Design and research of automatic road defects detection model for intelligent unmanned vehicle
As the transportation network expands,the road mileage continues to grow,and the pressure on road maintenance is also increasing.In response to this problem,this paper proposes an automatic defect detection model for intelligent unmanned ve-hicles that is suitable for roads.The model is based on object detection and uses YOLOv8 as the base model.The model is trained using defect data.In this paper,the training parameters of the model are adjusted first,and the best training parameters that achieve the best results in the same training cycle are obtained.The entire dataset is then trained,and the mAP value of the model is 45.8%.When the model is applied to intelligent autonomous vehicles,it can accurately and quickly identify various road defects.
万方数据
