1989 Volume 7 Issue 5 Pages 414-425
A motion-planning algorithm is proposed, which fulfills its function fast even if shapes of a mobile robot and its obstacles are complicated. A motion-planning algorithm generates robot motions tying start and goal positions on condition that the mobile robot does not collide with its obstacles. First considering globally obstacle allocation in the robot workspace, the algorithm selects intermediate positions where the mobile robot should pass. Secondly using the closest-points between the robot and the obstacles, the algorithm generates collision-free robot motions to join these intermediate positions systematically.
The proposed algorithm runs on the quadtree representing the robot workspace. The quadtree controls all obstacle regions hierarchically in positioning, and therefore these intermediate positions and the closest-points can be fast selected out of the quadtree even if shapes of the robot and its obstacles are complex. Thus the algorithm generates efficiently collision-free robot motions between start and goal positions in such a case.
Finally in comparison with several motion-planning algorithms presented previously, it is shown from some experiments that the proposed algorithm selects fast collision-free robot motions.
