A method of progressive constraints for nonholonomic motion planning

Presents a general method for nonholonomic motion planning (NMP), that is able to solve problems for systems with high-dimensional configuration spaces. The original NMP problem is replaced by a series of progressively constrained ones that are solved successively, and whose solution path converges towards a solution of the original problem. The nonholonomic constraints are introduced progressively in an iterative algorithm. Each iteration consists of exploring a neighborhood of the path obtained from the previous iteration, searching for a path that satisfies more accurate constraints. The explorations proceed according to a dynamic programming procedure able to deal with optimization criteria.