This paper presents a new method for planning collision-free paths for robots with any number of joints. It is particularly well suited for use with kinematically redundant robots. The algorithm is general in that it does not impose restrictions on the geometry, motion, or payload of the robot. It does not try to locate an optimal path. Instead, it attempts to locate a reasonable path while mapping a minimal amount of configuration space (c-space). The method involves iteratively modifying a connected path between the initial and goal configurations to avoid all intervening obstacles. Information from the world model is used to guide path modification. This approach is of particular value in high-dimensional cases for which exhaustive searches are impractical. In the worst case, the algorithm maps a straight-line path in c-space to the goal and the surfaces of the interfering obstacles along this path. An example for a seven-degree-of-freedom robot is included.

This content is only available via PDF.
You do not currently have access to this content.