Automatic cruise and intelligent obstacle avoidance method of geological radar auxiliary equipment based on YOLOX-nano lightweight target detection

Automatic cruise and intelligent obstacle avoidance method based on YOLOX-nano lightweight object detection, include, Single-stage (One-Stage) target detection algorithm based on regression; YOLOX-nano road detection method based on the least squares method; YOLOX-nano-obstacle recognition distance estimation method based on single visual ranging method; Autonomous cruise control strategy based on a PID control algorithm The proposed method of the present invention provides, by means of indirect perception, Integrate the lane boundary line inferred by target detection and the information of obstacle category and location size, and the obstacle distance information calculated by single visual ranging, Develop automatic cruise and intelligent obstacle avoidance strategies for geological radar auxiliary equipment, The method utilizes the powerful feature learning capabilities of deep learning methods, While having excellent generalization and robustness, Also have the interpretability of traditional methods, Easy the analysis, debugging and optimization of geological radar auxiliary equipment, Improve the system performance and stability.