A Low‐Cost Curvature Detection Method for an Untethered Biped Pipeline Climbing Robot with Energy‐Efficient Strategy

摘要

Daily cleaning and maintenance of high‐rise buildings with curved surfaces present high‐risk tasks for human workers due to the complexity of working on irregular surfaces. While robots are commonly used for cleaning flat surfaces in buildings, they encounter significant difficulty in automatically aligning the suckers with the target curved surface, which limits their applicability when climbing and cleaning on curved surfaces. This article proposes a low‐cost pipeline curvature detection method and applies it to an untethered biped pipeline surface climbing robot (BPCR) system with a five‐degree‐of‐freedom structure. The BPCR features adaptive vacuum suction cups equipped with multiple laser sensors on the suction modules to detect the curvature of various surfaces. The sensors measure the distance between the end‐effector and pipeline surface in real time, enabling the robot to adjust the posture and securely adhere to the surfaces. The entire system is integrated with a wireless communication system for remote control and real‐time monitoring. Moreover, an energy‐optimized strategy is proposed for improving the endurance of robot remote operations. Experimental results demonstrate that the robot can effectively navigate and attach to the exterior surface of pipelines, providing reliable and efficient cleaning solutions for high‐risk tasks with low energy costs.