An Autonomous Clutter Inspection Approach for Radiological Survey Using Mobile Robot

In the event of a radiological emergency, scattered clutters around the accident sites should be thoroughly inspected as they are potentially contaminated and require proper safety handling. The goal of this work was to localise the position of the hotspot within the clutters and estimate their respective radiation intensity using a mobile robot. Given the information from a 2D occupancy map, we proposed a strategy to identify the position of clutters and formulate waypoints that simultaneously serve as sampling positions to acquire radiation measurement. Particle filter was adapted to fuse the measured data and provide the results in real-time. We also introduced clutter contamination analysis that takes into account the limitations of the detector to derive the worst-case scenario; thus, preventing misclassifications of hotspots with low radiation intensity as clean areas. The approach was implemented in ROS and a commercial mobile robot with a Geiger Muller detector. Finally, we evaluated the overall performance in simulation and real-world experiments. Note to Practitioners—This work was motivated by the goal of realising automation of clutter inspection using a mobile robot to complement the current achievement in robotics radiation mapping. Currently, we still rely on human workers to conduct a thorough radiation survey, including detailed inspections on all clutters, before clearing any radiological accident or incident site. The theoretical framework in this paper proposed a practical solution to localise and estimate the intensity of the hotspot within the clutter in real-time. We demonstrated the versatility and performance of our proposed approach using a small commercial robot and Geiger Muller tube (typically available in any radiation facilities). This method reduces radiation exposure to human workers and provides a cost-effective solution for preliminary screening in radiological emergency situations.