Petrochemical gas pipelines have an extensive network in urban areas for various transportation. These pipelines suffer through different loads originated from both interior and exterior of the pipe, as well as oxidation and corrosion of pipe surface that sometimes leads to leakage. Therefore, it has to be inspected at frequent intervals but some parts of the network are difficult to monitor. So, the Pipe Inspection Robot for Autonomous Exploration (PIRATE), a six-segments wheeled snake-like robot is used for inspection. Work has been done on the autonomous motion of the robot, but always with the assumption of known state and environment (through the use of a camera on top of the scene). To achieve the next step of autonomy, localization and mapping of the pipeline network need to be done. The problem of getting a robot to build a map of the environment while moving through the same environment is known in robotics as the Simultaneous Localization and Mapping (SLAM) problem.
Thus, this project focuses on developing a Simultaneous Localization and Mapping (SLAM) framework for the PIRATE in a pipeline network so that it will be easier to inspect and identify defects by just localizing the PIRATE within the pipe.