Review on reliability in pipeline robotic control systems

摘要

With the increasing use of the pipeline transport facilities, the safety and efficiency of the pipeline systems are very important to energy security. Almost all traditional pipeline inspection methods have shortcomings in one way or another, such as heavy workload and low accuracy. Pipeline robot, effective detection equipment, can do detection work in the narrow place beyond human reach. But to the overall technical level, the research still stays in development stage, far away from massive practical applications because of the reliability issues. This paper reviews the significance and prospects, and representative pipeline robots developed at home and abroad. On the basis of the survey of various locomotion modes, the popular methods used in reliability design such as neural network and reinforcement learning are commented for the future main development.