A Real-Time Negative Obstacle Detection Method for Autonomous Trucks in Open-Pit Mines

Negative obstacles such as potholes and road collapses on unstructured roads in open-pit mining areas seriously affect the safe transportation of autonomous trucks. In this paper, we propose a real-time negative obstacle detection method for self-driving trucks mines. By analyzing characteristics mines, target model based Yolov4 network was built. It uses RepVGG backbone feature extraction network, applying SimAM space channel attention mechanism to multiscale fusion. addition, classification prediction modules are optimized improve accuracy with which it detects targets. A non-maximum suppression optimization algorithm (CIoU Soft Non-Maximum Suppression, CS-NMS) is proposed post-processing stage detection. The CS-NMS calculates confidence each frame weighted solve problems encountering obscure or poor positioning boxes. experimental results show that research achieves 96.35% mAP detecting speed 69.3 fps, can effectively identify mines complex backgrounds.