Multi-Robot Path Planning Method Based on Prior Knowledge and Q-learning Algorithms

Abstract

The path planning method based on prior knowledge and Q-learning algorithm proposed in this paper solves the problems of low operating efficiency and slow learning speed in existing multi-robot coordination and collision avoidance. This method can be divided into two stages. Firstly, the improved Q-learning algorithm is used to plan the static collision free path of a single robot in the built model; After the Q-table is initialized with the prior knowledge obtained in the previous step, the Q-learning algorithm is used to achieve conflict-free motion among multiple robots. Finally, the simulation experiment shows that the proposed path planning method based on prior knowledge and Q-learning algorithm performs well in handling collision avoidance path planning of multiple robots.