Key points are not available for this paper at this time.
Climbing robots present transformative potential for automated structural inspections, yet their deployment remains limited by the reliance on manual control due to the absence of effective environment perception and path-planning solutions. The critical bottleneck lies in the difficulty of generating accurate planning maps solely through onboard sensors due to the challenge of capturing open, large-scale, and irregular environments (e.g., cable-stayed bridge towers). This study proposes a building information modeling (BIM)-based complete-coverage path planning (BCCPP) framework, leveraging BIM to enable autonomous robotic inspection. The framework constructs accurate grid maps through BIM data, addressing the map-perception problem for robots in open, large-scale, and irregular environment while refining the boustrophedon-A* algorithm with multi-heuristic optimization, which reduces path repetition and improves energy efficiency. Field and simulated experiments on a cable-stayed bridge tower show the BCCPP achieves 93.5% coverage with 9.1% repetition, and planned paths were executable within a 0.2 m tolerance and collisions avoided. This work bridges BIM, climbing robot, and path planning, offering a scalable solution for intelligent infrastructure inspection.
Xia et al. (Fri,) studied this question.