The PIRATE, a Pipe Inspection Robot for AuTonomous Exploration, is currently being developed at the Robotics & Mechatronics (RaM) research group. For efficient inspection of pipelines, operating the PIRATE has to be straightforward, such that the operator can mainly focus on examining the condition of the pipe. Therefore, a certain degree of autonomy of the robot is required, i.e. autonomous movement through pipe bends and obstacle detection and avoidance.
The objective of this assignment was to create a sensing system that enables the PIRATE to accurately identify pipe-junctions and obstacles in front of it. This, to eventually navigate these autonomously without interference from an operator. The focus was on determining the proper sensor hardware and software methods to implement. Some key requirements were to obtain a wide Field of View (FoV) and accurate sensor data, that can easily be interpreted by the detection algorithm, which has to operate as efficiently as possible.
