Multirobot Allocation In A Flexible Manufacturing System, Using Reinforcement Learning For Decision-Making, Case of Study

This paper presents a reinforcement learning approach for decision-making in assigning multiple robots in a flexible manufacturing system. The algorithm allows for learning the optimal action policy for assigning tasks to multiple robots, outperforming other methods. The case study involved three robots and four tasks, some requiring cooperation. The study demonstrates the effectiveness of the approach in comparison to other methods and the results indicate that the reinforcement learning approach increases production efficiency and reduces task completion time compared to traditional assignment methods.