DESIGN AND IMPLEMENTATION OF A MULTIPLE AGV SCHEDULING ALGORITHM FOR A JOB-SHOP

Considering the actual needs of the job-shop, this paper establishes an experimental platform for automated guided vehicles (AGVs) with a six-wheel dual-drive mechanical structure, and designs a multi-AGV scheduling system for unmanned factories. The scheduling system works in the following steps: Firstly, the deviation of each AGV from the magnetic strip is calculated based on the data of the magnetic sensor, and the speeds of left and right drive wheels are adjusted based on the deviation, keeping the AGV moving stably on the magnetic strip. Next, the A* algorithm is called to plan a collision-free and efficient path. Finally, the conflict points and AGV priorities are determined by comparing the paths of different AGVs, and the paths are planned again if head-on collision may happen on the conflict points. In this way, multiple AGVs can operate coordinately on the same map. The proposed multi-AGV scheduling system was proved feasible through simulation and experiments. Our system can be applied widely in manufacturing factories to replace traditional manual handling and conveyor belt transmission, reduce labour cost, and improve production efficiency.