Millimeter-Wave Radar-Based Autonomous UAV Tracking Technology and ROS Simulation Environment

To address the challenges of accuracy and robustness in UAV autonomous tracking under multi-target dynamic scenarios, this study proposes a millimeter-wave radar-based UAV tracking system incorporating improved DBSCAN clustering and Kalman-PID control. The implementation involves three key phases: Firstly, an autonomous UAV tracking system with millimeter-wave radar is established on the ROS platform. Secondly, time-domain sliding average filtering is applied to compensate for radar signal distortion, effectively resolving motion-induced point cloud ambiguity. Finally, front-end multi-target separation is achieved through distance-weighted DBSCAN algorithm, while back-end trajectory tracking is accomplished via cascaded Kalman filtering and PID controller. Experimental results demonstrate static target positioning accuracy of less than or equal to 0.18 meters and dynamic target tracking error of less than or equal to 0.4 meters, representing a 31.5 percent improvement in accuracy compared to traditional fixed-parameter DBSCAN-Kalman filter hybrid methods.