Abstract
SUMMARYElbows and branches in pipelines produce unique image patterns, which can be used as landmarks for autonomous navigation inside the pipelines. This paper presents two landmark detection methods, known as shadow-based method and laser projection method. The first method uses the landmark's unique patterns of shadow produced by the robot's illuminator. The other method exploits special line features generated by its own line-laser beam projector. The basic algorithms for extracting the landmarks are given and special sensor mechanisms are addressed respectively. Finally, the detection performances of each method are validated in various pipeline conditions by using an in-pipe robot, called MRINSPECT-V.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
