Autonomous Navigation: Integrating Stability Concepts with RRT* in Complex Environments
Youssef Mhanni, Youssef Lagmich · 2024
Navigating complex, multi-obstacle environments pose significant challenges for autonomous robotic systems. To address these challenges, this study introduces a novel hybrid approach that combines Lyapunov control with Rapidly-exploring Random Trees (RRT*) to enhance path stability and efficiency. Lyapunov control ensures the robot maintains a stable trajectory, actively correcting deviations to prevent collisions, while RRT* efficiently explores and generates viable paths around multiple obstacles, adapting to dynamic changes in the environment. Our integrated method leverages the deterministic robustness of Lyapunov control alongside the probabilistic pathfinding capabilities of RRT*, creating a powerful tool for real-time navigation in cluttered and unpredictable settings. Through extensive simulations, we demonstrate that this approach not only improves the safety and reliability of autonomous navigation but also optimizes pathfinding in terms of speed and computational resource usage. The results underscore the potential of this hybrid method to significantly advance the field of robotics, offering a scalable solution that can be adapted to various real-world applications where autonomous systems must operate safely and efficiently amidst numerous obstacles.