Accurate Real-time Obstacle Detection Of Coal Mine Driverless Electric Locomotive Based On Odel-yolov5s

This article presents a deep learning-based ODEL-YOLOv5s detection model for the accurate identification and real-time detection of obstacles in coal mine driverless electric locomotives. The proposed model is based on the conventional YOLOv5s and introduces several data augmentation methods, an attention mechanism, a four-scale prediction, and an optimized localization loss function and non-maximum suppression method to improve the detection accuracy and speed. The experimental results show that the proposed model has a mean average precision (mAP) of 98.9%, an average precision of 97.9% for small obstacles, and a detection speed of 60.2 FPS.