"This project will deliver a high TRL system to address Sellafield’s stated objectives 1-5 in the call: 1. Optimise activities before entering area: robot will first enter and autonomously build an ""as-built"" 3D model the work-zone, with annotated contaminant/dose data. This model will enable enhanced planning pre human entry. 2. Protect the operator: co-bot bears the weight of heavy tools, while being directly guided by the operator (grasping a teach-handle) in force-amplification mode (avoiding back/physical injuries and heat stress), or in semi-autonomous remote operation mode (human keeps away from higher dose regions and shine paths while supervising robot). 3. Better real time information: the 3D model, generated from initial robot entry, is used to overlay radiation/hazard data on the operator’s Augmented Reality headset display. Real-time model updates, as cell demolition progresses, will facilitate task monitoring. Note that AR headsets (commonly demoed in labs) *cannot work off-the-shelf* for free movement of operator within a large zone, requiring advanced methods. 4. Reduce waste. Each operator will accomplish more per shift, leading to reduced secondary waste (contaminated suits). The co-bot will also facilitate cutting and size reduction, thereby maximising waste per container. 5. Support teams. Heavier tasks done by fewer people. The sensored robot could also assist with scanning operators for contamination before exiting work-zone."