I.M. Antonov1, A.R. Safin2

1 V.A. Kotelnikov Institute of Radio Engineering and Electronics of the Russian Academy of Science (Moscow, Russia)
2 National Research University Moscow Power Engineering Institute, V.A. Kotelnikov Institute of Radio Engineering and Electronics (Moscow, Russia)

1 mcplusplus@mail.ru, 2 arsafin@gmail.com

In modern conditions of rapid development of technologies of unmanned aircraft, robotics and artificial intelligence, the use of high-precision means of virtual modeling and simulation is becoming increasingly important. Aviation simulators are becoming a key tool at the design, testing and optimization stages of both hardware and software components of aircraft. In this context, the open FlightGear platform is of particular interest. It is one of the most popular and functional open source flight simulators. The purpose of this study is to analyze the capabilities of FlightGear as a universal environment for development, verification and debugging of control, guidance and diagnostics systems for unmanned aerial vehicles (UAVs) in a virtual simulation environment.

FlightGear has a number of advantages that make it a promising tool for scientific and industrial research: modular architecture, the ability to deeply configure parameters, support for external tools (for example, MATLAB/Simulink), extensive capabilities for modeling physical processes and interacting with the environment. Special attention is paid to the mathematical engines JSBSim and YASim, which provide accurate reproduction of flight dynamics and aerodynamic characteristics of aircraft. JSBSim, which is the main default engine, offers flexible configuration via XML files, which allows you to effectively simulate different types of aircraft and their behavior in different conditions. Additionally, the possibility of integrating external models implemented in specialized modeling packages has been considered, which expands the analytical and computational capabilities of the simulator.

One of the key uses of FlightGear is prototyping and debugging algorithms for autonomous control and guidance of UAVs. Two main methodologies have been presented: Software-in-the-Loop (SITL) and Hardware-in-the-Loop (HITL). The SITL methodology involves testing the management system software in a virtual environment without using real hardware, which allows you to quickly check and adjust algorithms. The implementation of HITL in FlightGear has been carried out through integration with the MAVLink protocol, which opens up the possibility of testing real flight controllers in simulated flight conditions. This approach minimizes the risks associated with early trials in real-world conditions, while at the same time ensuring a high degree of reliability of the results obtained.

Another important task solved with the help of FlightGear is the modeling of the external environment and influencing factors such as wind, weather, terrain and radio navigation conditions. The simulator supports two wind models: a simplified one suitable for basic scenarios, and an extended one based on spectral methods (Dryden, Karman), which takes into account local meteorological conditions and topography. It also provides opportunities for modeling clouds, precipitation, changes in visibility and other weather events, which is especially important when testing computer vision and navigation systems.

Special attention has been paid to the use of FlightGear to generate training data necessary for training and retraining neural network models. The use of synthetic data obtained in a virtual environment makes it possible to generate a variety of sample sets covering standard and critical flight modes. This is especially important in the development of autonomous control systems, where neural network algorithms are used to predict motion parameters, aim at a target, and recognize objects and obstacles. Combining synthetic data with empirical data obtained in real flights ensures high adaptability and accuracy of machine learning models that take into account the specifics of a particular UAV.

FlightGear provides a unique opportunity to model multicomponent systems, including groups of UAVs, which opens up new horizons in the development of group management and coordination systems. By running multiple simulator instances and organizing data exchange via the UDP protocol, you can create complex scenarios for interaction between devices, such as intercepting targets, joint patrols, or performing missions in a changing environment. This allows you to pre-evaluate the effectiveness of algorithms in real-world interaction and conduct scalable testing.

The conclusion highlights the importance of FlightGear as a powerful tool for research and development in the field of unmanned aircraft. Its ability to accurately simulate physical processes, reproduce complex meteorological conditions, and provide comprehensive testing of control algorithms makes it a valuable resource for academic and industrial applications. The use of the simulator not only reduces costs during the development and flight testing stages, but also increases the level of safety, reliability and fault tolerance of the final solution.

Antonov I.M., Safin A.R. Using the FlightGear flight simulator in the development of systems and algorithms for unmanned aerial vehicles. Achievements of modern radioelectronics. 2025. V. 79. № 11. P. 60–69. DOI: https://doi.org/10.18127/j20700784-202511-07 [in Russian]

1. Horri N., Pietraszko M. A tutorial and review on flight control co-simulation using Matlab/Simulink and flight simulators. Automation. 2022. V. 3. P. 486–510. DOI: 10.3390/automation3030025ff.
2. Tahir M.A., Mir I., Islam T.U. Control algorithms, Kalman estimation and near actual simulation for UAVs: State of art perspective. Drones. 2023. V. 7. № 6. P. 339. DOI: 10.3390/ drones7060339.
3. Zhang J., Geng Q., Fei Q. UAV flight control system modeling and simulation based on FlightGear. Proceedings of the International Conference on Automatic Control and Artificial Intelligence (ACAI 2012). Xiamen, China. 2012. P. 2231–2234.
4. Arif A., Sasongko R., Stepen. Numerical simulation platform for a generic aircraft flight dynamic simulation. Proceedings of the International Conference on Aviation Technology and Management (ICATeM 2018). Kuala Lumpur, Malaysia. 2018.
5. Vogeltanz T., Jasek R. FlightGear application for flight simulation of a mini-UAV. AIP Conference Proceedings. 2015. № 1648. P. 550014. DOI: 10.1063/1.4912769.
6. Aschauer G., Schirrer A., Kozek M. Co-simulation of Matlab and FlightGear for identification and control of aircraft. IFAC-PapersOnLine. 2015. 48. P. 67–72. DOI: 10.1016/j.ifacol.2015.05.071.
7. De Marco A., D’Onza P.M., Manfredi S. A deep reinforcement learning control approach for high-performance aircraft. Nonlinear Dynamics. V. 111. № 18. P. 17037–17077. DOI: 10.1007/s11071-023-08725-y.
8. Berndt J., De Marco A. Progress on and usage of the open source flight dynamics model software library, JSBSim. AIAA Modeling and Simulation Technologies Conference. Chicago, Illinois. 2009. DOI: 10.2514/6.2009-5699.
9. NASA-TM-2015-218675. End-of-Mission Check Case Appendices. NESC Academy. URL: https://nescacademy.nasa.gov/ src/flightsim/Reports/NASA-TM-2015-218675-EOM_checkcase_appendices.pdf
10. Akmal H., Qu Y. Estimation of wind field velocity and aircraft states with an F-16 lateral guidance control system. International Conference on Information and Automation. 2015. P. 2293–2298. DOI: 10.1109/ICINFA.2015.7279668.
11. Scenery. FlightGear wiki. URL: https://wiki.flightgear.org/Scenery
12. Setting up SITL on Linux. ArduPilot Development Site. URL: https://ardupilot.org/dev/docs/setting-up-sitl-on-linux.html
13. FlightGear Hardware-in-the-Loop Simulation. ArduPilot Development Site. URL: https://ardupilot.org/dev/docs/ flightgear-hardware-in-the-loop-simulation.html
14. Aeroplane flight simulator evaluation handbook. London: Royal Aeronautical Society. 2009.