Research on AGV Real-Time Path Planning and Obstacle Detection Based on Machine Vision

With the widespread application of automated guided vehicles (AGVs) in logistics and manufacturing, real-time path planning and obstacle detection have become key factors to improve their operational efficiency and safety. Aiming at the problem of path planning in dynamic AGV environment, this paper proposes a new hybrid path planning algorithm combining the advantages of A* algorithm and Dijkstra algorithm. This algorithm uses the heuristic search ability of algorithm A* to improve the efficiency of path search, and combines the shortest path search feature of Dijkstra algorithm to ensure the optimality of the path. In addition, machine vision technology is used for obstacle detection, environmental information is obtained through the camera, and obstacles are identified and located in real time based on image processing algorithms, so that the AGV can adjust the path in time. Experimental results show that the proposed algorithm can effectively generate safe and efficient driving paths in complex environments. This paper provides a new scheme for the intelligent and automatic application of AGV.