ICRA 2024 Workshop: Robotics for Nuclear Environments

Name: Kuniaki Kawabata

Company/Institute: Japan Atomic Energy Agency

Bio: Dr. Kawabata received Ph.D from Hosei University and was a Special Postdoctoral Researcher, The Institute of Physical and Chemical Research (RIKEN) in 1997. Then he was Research Scientist then Unit Leader at RIKEN until joining Japan Atomic Energy Agency in 2015 where he became Principal Investigator at the CLADS, JAEA in 2017.

He  is a senior member of IEEE and a member of JSME, SICE, RSJ and AESJ.

Title: JAEA’s research and development activities toward safe and effective decommissioning of the Fukushima Daiichi Nuclear Power Plant using remote control technology

Abstract: It is thought that it will take 30 to 40 years to complete the decommissioning of the Fukushima Daiichi Nuclear Power Plant, and remote-control technology will continue to be essential for decommissioning work. In this presentation, we will introduce the research and development activities that JAEA is conducting to support the safe and effective promotion of long-term decommissioning work at the Fukushima Daiichi Nuclear Power Plant. In particular, we will introduce the current status and future prospects of research and development related to basic technology for operator support during remote operation, which we have been focusing on in recent years.

Name: Wataru Sato

Company/Institute: TEPCO

Bio: Wataru Sato is an engineer of Tokyo Electric Power Company Holdings, Inc. (TEPCO) with nine years of experience working on the Fukushima Daiichi (1F) decommissioning. Recently, he Joined LongOps Project with United Kingdom Atomic Energy Authority (UKAEA) and developed “Next-Generation Digital Mock-Up (NG-DMU)”. Currently, he is responsible for deploying the system developed in LongOps project at 1F to incorporate digital technology into decommissioning, aiming to achieve a “cheaper”, faster” and “safer” decommissioning process.

Title: Research and Development of Next-Generation Digital Mock-Up system for decommissioning of Fukushima Daiichi Nuclear Power Plant

Abstract: The direct on-site work by human for long time at Fukushima Daiichi (1F) is unrealistic due to the high radiation environment. Therefore, many tasks require remote operation using robots. Tokyo Electric Power Company Holdings (TEPCO) has developed the “Next-Generation Digital Mock-Up system under LongOps project which is international research collaboration with the United Kingdom Atomic Energy Authority (UKAEA). NG-DMU makes it possible to operator assistance, training and task verification within digital environment. This system will provide more faster, cheaper, safer decommissioning. In this workshop, I will talk 1F overview, development details of LongOps and the direction of future application at 1F.

Name: Shu Shirai

Company/Institute: TEPCO

Bio: Shu Shirai is a researcher of Tokyo Electric Power Company Holdings, Inc. (TEPCO) R&D Department. He has been focusing specifically on the research and development for the decommissioning of the Fukushima Daiichi nuclear power station(1F). Recently, he joined LongOps Project with UK Atomic Energy Authority (UKAEA) and contributed to the development of “Next-Generation Digital Mock-Up (NG-DMU)”. He is currently responsible for the project to deploy the NG-DMU developed in the LongOps project at the 1F. This project is the first step in utilizing NG-DMU to enhance safety in future decommissioning operations.

Title: Research and Development of Next-Generation Digital Mock-Up system for decommissioning of Fukushima Daiichi Nuclear Power Plant

Abstract: The direct on-site work by human for long time at Fukushima Daiichi (1F) is unrealistic due to the high radiation environment. Therefore, many tasks require remote operation using robots. Tokyo Electric Power Company Holdings (TEPCO) has developed the “Next-Generation Digital Mock-Up system under LongOps project which is international research collaboration with the United Kingdom Atomic Energy Authority (UKAEA). NG-DMU makes it possible to operator assistance, training and task verification within digital environment. This system will provide more faster, cheaper, safer decommissioning. In this workshop, I will talk 1F overview, development details of LongOps and the direction of future application at 1F.

Name: Sungmoon Joo

Bio: Dr. Sungmoon Joo is a principal researcher within the Nuclear and Extreme Robotics Team at the Korea Atomic Energy Research Institute (KAERI), where his research focus lies in the development of robotic systems tailored for nuclear applications. He obtained his Ph.D. in aerospace engineering from Stanford University, specializing in space robotics.

Before joining KAERI, Dr. Joo held roles as a research scientist II in the School of Interactive Computing at the Georgia Institute of Technology in US, and served as a principal research engineer in the Mechatronics Center at Samsung Heavy Industries.

Title: Robotic Dismantling for Nuclear Decommissioning

Abstract: Over the past four decades, the Korea Atomic Energy Research Institute (KAERI) has been at the forefront of research and development in robotics and remote systems technology for nuclear applications in South Korea. As the South Korean government mandates the decommissioning of nuclear power plants reaching the end of their design lifespan, the demonstration of decommissioning technologies for field deployment have become critical missions for the Korean nuclear industry.

As part of the mission, KAERI has recently focused on advanced remote dismantling technologies, incorporating high-precision heavy-duty manipulators, underwater laser cutting and scanning technologies to seamlessly integrate physical and virtual robotic operations. This presentation provides a general overview of KAERI’s robotic dismantling system, highlighting recent advancements.

Name: Rob Buckingham 

Company/Institute: UK Atomic Energy Authority

Bio: Professor Rob Buckingham OBE FREng is an Executive Director of the UK Atomic Energy Authority (UKAEA). He is responsible for RACE, the UKAEA’s centre for Remote Applications in Challenging Environments; RAICo, the Robotics and Artificial Intelligence Collaboration with the Nuclear Decommissioning Authority and the University of Manchester; and the JET Decommissioning and Repurposing Programme. Rob was lead author of the UK’s Robotics and Autonomous Systems 2020 Strategy and is a member of the UK Robotics Growth Partnership. Before re-joining the UKAEA, Rob co-founded and was Managing Director of OC Robotics which developed and commercialised snake-arm™ robots. He is a Fellow of the Royal Academy of Engineering and a Fellow of the Institute of Engineering and Technology. He received an OBE for services to robotics engineering in the 2021 New Year Honours.

Name: Dohee Lee

Company/Institute: Korea Institute of Fusion Energy

Bio: Dohee Lee is a senior researcher at the Korean Institute of Fusion Energy (KFE). He received a Ph.D. degree in computer science from Ulsan National Institute of Science and Technology (UNIST). His research interest is applying AI to the robotics system. His current research field is remote maintenance systems for fusion reactors, including developing maintenance strategies & robotic systems.

Title: A remote maintenance for Korean fusion demonstration reactor (K-DEMO): Challenges, Progress, and Collaboration

Abstract: This presentation delves into the development of remote maintenance systems for the Korean Fusion Demonstration Reactor (K-DEMO), highlighting the critical role of robotic technologies in overcoming the operational challenges of fusion reactors. It starts with a concise overview of K-DEMO, stressing the imperative for remote maintenance within its high-radiation environment. Next, the focus shifts to current research in robotic systems for K-DEMO maintenance tasks. Furthermore, the presentation highlights the significance of global cooperation and introduces current international collaborative research for maintenance tasks. This approach provides guide values to accelerate K-DEMO design by addressing remote maintenance systems.

Name: Marie Farrell

Company/Institute: The University of Manchester

Bio: Marie Farrell is a Royal Academy of Engineering Research Fellow at the University of Manchester. Her work aims to devise new ways of describing, analysing and assuring the autonomous behaviour of robotic systems. In her current and previous work, she uses and combines formal methods to reason about and provide certification evidence for robotic and aerospace systems that are to be deployed in hazardous environments. She is secretary of the working group developing the IEEE P7009 Standard on Fail-Safe Design of Autonomous and Semi-Autonomous Systems.

Title: Formalising Requirements for Autonomous Robotic Systems

Abstract: Formal methods, combined with testing and simulation-based approaches to verification, all play a role in analysing the robustness of autonomous robotic systems. But these techniques cannot be used effectively unless the requirements of the system are clearly and unambiguously defined. In fact, requirements specification for autonomous systems is an especially difficult part of the development process. This talk will provide an overview of our recent work on examining requirements for autonomous robotic systems using NASA’s Formal Requirements Elicitation Tool (FRET).

Name: Richard Mills

Company/Institute: Kongsberg Discovery

Bio: Richard joined the underwater robotics industry in 2008. His primary focus is the adoption of autonomous systems across government, defence, commercial and scientific users. He joined Kongsberg in 2012, leading a team and contributing to the development of more capable robotic systems for underwater exploration.

Prior to joining the subsea industry, Richard spent 17 years flying various types of military helicopters in different roles. Several of these helicopters had autopilot and autonomous capabilities that fostered his interest in autonomy. He has lived in the UK, Middle East, Canada, Norway and now resides in the USA.

Title: Building Trust in Autonomy: A Tale from the Deep

Abstract: The advent of AUVs capable of spending weeks at sea requires a different approach to autonomy. Goal-based mission plans do not plot lines on a chart, rather the operator sets an objective and defines an operating area. The system then compares the predicted performance of the sensors on-board with a world model to determine the most efficient survey pattern.  Translating this to in-mission performance-based autonomy takes that plan a step further by introducing a measure of real-time performance and comparing it against not only the predictive world model, but also environmental conditions measured during the mission. From this comparison, mission changes can occur on the fly.  The in-mission decisions of how best to proceed may well be made in isolation without operator input or awareness. That means the level of trust required to field these systems is different, with operators, supervisors and even investors needing to build an understanding of how they work, and what the impact of performance-based autonomy may be. Trust can be built through repetitive deployments, predictive modelling and statistical or probabilistic assessment of the system, environment and mission requirements.  A combination of these approaches can assist with the safe implementation of long range AUVs and increase the chances of getting it back with actionable information on board. This presentation will outline some of the capabilities, technical challenges and operational practicalities of fielding complex autonomous systems for subsea operations. It will also discuss the process of building trust in complex autonomous systems operating beyond communications range.

Name: Phill Mulvana

Company/Institute: UK Atomic Energy Authority

Bio: Phill is the Safety Manager and Principal Technologist for RAICo, the Robotics and AI Collaboration, part of the UK Atomic Energy Authority. Specialising in the safety and regulation of innovative technologies, he now works with industry and regulators on nuclear and robotics challenges, having previously spent time in defence, advanced manufacturing and energy systems. He holds a MSc in Systems Safety Engineering, is a Chartered Manager and a Fellow of the IET.

Title: Use of AI in robotics in the nuclear space – targeting a new paradigm in safety cases

Abstract: Nuclear decommissioning poses many unique challenges, historically these have been addressed by highly competent individuals, working methodically under strict controls designed to minimise risk. Even for small items however, the decommissioning process can be complex, hazardous, and time-consuming. Skilled operators undergo extensive training to work in physically demanding environments, creating a bottleneck in the workforce and prolonging the processing time for an expanding inventory of nuclear materials. Our project aims to develop a strawman safety case for a Robotic Glovebox with AI capability, designed to assist in the preparation and processing of nuclear material. This innovative solution has the potential to enable safer, more efficient round-the-clock decommissioning operations. By leveraging the combined expertise of UKAEA and RAICo in utilizing AI datasets, machines, and operational experience, we can speculate on the functionality of a fully operational AI glovebox system when integrated. Our safety case applies the SACE approach developed by the Assuring Autonomy International Programme to generate confidence in our initial AI glovebox design. Inputting this into the regulatory innovation sandbox of the Office for Nuclear Regulation (ONR) we hope to establish a new paradigm in safety cases for autonomous systems in nuclear environments.