Factor-guided motion planning for a robot arm

摘要: Motion planning for robotic arms is important real, physical world applications. The with high-degree-of-freedom (DOF) hard because its search space large (exponential in the number of joints), and links may collide static obstacles or other joints (self-collision). In this paper we present a motion algorithm that finds plans from one arm configuration to goal 2D assuming no self-collision. Our unique two ways: (a) it utilizes topology factor reduce complexity problem using dynamic programming; (b) takes only polynomial time under some conditions. We provide sufficient condition polytime 2D-space arms: if there path between homotopic configurations, an embedded local planner within time. experimental results show proposed improves performance arms.