Self-learning robotics for industrial contact-rich tasks (ATARI): enabling smart learning in automated disassembly
Lead Research Organisation:
UNIVERSITY OF BIRMINGHAM
Department Name: Mechanical Engineering
Abstract
Disassembly is an essential operation in many industrial activities including repair, remanufacturing and recycling. Disassembly tends to be manually carried out - it is labour intensive and usually inefficient.
Disassembly requires high-level dexterity in manipulations and thereby can be more difficult to robotise in comparison to the tasks that have no physical contacts (e.g. computer visual inspection) or simple contacts (e.g. cutting, welding, pick-and-place). Robotic disassembly has the potential to improve the productivity of repair, remanufacturing, recycling, all of which have been recognised as key components of a more circular economy.
The existing procedure and state-of-the-art techniques for disassembly automation usually require a comprehensive analysis of a disassembly task, correct design of sensing and compliance facilities, efficient task plans, and a reliable system integration. It is usually a complex, expensive and time-consuming process to implement a robotic disassembly system.
This project will develop a self-learning mechanism to allow robots to learn disassembly tasks and the respective control strategies autonomously, by combining multidimensional sensing and machine learning techniques. This capability will help build a more plug-and-play disassembly automation system, and reduce the technical difficulties and the implementation costs of disassembly automation.
It is expected the next generation industrial robotics can be adopted in more complex and uncertain tasks such as maintenance, cleaning, repair, remanufacturing and recycling, where many processes are contact-rich. Disassembly is a typical contact-rich task. The Principal Investigator envisages that self-learning robotic disassembly will provide key understandings and technologies that can be adopted to the automation of other types of contact-rich tasks in the future to encourage a wider adoption of robots in the UK industry.
Disassembly requires high-level dexterity in manipulations and thereby can be more difficult to robotise in comparison to the tasks that have no physical contacts (e.g. computer visual inspection) or simple contacts (e.g. cutting, welding, pick-and-place). Robotic disassembly has the potential to improve the productivity of repair, remanufacturing, recycling, all of which have been recognised as key components of a more circular economy.
The existing procedure and state-of-the-art techniques for disassembly automation usually require a comprehensive analysis of a disassembly task, correct design of sensing and compliance facilities, efficient task plans, and a reliable system integration. It is usually a complex, expensive and time-consuming process to implement a robotic disassembly system.
This project will develop a self-learning mechanism to allow robots to learn disassembly tasks and the respective control strategies autonomously, by combining multidimensional sensing and machine learning techniques. This capability will help build a more plug-and-play disassembly automation system, and reduce the technical difficulties and the implementation costs of disassembly automation.
It is expected the next generation industrial robotics can be adopted in more complex and uncertain tasks such as maintenance, cleaning, repair, remanufacturing and recycling, where many processes are contact-rich. Disassembly is a typical contact-rich task. The Principal Investigator envisages that self-learning robotic disassembly will provide key understandings and technologies that can be adopted to the automation of other types of contact-rich tasks in the future to encourage a wider adoption of robots in the UK industry.
Organisations
- UNIVERSITY OF BIRMINGHAM (Lead Research Organisation)
- Ocado Technology (Collaboration)
- University of Castile-La Mancha (Collaboration)
- Cranfield University (Collaboration)
- University of Sheffield (Collaboration)
- Toshiba (Collaboration)
- Ford Motor Company (Collaboration)
- Airbus Group (Collaboration)
- Satellite Applications Catapult (Collaboration)
- Dyson (Collaboration)
- Royal Institute of Technology (Collaboration)
- International Telecommunication Union (Collaboration)
- Caterpillar Inc. (Collaboration)
- University of Oklahoma (Collaboration)
- University of Southern California (Collaboration)
- Karlsruhe Institute of Technology (Collaboration)
- United Nations Educational, Scientific and Cultural Organization (Collaboration)
- Ecobat Technologies (Collaboration)
- Boston Dynamics (Collaboration)
- Beihang University (Project Partner)
- KUKA Robotics UK Limited (Project Partner)
- KEYENCE (UK) Ltd (Project Partner)
- Wuhan University of Technology (Project Partner)
- Manufacturing Technology Centre (United Kingdom) (Project Partner)
Publications
Deng W
(2024)
Learning by doing: A dual-loop implementation architecture of deep active learning and human-machine collaboration for smart robot vision
in Robotics and Computer-Integrated Manufacturing
Deng W
(2025)
Learning from new products: A robust end-of-life object detection model for robotic disassembly using the dual constraints of anchors and corners
in Journal of Cleaner Production
Deng W
(2024)
Predictive exposure control for vision-based robotic disassembly using deep learning and predictive learning
in Robotics and Computer-Integrated Manufacturing
Goli F
(2024)
Characterizing the mechanics of rectangular peg-hole disassembly and the effect of the active compliance centre on the extraction force.
in Royal Society open science
Goli F
(2024)
Jamming problems and the effects of compliance in dual peg-hole disassembly
in Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences
| Description | The ATARI project fundamentally advanced the integration of machine learning and multidimensional sensing to automate complex, contact-rich disassembly tasks. The core scientific and technological breakthroughs include: Disassembly Mechanics & Compliance: The project was the first to formalise the characteristics of disassembly mechanics, proving that mechanical compliance-a six-dimensional property of motion-critically dictates contact conditions. The team demonstrated that establishing an accurate remote compliance center drastically reduces the likelihood of "jamming" during the disassembly process. Programmable Material Structure (FMHE): Building on the mechanics research, the team developed FMHE, a novel material structure featuring self-tunable conductivity and stiffness. This resulted in a first-of-its-kind mechanical device capable of programmable mechanical compliance. Autonomous Skill Acquisition: ATARI pioneered a world-first mechanism that allows machines to learn disassembly skills autonomously. This system requires zero human instruction or pre-programming and has demonstrated the capability for seamless skill transfer between different robotic platforms. EV Battery Disassembly Prototype: The team designed, built, and evaluated an easily reconfigurable robotic cell specifically for the safe, cost-effective module-to-cell disassembly of electric vehicle lithium-ion batteries (EV-LiB), utilising a mixture of destructive and non-destructive extraction operations. |
| Exploitation Route | To ensure the transition of these scientific discoveries into real-world applications, the following exploitation and commercialisation strategies are actively underway: Industrial Piloting: Active industrial pilot programs are progressing in collaboration with major partners, including Airbus, ECOBAT, and Gomes Technology. Furthermore, the team is exploring the use of legged robots for disassembly tasks alongside Boston Dynamics. Commercialisation & Intellectual Property: A patent covering the project's core technologies is currently in preparation, and active negotiations with investors are ongoing to launch a dedicated spin-out company. Follow-On Innovation Pipeline: The foundational technologies developed in ATARI are being directly scaled through substantial follow-on funding. This includes an EPSRC Impact Acceleration Account (IAA) to develop modular disassembly systems, alongside multi-million-pound EPSRC investments (such as the STAMAN and RoboTriage projects) to further industrialise robotic skill transfer and circular economy value retention. |
| Sectors | Aerospace Defence and Marine Digital/Communication/Information Technologies (including Software) Electronics Environment Manufacturing including Industrial Biotechology |
| Description | ATARI has delivered profound academic, societal, and economic value, significantly accelerating the UK's capabilities in the circular economy: Policy & Global Sustainability: Project outcomes were formally reviewed by the International Telecommunication Union (ITU) and incorporated into an official ITU report. This report now serves as United Nations guidance on utilising AI to support the circular economy. Scientific Dissemination: The project generated over 40 academic publications across top-tier platforms (including Science, The Royal Society, and IEEE) and established 16 new strategic partnerships with world-leading technology and engineering firms. Economic & Environmental Transformation: By lowering the technical barrier and cost of implementing contact-rich automation, ATARI enables "plug-and-play" disassembly. This directly tackles the massive bottleneck in end-of-life product processing (such as e-waste and EV batteries), making mass recycling, remanufacturing, and repair financially viable and highly scalable for the UK manufacturing sector. |
| Sector | Aerospace, Defence and Marine,Digital/Communication/Information Technologies (including Software),Electronics,Healthcare,Manufacturing, including Industrial Biotechology |
| Impact Types | Economic Policy & public services |
| Description | United Nations Guidelines on the Implementation of Eco-friendly Criteria for AI and other Emerging Technologies |
| Geographic Reach | Multiple continents/international |
| Policy Influence Type | Implementation circular/rapid advice/letter to e.g. Ministry of Health |
| Description | EPSRC IAA (2022-25): Affordable And Modular Robotic Disassembly Systems |
| Amount | £50,000 (GBP) |
| Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
| Sector | Public |
| Country | United Kingdom |
| Start | 03/2023 |
| End | 12/2023 |
| Description | EPSRC Manufacturing Research Hub in Robotics, Automation & Smart Machine Enabled Sustainable Circular Manufacturing & Materials (RESCu-M2) |
| Amount | £11,839,509 (GBP) |
| Funding ID | EP/Z532873/1 |
| Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
| Sector | Public |
| Country | United Kingdom |
| Start | 08/2024 |
| End | 09/2031 |
| Description | MTC-UoB PhD Scholarship |
| Amount | £100,000 (GBP) |
| Organisation | Manufacturing Technology Centre (MTC) |
| Sector | Private |
| Country | United Kingdom |
| Start | 09/2024 |
| End | 05/2028 |
| Description | Project 2766306: Electric Motor Disassembly - A FEMM Hub Feasibility Study |
| Amount | £75,000 (GBP) |
| Organisation | University of Sheffield |
| Sector | Academic/University |
| Country | United Kingdom |
| Start | 04/2024 |
| End | 04/2025 |
| Description | Robotic Triage for Value Retention in a Circular Economy (RoboTriage) |
| Amount | £1,510,665 (GBP) |
| Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
| Sector | Public |
| Country | United Kingdom |
| Start | 02/2025 |
| End | 01/2028 |
| Description | Robotic skill transfer and augmentation for contact-rich tasks in manufacturing (STAMAN) |
| Amount | £1,035,396 (GBP) |
| Funding ID | EP/Y02270X/1 |
| Organisation | Engineering and Physical Sciences Research Council (EPSRC) |
| Sector | Public |
| Country | United Kingdom |
| Start | 07/2024 |
| End | 08/2027 |
| Description | Boston Dynamics |
| Organisation | Boston Dynamics |
| Country | United States |
| Sector | Private |
| PI Contribution | The ATARI team is collaborating with Boston Dynamics on robotic disassembly techniques for using legged robots to perform disassembly. |
| Collaborator Contribution | The ATARI team is collaborating with Boston Dynamics on robotic disassembly techniques for using legged robots to perform disassembly. A support letter was obtained in July 2022. |
| Impact | Exchange of information and joint new research proposal submitted. |
| Start Year | 2022 |
| Description | Collaboration with KIT |
| Organisation | Karlsruhe Institute of Technology |
| Country | Germany |
| Sector | Academic/University |
| PI Contribution | A joint research initiative that has led to a £6.5M MSAC Doctoral Network proposal |
| Collaborator Contribution | A joint research initiative that has led to a £6.5M MSAC Doctoral Network proposal |
| Impact | A joint research initiative that has led to a £6.5M MSAC Doctoral Network proposal |
| Start Year | 2024 |
| Description | Collaboration with KTH |
| Organisation | Royal Institute of Technology |
| Country | Sweden |
| Sector | Academic/University |
| PI Contribution | A joint research initiative that has led to a £6.5M MSAC Doctoral Network proposal |
| Collaborator Contribution | A joint research initiative that has led to a £6.5M MSAC Doctoral Network proposal |
| Impact | A joint research initiative that has led to a £6.5M MSAC Doctoral Network proposal |
| Start Year | 2024 |
| Description | Collaboration with the Univeristy of Southern California |
| Organisation | University of Southern California |
| Country | United States |
| Sector | Academic/University |
| PI Contribution | A new research initiative has led to a joint proposal to the NSF and UKRI |
| Collaborator Contribution | A new research initiative has led to a joint proposal to the NSF and UKRI |
| Impact | A new research initiative has led to a joint proposal to the NSF and UKRI |
| Start Year | 2024 |
| Description | Collaboration with the University of Oklahoma |
| Organisation | University of Oklahoma |
| Country | United States |
| Sector | Academic/University |
| PI Contribution | A new research initiative has led to a joint proposal to the NSF and UKRI |
| Collaborator Contribution | A new research initiative has led to a joint proposal to the NSF and UKRI |
| Impact | A new research initiative has led to a joint proposal to the NSF and UKRI |
| Start Year | 2024 |
| Description | Ford |
| Organisation | Ford Motor Company |
| Country | United States |
| Sector | Private |
| PI Contribution | New collaboration on robotic triage for value retention in the circular economy |
| Collaborator Contribution | New collaboration on robotic triage for value retention in the circular economy. Support letter obtained in January 2024. |
| Impact | Exchange of information and joint new research proposal submitted. |
| Start Year | 2024 |
| Description | ITU |
| Organisation | International Telecommunication Union |
| Country | Senegal |
| Sector | Learned Society |
| PI Contribution | New collaboration on robotic triage for value retention in the circular economy |
| Collaborator Contribution | New collaboration on robotic triage for value retention in the circular economy. Support letter obtained in January 2024. |
| Impact | Joint report published; Exchange of information and joint new research proposal submitted. |
| Start Year | 2022 |
| Description | New partnership with Caterpillar |
| Organisation | Caterpillar Inc. |
| Country | United States |
| Sector | Academic/University |
| PI Contribution | The ATARI team is collaborating with Caterpillar on robotic disassembly techniques for battery and engine remanufacturing for the defense sector. |
| Collaborator Contribution | The ATARI team is collaborating with Caterpillar on robotic disassembly techniques for battery and engine remanufacturing for the defense sector. Suppor letter sent in Jan 2023. |
| Impact | Exchange of information and joint new research proposal submitted. |
| Start Year | 2023 |
| Description | Ocado |
| Organisation | Ocado Technology |
| Country | United Kingdom |
| Sector | Private |
| PI Contribution | The ATARI team is collaborating with Ocado on robotic disassembly techniques for EoL assessment technologies. |
| Collaborator Contribution | The ATARI team is collaborating with Ocado on robotic disassembly techniques for EoL assessment technologies. Support letter obtained in January 2024. |
| Impact | Exchange of information and joint new research proposal submitted. |
| Start Year | 2024 |
| Description | Research collabration with Airbus |
| Organisation | Airbus Group |
| Country | France |
| Sector | Academic/University |
| PI Contribution | Regular talks with the collaborator to share research activity information and ideas |
| Collaborator Contribution | Members of Airbus visit the UoB team to contribute to the project |
| Impact | We are joining forces in research activities and the preparation of HORIZON proposals. |
| Start Year | 2022 |
| Description | Research collabration with Cranfield Univeristy |
| Organisation | Cranfield University |
| Country | United Kingdom |
| Sector | Academic/University |
| PI Contribution | A new research area was identified and funded by a £1.8M EPSRC project |
| Collaborator Contribution | A new research area was identified and funded by a £1.8M EPSRC project |
| Impact | £1.8M EPSRC project RoboTriage |
| Start Year | 2023 |
| Description | Research collabration with Dyson |
| Organisation | Dyson |
| Country | United Kingdom |
| Sector | Private |
| PI Contribution | Regular talks with the collaborator to share research activity information and ideas |
| Collaborator Contribution | Members of Dyson visit the UoB team to contribute to the project |
| Impact | We are joining forces in research activities and the preparation of new research proposals. |
| Start Year | 2022 |
| Description | Research collabration with ECOBAT |
| Organisation | Ecobat Technologies |
| Country | Germany |
| Sector | Private |
| PI Contribution | Regular talks with the collaborator to share research activity information and ideas |
| Collaborator Contribution | ECOBAT has agreed to host onsite tests for ATARI developments |
| Impact | We are preparing for an onsite test of ATARI technologies |
| Start Year | 2022 |
| Description | Research collabration with Sheffield University |
| Organisation | University of Sheffield |
| Department | Sheffield Biorepository |
| Country | United Kingdom |
| Sector | Academic/University |
| PI Contribution | Regular talks with the collaborator to share research activity information and ideas |
| Collaborator Contribution | Members of Sheffield University visit the UoB team to contribute to the project |
| Impact | We are joining forces in research activities and the preparation of new research proposals. |
| Start Year | 2022 |
| Description | Research collabration with University of Castilla-La Mancha - UCLM |
| Organisation | University of Castile-La Mancha |
| Country | Spain |
| Sector | Academic/University |
| PI Contribution | Regular talks with the collaborator to share research activity information and ideas |
| Collaborator Contribution | Members of Universidad de Castilla-La Mancha visit the UoB team to contribute to the project |
| Impact | We are joining forces in research activities and the preparation of HORIZON proposals. |
| Start Year | 2022 |
| Description | Satellite Applications Catapult |
| Organisation | Satellite Applications Catapult |
| Country | United Kingdom |
| Sector | Charity/Non Profit |
| PI Contribution | New collaborations on robotic disassembly techniques for space manipulation and satellite maintenance. |
| Collaborator Contribution | New collaborations on robotic disassembly techniques for space manipulation and satellite maintenance. A support letter was obtained in August 2022. |
| Impact | Exchange of information and joint new research proposal submitted. |
| Start Year | 2022 |
| Description | TOSHIBA |
| Organisation | Toshiba |
| Country | Japan |
| Sector | Private |
| PI Contribution | New collaboration on robotic triage for value retention in the circular economy |
| Collaborator Contribution | New collaboration on robotic triage for value retention in the circular economy. Support letter obtained in January 2024. |
| Impact | Exchange of information and joint new research proposal submitted. |
| Start Year | 2023 |
| Description | UNESCO |
| Organisation | United Nations Educational, Scientific and Cultural Organization |
| Country | France |
| Sector | Academic/University |
| PI Contribution | New collaboration on robotic triage for value retention in the circular economy |
| Collaborator Contribution | New collaboration on robotic triage for value retention in the circular economy. Support letter obtained in January 2024. |
| Impact | Exchange of information and joint new research proposal submitted. |
| Start Year | 2022 |
| Title | A CASING REMOVAL TOOL |
| Description | A tool for removal of a casing from a body comprises a first end and a second end and a casing attachment configured to attach to the first end of the casing and a body attachment configured to attach to the body. A retainer is configured to restrict movement of the casing attachment with respect to the body such that when the casing attachment and body are moved apart, the casing is peeled away from the body from the first end to the second end. The body may be an electric battery module. The retainer may be a circumferential retainer configured to cause the casing attachment to move circumferentially about a centre of rotation near to the second end. The retainer may comprise a retaining slot engageable with a retaining pin. The tool may comprise a stop to prevent complete removal of the casing from the body. |
| IP Reference | WO2025172717 |
| Protection | Patent / Patent application |
| Year Protection Granted | 2025 |
| Licensed | No |
| Description | A conference presentation at ICAC2022 - Mr Yue Zang |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Schools |
| Results and Impact | A conference presentation at ICAC2022 - Mr Yue Zang |
| Year(s) Of Engagement Activity | 2022 |
| Description | A conference presentation at ICAC2022 - Ms Farzaneh Goli |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Schools |
| Results and Impact | A conference presentation at ICAC2022 - Ms Farzaneh Goli |
| Year(s) Of Engagement Activity | 2022 |
| Description | An invited talk at ICAC2022 - Dr Yongjing Wang |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Schools |
| Results and Impact | Dr Yongjing Wang has been invited to give a talk at ICAC2022 |
| Year(s) Of Engagement Activity | 2022 |
| Description | An invited talk at IEEE International Conference on Universal Village (UV) |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Schools |
| Results and Impact | Dr Yongjing Wang has been invited to give a talk at IEEE International Conference on Universal Village (UV), Boston, US. |
| Year(s) Of Engagement Activity | 2022 |
| Description | Invited talk at Donghua University |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Schools |
| Results and Impact | Invited talk at an workshop at Donghua University, China |
| Year(s) Of Engagement Activity | 2023 |
| Description | Invited talk at Imperial college |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Schools |
| Results and Impact | Invited talk at an international workshop involving researchers from the UK, Netherlands, and China. |
| Year(s) Of Engagement Activity | 2023 |
| Description | Invited talk at Virginia Tech |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Schools |
| Results and Impact | Invited talk at an international workshop involving researchers from Virginia Tech, Mississippi State, Auburn, and NC State |
| Year(s) Of Engagement Activity | 2023 |
| Description | Participation in an activity, workshop or similar - Special session at 2024 IEEE ICIT |
| Form Of Engagement Activity | A formal working group, expert panel or dialogue |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Schools |
| Results and Impact | We are hosting a special session about smart remanufacturing at the 2024 IEEE ICIT. ICIT is the flagship automation conference of the IEEE IE society and constitutes the primary forum for cross-industry and multidisciplinary research in automation. |
| Year(s) Of Engagement Activity | 2024 |
| Description | Presentation at CASE 2024 |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Professional Practitioners |
| Results and Impact | Presentation at IEEE CASE 2024 |
| Year(s) Of Engagement Activity | 2024 |
| Description | Presentation at DigiTwin 2024 |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Industry/Business |
| Results and Impact | Presentation at DigiTwin 2024 |
| Year(s) Of Engagement Activity | 2024 |
| Description | Special session at 2023 IEEE International Conference on Automation Science and Engineering (CASE) |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | International |
| Primary Audience | Professional Practitioners |
| Results and Impact | We are hosting a special session about smart remanufacturing at the 2023 IEEE International Conference on Automation Science and Engineering (CASE). CASE is the flagship automation conference of the IEEE Robotics and Automation Society and constitutes the primary forum for cross-industry and multidisciplinary research in automation. "Smart Remanufacturing Technologies" (Code: 8wf47). |
| Year(s) Of Engagement Activity | 2023 |
| URL | https://case2023.org/special-session-proposals/ |
| Description | UKRAS Early career workship |
| Form Of Engagement Activity | A formal working group, expert panel or dialogue |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Schools |
| Results and Impact | Hosting a workshop for the UKRAS network early career community. |
| Year(s) Of Engagement Activity | 2023 |