Constrained Path Planning on Pipeline Surface for Wall Climbing Robots

Han Li,Jianfeng Liao,Shiqiang Zhu,Hongjiang Ge,Wei Song,Xianlei Chen,Jason Gu

DOI: https://doi.org/10.1109/robio55434.2022.10011828

2022-01-01

Abstract:Pipeline structures exist widely in industrial sce-narios and regular maintenance is an essential for sustainability. Manual maintenance suffers from high costs and risks and low efficiency and reliability. Thus, wall climbing robots (WCRs) play an increasingly important role in the field of pipeline non-destructive testing and repairing. Although many commercial wall climbing robots have been put on the market, they rely heavily on manual remote control and are difficult to operate autonomously. In order to improve their automation efficiency, we develop a path planning method based on the characteristics of the scenarios. In this paper, the pipeline surface is be reconstructed as a triangular mesh, thereby transforming the path planning problem into a graph search problem. The step size and the maneuverability and stability of the robot are considered as factors of cost function. By analyzing the attitude transformation of the robot from one vertex of the meshes to another, we then propose a specific mathematical form of the cost function and verify the effect of each term. Given the special topological morphology of the pipeline surface, the geodesic distance to the target point is used as a heuristic function to reduce unnecessary expansion and improve the efficiency of the search. Simulations are carried out to verify the proposed method.