Stereovision-based robot motion planning with negotiation-selection mechanism in complex environments

Abstract

An interactive mechanism and modular approaches are proposed for hybrid motion planning. A bidirectional interaction is designed by deliberative candidates negotiating with the feedback on reactive evaluation. The deliberative module constructs a multilayer state lattice with switching control sets. Furthermore, a heuristic multitask-parallel algorithm efficiently generates desirable paths. The reactive module designs a hierarchical structure to integrate reaction optimisation and situation-dependent adjustment. Based on manifold correlations, piecewise criteria rather than a single function are proposed to cater for different stages of planning. Extensive experiments verify the efficacy, reliability and robustness of the approach in complex environments.