An improved joint space Astar algorithm for a 6-DOF manipulator with pre-planning strategy

Efficiency in manipulator path planning and obstacle avoidance represents a crucial technological challenge in manipulator research. Traditional path planning methods have certain limitations when dealing with the obstacle avoidance for links of manipulator, such as large calculational amount or unstable planned paths. To address this, this paper presents an improved joint space Astar algorithm for a 6 degree of freedom (DOF) manipulator with a pre-planning strategy. Firstly, we propose a pre-planning strategy that can refresh start positions and target positions to reduce computational load. Next, we perform a three-dimensional Astar path planning method in joint space, which avoids numerous inverse kinematics computations and singularities, and reduces the amount of high-dimensional calculations. In addition to the obstacle avoidance for manipulator end-effector, we also realize the obstacle avoidance for manipulator link. In simulation and experiments, the effectiveness of the proposed path planning algorithm is validated and the path tends to be optimal compared to the conventional Astar, RRT and Dijkstra algorithms.