Dynamic path planning of a mobile robot with improved Q-learning algorithm

Path planning of a mobile robot under dynamic environment is a difficult part of robot navigation. In this paper, a new path planning method based on improved Q-learning (IQL) algorithm and some heuristic searching strategies is proposed for mobile robot in dynamic environment. A new exploration strategy which combines ε-greedy exploration with Boltzmann exploration is used in IQL. In addition, the heuristic searching strategies are provided to reduce the search space and limit the variation range of orientation angle. From simulations, the better performance of the proposed method was certified in terms of time taken and optimal path comparison with classical Q-learning (CQL) and other planning methods. Meanwhile, the reduction in orientation angle and path length has significance in the robotics literature of the energy consumption.