Remote Inspection

High quality data about nuclear environments is essential to accurately plan for dismantling and decommissioning, as well as monitoring and maintaining sites.

The Remote Inspection Working Group (RIWG) focuses on developing robotic and AI technology for both the characterisation of unknown nuclear environments, and change or anomaly detection of previously characterised environments. Regular testing and feedback from industrial partners is helping the working group shape their solution for remote inspection.

We are continuing to build collaborative relationships with challenge owners. In particular, we are seeking further inspection sites or challenges that could be addressed using this platform and capability. Research priorities include integrating information from 3D data sources and further sensor types, and developing tools for anomaly and change detection of previously inspected sites.

The RIWG demonstrator

The RIWG demonstrator has been constructed based on feedback from members of the working group and RAIN researchers. It is built on a Clearpath Husky 4 wheeled robotic platform. There are 3 x Hoyoku UTM-30LX LIDARs, for navigation and safety, a Skull Canyon Intel NUC computer and Red Pitaya FPGA board for signal processing. An FLIR D48 pan-tilt unit carries: a FLIR Tau 2 thermal camera with Workswell mounting unit, a SICK TIM571 LIDAR, and a detector and collimator that is interchangeable with a Kromek GR1. Hardware components can be changed according to the requirements. The RIWG has done a lot of development on the software. These include:

- ROS and Clearpath packages, for drivers, localisation, and navigation

- RViz including custom plugins, for visualisation of robot data and inspection outputs

- Robust topological navigation and task execution (from the EU STRANDS project)

- ROS radiation message type for use across a range of detectors

- ROS node to bridge Red PItaya signal processing to ROS

- Pointcloud painter software that builds a composite pointcloud from input LIDARS and combines it with outputs from RGB, thermal and radiation sensors

- Mission Control software for managing software at runtime

The Remote inspection working group corresponds with research themes 1 & 2

Working Group Lead - Nick Hawes

Theme 1: Mobility, Inspection and Operations - Ioannis Havoutis

Theme 2: Reusability and Robustness - Joaquin Carrasco

The working group builds upon the research going on across the institutions within the hub and finds industrial challenges that could be solved by applying this knowledge. The Remote Inspection Working Group (RIWG) is particularly developing tools for both characterisation and long-term monitoring of nuclear environments.

Research Challenges Theme 1:

  • 1

    Improved dynamic mobility is essential to allow legged vehicles to cross rough terrain, climb ladders etc. This requires responsive re-planning of high dimensional robot systems and current approaches struggle to marry the uncertainty of real world sensing, with the limits of high frequency feedback control
  • 2

    Robust situational awareness in low-light conditions and poor visibility. This requires the fusion of inertial, LIDAR, vision, radiation sensing and other measurements to achieve reliable and accurate navigation
  • 3

    Autonomous navigation algorithms should effectively pair with human operators. Map representations communicated to human operators should be labelled with estimates of physical properties (radiation, temperature) as well as geometric structure. Reconstructions should support change detection on a semantic level to identify subsidence and fissure.

Research Challenges Theme 2:

  • 1

    To achieve high reliability, failure modes in high gamma and neutron environments must be quantified
  • 2

    Improved rad-tolerance needs to be developed through software, low-level hardware redundancy, electronic reconfigurability and by dynamic assessment of balance between on-board and off-board system
  • 3

    Extreme environments demand high levels of mechanical system reliability. Approaches will detect and compensate for failures using redundancy and fault tolerant control systems
  • 4

    Where power is limited, minimally actuated robotic systems are necessary, with reduced electronics and power scavenging