Abstract
There is a growing global demand for safe, environmentally conscious, and long-term radioactive waste disposal solutions. Owing to the potentially deleterious effects of waste evolution during containment, there are strict guidelines and waste packaging specifications which must be adhered to, including rigorous documentation of waste inventories. This mandates a detailed characterisation of waste materials prior to packaging. Waste must be characterised in terms of its material, chemical and radiometric properties. Currently, this characterisation is a highly dangerous and labour intensive process which puts workers at risk. In addition, frequent estimations are made on waste items on the basis of ex-situ sub-samples. These not only sacrifice accuracy in characterisation, but also introduce lengthy time delays into the process.The primary objective of this thesis was to explore the use of robotic systems and other advanced emerging technologies that could be integrated and then implemented to resolve these challenges, fully autonomously and without human intervention. Ultimately, any integrated technology must fulfil a requirement to facilitate, accurate, in-situ characterisation of individual waste objects, while improving worker safety, increasing characterisation accuracy and streamlining an otherwise time consuming process. Hence, the research and development of a prototype fully-autonomous, robotic waste sorting system which can fulfil each of these aforementioned objectives is detailed. The fully-autonomous components were delivered by a variety of robotic manipulator systems integrated with sensory hardware for both manipulation and characterisation. For object manipulations, depth vision cameras were used to intelligently deliver the robotic end-effector to relevant locations to enable grasping. To provide short distance stand-off characterisation of the waste objects, a variety of sensors were integrated onto robotic arms, including micro gamma-spectrometers, Laser Raman Spectroscopy probes and a portable X-ray Fluorescence (XRF) device. These were able to classify the material and radiometric composition of waste objects. AI algorithms were frequently employed to assist all processes.
Following a progressive development sequence, robotic integration of each individual characterisation system is demonstrated in its own chapter. From the technology development demonstrated in this thesis, it may be concluded that a fully integrated robotic system may be deployable as part of an integrated waste sorting and segregation solution which could autonomously handle, measure and determine long-term disposal routes for mixed assorted nuclear wastes.
| Date of Award | 6 Dec 2022 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Thomas Bligh Scott (Supervisor) & David Megson-Smith (Supervisor) |