A UAV Trajectory Tracking Algorithm Considering Dynamic Obstacles

In contemporary aerial combat scenarios, the increasingly complex and contested operational environment necessitates robust and agile autonomous navigation capabilities for Unmanned Aerial Vehicles (UAVs). Specifically, enhanced path tracking coupled with real-time dynamic obstacle avoidance is critical for mission success. To address the challenge of UAVs encountering unforeseen or unmapped obstacles during trajectory following, this paper introduces a novel algorithmic framework that synergistically integrates L1 adaptive control with the Dynamic Window Approach (DWA). Dynamic obstacles are modeled through time-parameterized coordinate functions, and real-time obstacle information is assumed to be acquired via radar or equivalent sensor systems. In the absence of detected threats, the L1 adaptive controller ensures precise trajectory tracking along a predefined path. Upon obstacle detection, the DWA is activated to enable reactive collision avoidance maneuvers. To further enhance safety and minimize potential conflicts with dynamic obstacles, a dynamic obstacle evaluation metric is incorporated. Furthermore, fuzzy logic control is strategically implemented to dynamically modulate the DWA weight parameters, optimizing trajectory generation for enhanced conflict resolution. Extensive trajectory tracking simulations, conducted within diverse and challenging operational environments, rigorously validate the robust self-adaptive capabilities of the proposed methodology. The algorithm demonstrably achieves effective trajectory tracking performance across various and speed regimes, while simultaneously and successfully accomplishing dynamic obstacle avoidance.