Collision-free path planning of cable-driven parallel robots in cluttered environments

초록

Path planning of cable-driven parallel robots (CDPRs) is a challenging task due to cables which may cause various collisions. In this paper, three steps are suggested to perform path finding of CDPRs in cluttered environments. First, a way to visualize the cable collision of CDPRs is suggested to consider actual workspace of CDPRs. Second, a path finding algorithm based on rapidly exploring random trees (RRT) is presented to find a path free of various collisions of CDPRs including cable collisions and wrench feasible workspace. While conventional RRT algorithms are mainly focused on mazy environments, the modified RRT algorithm proposed here directly connects a sampled node and the tree to find a path faster in a non-mazy, but cluttered environment. Goal-biased sampling algorithm is also modified and employed to decrease computational cost. To deal with complicated collision detection of cables in the RRT, Gilbert-Johnson-Keerthi algorithm was employed. Finally, post-processing algorithm for any waypoint-based path is suggested to get a shorter and less winding path. A numerical study was carried out to suggest choosing proper meta parameters for the post-processing algorithm. The suggested algorithms were evaluated with 1000 times of simulations, and they were equally carried out for RRT* to compare. According to the results, the suggested algorithm found a shorter path with less computation time compared to RRT* and the post-processing algorithms made the already found path shorter.

키워드