A study on autonomous pipeline maintenance robots, 6

This paper deals with the sensing planning for a pipeline inspection and the maintenance robot, by which the robot can carry out inspection tasks to detect a malfunction location in a plant pipeline system autonomously. For this purpose, the robot needs knowledge of the plant map, plant function and plant diagnosis. In the previous report, the path planning expert system (PPES) was reported; if the plant map is given, the robot can automatically produce the path to reach a location from another location within the robot task level. In this paper, PPES is modified to adapt the sensing planning expert system (SPES) and generates executable robot commands for motion control. In addition, the plant knowledge system requires more information concerning plant operation states, such as standard value/status and up/down stream. Furthermore, the robot needs knowledge on inspection/repair and diagnosis, so that the robot can estimate the malfunction candidates and select one individually after some inspection trials. Together with PPES and SPES, the robot becomes intelligent enough to carry out given inspection tasks automatically. (author)