I have written about the use of robots in the nuclear industry in prior blog posts. A number of robots were used in the aftermath of the Fukushima nuclear disaster. The primary use of these robots was to explore the areas of the ruined reactors that were too radioactive for human beings to inspect. Unfortunately, many of these robots were rapidly destroyed by the heat and radioactivity. I recently blogged about a snake robot with a laser cutting head that was used to cut out a radioactive component of a nuclear reactor that was being decommissioned. Now, some scientists are talking about the use of artificial intelligence in nuclear industry robotics.
      Machine learning (ML) is an application of artificial intelligence. ML allows computer systems to collect information and use it to automate the process of decision making. If successful, ML can permit a robot to make improvements in its actions without the need for the changes to be explicitly programmed.
      Scientists at the University of Lincoln in the U.K. just got a grant of one and a half million dollars from the U.K. Engineering and Physical Sciences Research Council to research and develop artificial intelligence ML systems which can be employed by robots to teach themselves about nuclear sites. The ML will enable the robots to learn about decommissioning, nuclear waste handling and site monitoring. The AI systems will be used mapping and navigation, vision guided grasping, manipulation and cutting. It is hoped that the new AI systems will allow the robots to adapt to unique, hostile and radioactive nuclear sites.
      One of the principle UoL researchers said, “Clean-up and decommissioning of nuclear waste is one of the biggest challenges for our generation and the next, and the predicted costs are enormous: up to £200 billion over the next 100 years. Recent disaster situations such as Fukushima have shown the crucial importance of robotics technology for monitoring and intervention, which is missing up to date, making our work even more vital.”
       The Lincoln project is being carried out by the National Center for Nuclear Robotics which is a multidisciplinary EPSRC RAI (Robotics and Artificial Intelligence) Hub led by the University of Birmingham. The Queen Mary University of London, the University of West England, University of Bristol, University of Edinburgh, and Lancaster University.
      The UoL scientists are also exploring other ways to make nuclear work safer. One of these is augmented reality. At this time, it is unclear exactly how AR could be useful. Another project focuses on the operation of a bi-manual arm that will be mounted on a mobile platform. The arm could be controlled by a human being from another location or could be controlled by the AI systems themselves which would permit the robot to work autonomously.
       It is believed that robots with self-learning ability should have a better chance of surviving and functioning in dangerous nuclear sites because they will be able to make better decisions on their own.