PRRT: An Intelligent Planning Algorithm with Memory

Abstract This paper proposes a practically executable path planning method, namely, Pheromones-RRT(PRRT), for robots with large joint range in complex environment. To inter-activate the real world, point cloud is utilized as scene information, while sampling, pheromones approach designed to describe pheromone content carried by each sampling point. During process, random nodes are performed probability of ε, or those highest current tree selected 1-ε and sampled their neighborhood. avoid local minimum problem, concept volatile factor (PVF) proposed, expansion, double trees also generated PRRT both cartesian configuration spaces improve speed algorithm. The accumulation enables certain learning abilities, reducing randomness process. Simulation results show that proposed method can effectively plan an optimal obstacle avoidance satisfactory performances compared RRT-Connect.