An assistant algorithm model for a mobile robot to pass through a concave obstacle area

Abstract

When a mobile robot passes through a concave obstacle area, the limitations of the traditional control algorithm add a significant challenge to the mobile robot to move through the concave obstacle area efficiently. Improving the conventional path planning algorithm is a common choice to solve this problem. However, because various path planning algorithms adapt to the fixed scene and the algorithm steps are complex, it is difficult to popularize these algorithms. In addition, conventional path planning algorithms usually encounter difficulties when passing through recessed obstacle areas or cause robot deadlock. An auxiliary algorithm strategy is proposed to help mobile robots avoid concave obstacles quickly. The algorithm strategy plays an auxiliary role by incorporating the judgment of the starting position of the mobile robot, the building of a virtual wall to shield the concave obstacle area, and the judgment of the moving direction of the mobile robot. The advantage of this algorithm strategy is that it does not change the operation mechanism of traditional path-planning algorithms and is highly adaptable. The results of the comparison of simulation experiments with the APF (Artificial Potential Field), the ACO (Ant Colony Optimization), the RRT (Rapidly-Exploring Random Trees), the A-star algorithm, and the Dijkstra algorithm show the inclusion of the auxiliary algorithm strategy resulted in a significant increase in the efficiency of the mobile robot through the depressed obstacle zone.Article Highlights●This auxiliary algorithm strategy can avoid the self-locking caused by the mobile robot falling into the concave obstacle area.●This auxiliary algorithm strategy does not change the mechanism of the traditional path planning algorithm, and it is simple and easy to be used.●This auxiliary algorithm has wide adaptability (it can assist the mainstream path planning algorithm and adapt to various scenarios).