Genetic Fuzzy-LQR Control Enhances Satellite Servicing Automation in Cislunar Space

On April 21, 2025, researchers from Thales presented a cutting-edge framework that integrates genetic fuzzy logic with LQR control, enhancing the efficiency and reliability of robotic systems for satellite servicing in space.

The study combines Genetic Fuzzy Trees with LQR control using Thales’ TrUE AI Toolkit to develop a trusted and efficient controller for a two-degree-of-freedom planar manipulator designed for satellite maintenance in cislunar space. The proposed Genetic Fuzzy-LQR method demonstrated 18.5% better performance than optimal LQR on average. It exhibited exceptional robustness to uncertainty, addressing critical safety and efficiency needs for automated satellite servicing as orbital debris increases.

Spacecraft docking is a critical operation in space exploration, requiring precise and reliable manoeuvres to ensure mission success. Traditional control systems, such as linear quadratic regulators (LQR) and proportional-integral-derivative (PID) controllers, have been widely used but often struggle with the complexities and uncertainties inherent in docking scenarios.

Recent research has explored innovative approaches by integrating genetic algorithms (GA) with fuzzy logic to enhance spacecraft docking systems. This hybrid approach aims to address the limitations of conventional control methods, offering a more robust and adaptive solution. The study using Thales’ True AI platform demonstrates significant improvements in docking accuracy and stability under varying conditions.

The research focuses on developing a hybrid control system that combines GA with fuzzy logic to optimise spacecraft docking trajectories. Genetic algorithms are computational techniques inspired by natural selection and evolution, enabling the system to adapt and improve over time. Fuzzy logic allows for handling imprecise or uncertain information, making it suitable for complex real-world scenarios.

The researchers tested their approach under various uncertainties, including variations in spacecraft mass, external disturbances, and communication delays. By simulating these conditions, they evaluated the performance of their hybrid system against traditional control methods. The results showed that the GA-fuzzy logic combination outperformed LQR and PID controllers in terms of docking accuracy and stability.

The findings suggest that integrating genetic algorithms with fuzzy logic offers a promising solution for spacecraft docking systems. The hybrid approach demonstrated superior performance in handling uncertainties in space environments. This advancement could lead to more reliable and efficient docking operations, reducing the risk of mission failures and improving the overall success rate of space exploration missions.

The use of Thales’ True AI platform underscores the practicality of this research. By leveraging advanced artificial intelligence tools, the researchers could simulate and optimise their control system in a realistic environment, paving the way for potential real-world applications.

This study represents a significant step forward in spacecraft docking technology. By combining genetic algorithms with fuzzy logic, researchers have developed a more robust and adaptive control system that can handle the complexities of space operations. As space exploration continues to expand, such innovations will play a crucial role in ensuring the safety and success of future missions.

Integrating advanced AI tools like Thales’ True AI platform highlights the growing importance of computational methods in solving real-world challenges. With further development and testing, this hybrid control system has the potential to revolutionise spacecraft docking operations, making them safer, more efficient, and more reliable.