基于YOLOv5s和单目测距的无人机目标追踪算法设计

Design and implementation of UAV target tracking algorithm based on YOLOv5s and monocular ranging

针对无人机目标识别与追踪算法的系统设计,通过YOLOv5s目标检测算法与CSRT目标追踪算法,搭建了无人机基于视觉的目标识别追踪模型,提出了单目测距法快速实现目标测距,并结合鲁棒性强的PID控制算法对无人机进行高精度控制。实验结果表明,无人机总能保持在与追踪目标规定范围内,顺利实现目标追踪,证明此目标追踪算法具有可行性,且有较好的反应效率和追踪性能,为今后无人机目标检测与追踪算法的研究提供了一定的参考价值。

For the system design of UAV target recognition and tracking algorithm,the vision-based target recognition and tracking model of UAV is built by YOLOv5s target detection algorithm and CSRT target tracking algorithm,the monocular ranging method is proposed to achieve target ranging quickly,and combined with the robust PID control algorithm to control the UAV with high precision.The experimental results show that the UAV always stays within the prescribed range with the tracking target and achieves the target tracking smoothly,which proves that this target tracking algorithm is feasible and has better response efficiency and tracking performance,and it provides a certain reference value for the future research of the target detection and tracking algorithm of the UAV.