A 7-DOF practical motion planner based on sequential framework: theory and experiments

摘要

Presents an implementation of a path planner for a 7-DOF manipulator in 3-D environments. This implementation is based on a previously reported approach called the sequential framework to develop practical motion planners for many-DOF arms. The essence of the sequential framework is to exploit the serial structure of manipulator arms and decompose the n-dimensional problem of planning collision-free motions for an n-link manipulator into a sequence of smaller m-dimensional sub-problems (with backtracking), each of which corresponds to planning the motion of a sub-group of m-1 links along a given path. Each of the sub-problems (m=2) is solved by using numerical potential fields defined over bitmap-based representations of the 2-dimensional sub-spaces. Furthermore, an efficient backtracking mechanism based on a novel notion of virtual forbidden regions in these 2-dimensional sub-spaces is used. The author has implemented the above approach for a real 7-DOF manipulator in realistic 5-dimensional environments. Although it is not complete, the planner succeeded in several difficult examples with only a few levels (/spl les/2) of backtracking with run times ranging from ten minutes to an hour depending on the task difficulty.