A Collision-free Path Planning Algorithm for Industrial Robots with Adaptive Search Mechanism

In path planning for industrial robots, collision avoidance is a critical constraint. This paper studies the problem of collision-free motion planning for a spraying robot, i.e., obtaining a series of joint configurations along a given spraying waypoint that satisfies the collision constraints and satisfy the kinematics of the robot. Aiming at this problem, a collision-free motion optimization algorithm based on the adaptive search is proposed in this paper. First, a boundary volume hierarchy (BVH) of the robot and environment is quickly generated based on the GPU parallel architecture. Second, the redundancy of spray incidence angles is used to search for feasible spray incidence angle directions in the redundant space and update the robot's BVHs for collision detection. In addition, during the optimization iterations, the size of the search space is adaptively reduced according to the number of collision failures in the previous iterations to avoid falling into a local optimum. Finally, the effectiveness of the proposed algorithm is verified by simulation on a six-degrees-of-freedom robot performing a spraying task.