Path planning for manipulators based on an improved probabilistic roadmap method

lThe sampling strategy can make the distribution of samples more appropriate for practical application.lThe efficiency of constructing a high-connectivity roadmap is increased.lThe three-stage connection strategy ensures roadmaps have greater connectivity.lThe improved PMR algorithm is practical in environments with narrow passages. An indispensable feature of an intelligent manipulator is its capability to quickly plan a short and safe path in the presence of obstacles in its workspace. Among the path planning methods, the probabilistic roadmap (PRM) method has been widely applied in path planning for a high-dimensional manipulator to avoid obstacles. However, its efficiency remains disappointing when the free space of manipulators contains narrow passages. To solve this problem, an improved PRM method is proposed in this paper. Based on a virtual force field, a new sampling strategy of PRM is presented to generate configurations more appropriate for practical application in the free space. Correspondingly, in order to interconnect these configurations to form a roadmap, a new connection strategy is designed, which consists of three stages and can gradually improve the connectivity of the roadmap. The contributions of this paper are as follows. The new sampling strategy can increase the sampling density at the narrow passages of the free space and reduce the redundancy of the samples in the wide-open regions of the free space; the three-stage connection strategy for interconnecting samples can ensure a high-connectivity roadmap; through synthesizing the above strategies, the improved PRM method is more suitable for path planning of manipulators to avoid obstacles efficiently in a cluttered environment. Simulations and experiments are carried out to evaluate the validity of the proposed method, and the method is available for manipulator of any degrees of freedom. It shows the process of the new sampling strategy of the PRM algorithm based on a virtual force field. Based on this sampling strategy, the distribution of samples will become more appropriate for practical application, which can solve the problem of narrow passages in the free space of manipulators and improve the efficiency of constructing a roadmap with a high-connectivity. [Display omitted]