mathematical models we are calculating
Body rates (p, q, r) and Accelerations (ax, ay, az).

1.2 Need for Real Time Simulation: Simulation environment is creation of the
virtual scenario of the launch with execution of the real time models. Although the
unit level testing is done with respective subsystems but the integration and data
exchange with each other is essential and should happen exactly the same as per
the real flight. So the simulation computer is configured with the hard real time
execution of algorithm.

Problem Definition:
4 F. Author and S. Author

2 Problem Definition: There is requirement to update the old Intel P#4

based architecture to the latest technology for high computation complexities.

A. Performance Optimization: The advancement of avionics software complexity

and sophistication necessitates hardware architectures capable of delivering
optimal performance and computational efficiency.

B. Hardware and software obsolesce: Hardware is decade old and ISA bus
communication. In this era of optimization, all hardware are designed on PCI or SB
bus. Also, old RT-Linux is offering limited services and also work on some of the
hardware. It is essential and mandatory to migrate to latest technology based
architecture[5].

By leveraging the latest real-time kernel technologies, HILS environments can

achieve higher levels of determinism, precision, and accuracy in simulating complex
avionics software behaviors. This, in turn, enhances the fidelity and reliability of
HILS testing, enabling engineers to identify and address potential issues before
deployment.

3 Development Methodology:

3.1 Hardware Methodology:

The development approach has been taken with the selection of latest technology
based DAQ cards and the simulation software design and development. The brief of
the Hardware used for the development is given below:
A. Intel i7 based Computer: This is working on 3.4 GHz quad core processor for
performance enhancement. It is with 8 GB DDR4 with 64-bit operating
system.
B. Interface Card- ADC: The PCI based card with 16 channels, 16 bit and +/-10 V
range is selected with latest technology. The IPCAP 1602 is used for HILS.
C. Interface Card-1553: This is dual node MIL-STD 1553 card used for the
sending the data at 1Mbps serial interface. It is used as Remote Terminal (RT)
mode for the simulation.
D. Interface Card-DAC: This is 8 channels, 16 bits with output voltage of +/- 10V.
This card is also have DIO inbuilt that is required for simulation, so this card
is used as DAC as well as DIO interface. Adlink 6208 card is used for this
purpose.
This hardware architecture is used for the development of simulation system[6].
3.2 Software Methodology
Xenomai is a real-time development framework that enables the creation of real-
time applications alongside a general-purpose operating system (GPOS) like Linux.
It's designed to provide a robust real-time environment within a Linux-based system,
allowing for precise control over timing and execution which is crucial for
 Contribution Title (shortened if too long) 5

applications where timing predictability is essential. Here are some key details about
Xenomai.

Fig. 2. Xenomai Architecture block diagram.

1. Dual-Kernel Approach: Xenomai uses a dual-kernel approach, which allows it to

provide real-time capabilities without compromising the performance and flexibility
of the Linux kernel. It runs a real-time core (Cobalt) alongside the Linux kernel,
handling time-critical tasks while the Linux kernel handles non-critical tasks.

2. Real-Time APIs: Xenomai provides a variety of real-time APIs, such as POSIX,

VxWorks, and pSOS, allowing developers to port applications from other real-time
operating systems with minimal changes.

3. Low Latency: One of the primary advantages of using Xenomai is its ability to
achieve very low latency in task scheduling and interrupt handling, which is essential
for real-time applications.

4. Compatibility: Xenomai is compatible with a wide range of hardware platforms

and architectures, including x86, ARM, and PowerPC.

Integrating a 6 Degrees of Freedom (6DoF) sensor with RS-422, MIL-STD-1553,

ADC (Analog-to-Digital Converter), and DAC (Digital-to-Analog Converter)
involves combining multiple communication protocols and data acquisition systems
to create a cohesive system. As per the project configuration On board computer is
generating the command for every @2.5ms of iteration. For this thread is created ad
for @2.5ms periodicity in simulation computer. Using input files reading the
configuration of test process and thrust data, aero data. After that created pipes for
inter process communication (IPC) to share the mission parameters from kernel to
user space. Device drivers are for communicating subsystems to simulator PC. For
this we are using different types of device drivers in our project. As per the manual
we are initialized the drivers into Simulator PC. The Lift_off signal is given by the
OBC to the simulation computer. Then based on the deflection commands/feedback
6 F. Author and S. Author

input through ADC interface sixdof computation has been done. These computed
rates are sent to OBC via MIL-STD 1553 interface. All the critical events have been
captured through Digital I/O interface with polling @2.5 ms. In the case of Flight
motion simulator (FMS) for the SIL and SAIL test cases DAC channels are used to
send Euler angles to the FMS to validate Gyros in close loop. This all the simulation
has been carried out in the real time and the simulation data is stored in the files for
future analysis.
There are number of problems faced during the development phase. The IRQ
allocation to the respective interface devices have been solved. The periodicity of
2.5ms is measured and initially it was deviating due to the improper scheduling. The
scheduling have been programmed in a systematic way to resolve this problem. The
dual node MIL-STD 1553 nodes activation is the another problem faced.

Fig. 3. Software flowchart and the software code for the simulation software.

4 TESTING AND RESULT ANALYSIS

After the successful development of the simulation system many test cases has
been given to test the performance of the real time operation and interface device
communication[7]. Following are the configuration and tabulated in Fig. 4:
OBC In Loop Simulation (OIL): Control and guidance algorithm is validated.
Full Stimulation In Loop (FSIL): Only Navigation algorithm is validated.
Sensor In Loo (SIL): The dynamics of the Gyros & accelerometer validated.
Sensor Actuator In Loop (SAIL): both sensors and Actuator together validated.
 Contribution Title (shortened if too long) 7

Fig.4. Table of the List of Test-cases for evaluation of performance of system

The data analysis is carried out with the old system and the new system to compare
the performance of the simulation system. The Fig. 5 shows that all the events as well
as the data are closely matching with the proven system performance so the newly
developed system has been deployed in HILS test-bed for validation of the avionics.

Fig. 5. Comparision of the old and new System.

-3
x 10
6
RT-7 (EMA1)

4
0 0.5 1 1.5 2 2.5 3 3.5 4
4
x 10
-3
x 10
RT-11 (EMA2)

4
0 0.5 1 1.5 2 2.5 3 3.5 4
4
x 10
-3
x 10
RT-13 (EMA3)

4
0 0.5 1 1.5 2 2.5 3 3.5 4
4
x 10
-3
x 10
RT-14 (EMA4)

4
0 0.5 1 1.5 2 2.5 3 3.5 4
4
x 10

Fig. 6. Latency and Jitter Evaluation of the System.

The Latency and the jitters values has been monitored for number of runs. The
latency is minimum and within the specified values. The Fig. 6 shows that the data
captured @5ms have shown consistent performance and no sample has been dropped
8 F. Author and S. Author

or deviating from the sampling rate. This proves the latency and jitters values have no
affect on the performance of the system.

5 Conclusion and Suggestions

The development to the new technology is essential and mandatory as the old
technology is getting obsolete. On the similar lines, the HILS setup is developed on
the latest intel based technology with the real time kernel Xnomai. As discussed in the
paper, both hardware architecture and software operating system is upgraded to latest
technology based simulation system. The computational capabilities and the
performance is enhanced by using this latest technology with many of the features
like multi-core programming and many more has been added to the architecture. The
driver development with all the interface I/Os has been done successfully and sixdof
algorithm is executing with the specified real time execution period. The jitter and the
latency is very low and within the acceptable limit. Number of test-case have been
given to test the performance before deployment. The future suggestion is to integrate
simulation system with IoT kind of wireless control to monitor and configure HILS
system without human intervention.

References
1. Wagle, R., Sharma, P., Amin, M., Gonzalez-Longatt, F. (2023). Cyber-Physical Co-
simulation Framework Between Typhon HIL and OpenDSS for Real-Time Applications.
In: Tripathi, S.M., Gonzalez-Longatt, F.M. (eds) Real-Time Simulation and Hardware-in-
the-Loop Testing Using Typhoon HIL. Transactions on Computer Systems and Networks.
Springer, Singapore. https://doi.org/10.1007/978-981-99-0224-8_14
2. Karvande Rajesh (2014) cPCI based hardware-in-loop simulation system development
under real time operating system. In: ICACCI-2014, pp 736–739
3. Prasad R, Shanmukh G, Karvande Rajesh (2014) Analysis of various RF-interference
effects in GPS L1 C/A code receiver in tracking mode. In: IEEE conference, pp 1–6
4. Barik, C.R., Vijayan, K., Sha, O.P. et al. Hybrid modelling and analysis of cutter suction
dredger using hardware and software in the loop. J. Ocean Eng. Mar. Energy (2024).
https://doi.org/10.1007/s40722-024-00327-z
5. Karvande RS, Ramesh Kumar B (2013) Development of HILS test-bed to test newly
developed INS system. In: Proceedings of IEEE conference, ICMIRA. pp 536–539
6. Li, D., Ye, Q., Guo, X. et al. HILP: hardware-in-loop pruning of convolutional neural
networks towards inference acceleration. Neural Comput & Applic 36, 8825–8842 (2024).
https://doi.org/10.1007/s00521-024-09539-8
7. Todorović, I., Isakov, I., Gecic, M. (2023). Development of Electric Vehicles Applications
Using AURIX™ Microcontroller and Typhoon HIL. In: Tripathi, S.M., Gonzalez-Longatt,
F.M. (eds) Real-Time Simulation and Hardware-in-the-Loop Testing Using Typhoon HIL.
Transactions on Computer Systems and Networks. Springer, Singapore.
https://doi.org/10.1007/978-981-99-0224-8_7