Along with the development of robot technology,a robot not only need to complete a specific task, but asloneed to do path planning in the process of performing thetask. So, path planning is widly studied. This paperintroduce a method of robot path planning based onreinforcement learning,which aimed at Markovian decisionprocess. In this paper, we introduce the basic concept,principle and the method of reinforcement learning andsome other algorithms.Then,we do research from singlerobot’s path planning in the static invironment based on Qlearning,and describe the application of this algorithm onthe path planning by setting off state space and action spacereasonablly and designing reinforcement function.Byedditting Matlab program,we do some simulationexperiments,which incarnate the algorithm visually and getthe optimal path.
展开▼
