reinforcement learning application of multi-agent path planning

in this paper, we talk about the path planning problem of two robots. this robot routing has been trained through reinforcement learning. in this paper, then, we are referring to reinforcement learning and the q-learning method, and after defining the environment and problem constraints, defining the states vector and the actions vector, and defining the reward function, problem solved and simulation results are shown by using matlab software. the wnvironment is a 7x8 cell that bobots must learn to move towards the target point and pick up the target without colliding with obstacles, and then reach the goal point.