Time-optimal motion planning for a nonholonomic mobile robot on a barcode map

This paper presents a time-optimal motion planning approach for a nonholonomic mobile robot working in a typical warehouse-like environment, where barcodes are regularly attached on the ground for global localization. In such an application, the state lattice is inherently embedded in the barcode map with an extra orientation dimension, and a search based path planning scheme is adapted. Different from existing approaches, a new group of curvature continuous motion primitives and a corresponding motion primitive selection criterion are introduced. The discrete states and motion primitives are used to construct a search graph, based on which path planning is achieved through an existing heuristic search algorithm ARA* to coordinate the optimality and time performance of the solution. After obtaining the resultant path, numerical integration based time-optimal trajectory planning method is introduced. Simulation results show that the proposed approach could achieve real-time performance with highly efficient trajectory under kinematic constraints and the barcode localization constraint.