Research on the Design and Navigation Approaches of Autonomous Mobile Escort Robot Systems

Stable navigation and obstacle avoidance are crucial when a home care robot moves autonomously. This paper proposes a twice-optimized navigation method (RRT-Smooth-A*). Firstly, RRT is used for preliminary path planning, and the Moore cellular automata is adopted to expand the search range and extract potential areas. Secondly, after rasterization, A* is employed for precise path planning. Finally, the Bezier curve is used to smooth the path. Through simulation and tests on a self-developed home care robot, in terms of ineffective steps, it is reduced by 73.39% compared with the traditional A* algorithm and by 27% compared with the RRT algorithm. In terms of path smoothness, the average number of inflection points is decreased by approximately 85.7% compared with the RRT algorithm and by 56.8% compared with the A* algorithm. In addition, an experiment was carried out on the self-developed home care robot in a corridor. The results show that it has a short path, fast planning, high navigation efficiency and is feasible.