RRT*-APF Hybrid Path Planning for Reconfigurable Cable-Driven Parallel Robots

When Cable-Driven Parallel Robots (CDPRs) are doing some complex work, obstacles in the environment will interfere with cables and the mobile platform. It is a meaningful work to avoid these disturbances by planning the path of CDPRs. This paper presents an optimal path planning strategy for a reconfigurable CDPR. By a variant of Rapid-exploration Random Tree (RRT) to find the optimal collision avoidance path gradually, Artificial Potential Field (APF) is used to guide the generation of random tree nodes, reducing the time of finding the path, and ensuring the safe distance between the platform and obstacles, and generating the shortest collision-free path. Post-processing algorithm reduce the number of path points. Detect whether the generated path will cause obstacles to interfere with the cable. If there is interference, optimize the robot's stiffness and minimize cable tensions to determine the optimal configuration. By adjusting the location of the cable connection points on the fixed frame, the obstacle avoidance of the cable is realized. The simulation results demonstrate the RRT
${}^*-\text{APF}$
hybrid path planning algorithm's ability to successfully find the path to avoid collision of the mobile platform in the environment with obstacles, the reconstruction algorithm can find the best collision-free configuration of the cable.