Abstract
There is no safe disposal technique for waste drums containing low-level radioactive waste and toxic substances, so they are stored in remote warehouses and closely monitored for signs of deterioration. There is a potentially very cost-effective approach to drum inspection, based on autonomous robot navigation, computer vision, and change detection based on automatic image registration and subtraction. The operational concept is an autonomous mobile platform with a camera mounted on a custom designed positioning mechanism. The mobile robot will automatically navigate from its charging station to each of the aisles in the warehouse, positioning the camera in front of each drum in an aisle. An image of the drum taken at an earlier date will be compared directly and automatically with a new image of the drum. Any changes that occur may indicate deterioration, and these changes will be analyzed automatically to determine if a human inspector should examine the inspection images to interpret the changes. In this paper we explain the key intelligent systems technologies that make automatic inspection for change feasible, the challenges we face in implementing such a system, the image processing steps required for automatic change detection, and our results to date. We conclude with a discussion of other applications that could benefit from automatic robotic inspection.
Published Version
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