With the improvement of the level of social and economic development and the improvement of living standards, the amount of garbage output in China is increasing year by year, and the problem of garbage disposal is becoming more and more serious. In view of this situation, China began to carry out garbage classification. Garbage classification can give full play to the utilization value of resources and means of production and reduce the impact of human economic activities on nature. Therefore, aiming at the problems of difficult garbage classification and low efficiency of manual sorting, this paper designed and developed an intelligent garbage sorting vehicle. The system combines the background replacement algorithm to enhance the garbage data set, detects and classifies the garbage targets through the improved YOLOv5 network model, and finally realizes the garbage sorting work through the all-terrain intelligent vehicle equipped with mechanical arm. Through the test of garbage sorting in the real environment, the success rate of garbage identification by the intelligent garbage truck is 85%, which can meet the needs of garbage sorting in real life.