Autonomous Robot Evolution (ARE): Cradle to Grave
Lead Research Organisation:
University of the West of England
Department Name: Bristol Robotics Laboratory
Abstract
Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.
People |
ORCID iD |
Alan Winfield (Principal Investigator) |
Publications
Afroogh S
(2022)
Tracing app technology: an ethical review in the COVID-19 era and directions for post-COVID-19.
in Ethics and information technology
Angus M
(2023)
Practical hardware for evolvable robots.
in Frontiers in robotics and AI
Buchanan E
(2020)
Bootstrapping Artificial Evolution to Design Robots for Autonomous Fabrication
in Robotics
Buchanan E
(2020)
Evolution of Diverse, Manufacturable Robot Body Plans
Hale M
(2020)
Hardware Design for Autonomous Robot Evolution
Jones S
(2019)
Onboard Evolution of Understandable Swarm Behaviors
in Advanced Intelligent Systems
Le Goff L
(2023)
Morpho Evolution With Learning Using a Controller Archive as an Inheritance Mechanism
in IEEE Transactions on Cognitive and Developmental Systems
Lee C
(2021)
Negative updating applied to the best-of-n problem with noisy qualities
in Swarm Intelligence
Li W
(2023)
Evaluation of Frameworks That Combine Evolution and Learning to Design Robots in Complex Morphological Spaces
in IEEE Transactions on Evolutionary Computation
Description | One of the key initial challenges was to design, build and test a system for automatically fabricating evolved robots. This system is called RoboFab, and consists of 2 robots: one is a 3D printer and the other a robot arm. Under software control the 'skeleton' of the evolved robot is fabricated in the 3D printer. Then to complete assembly of the robot the robot arm removes the skeleton from the print bed, then places it into an assembly fixture. The robot then picks and places the pre-fabricated 'organs' (controller, wheels, sensors etc) from the 'organ bank' into the skeleton in the positions determined by the robot's genome. Finally the robot arm connects wires between the controller organ and the wheel & sensor organs in order to complete the robot's 'nervous system'. The robot is then complete. We have successfully demonstrated the autonomous fabrication and assembly of complete robots, as reported in our papers: Hardware design for autonomous robot evolution (Hale et al, 2021) and Towards Autonomous Robot Evolution (Eiben et al, 2021). This was a significant milestone in the ARE project. A movie clip of the RoboFab can be found at https://www.youtube.com/watch?v=mWjZya9PJQg. In 2022 we have extended the RoboFab by successfully incorporated articulated limb (leg) organs, so that robots with legs and/or wheels can be evolved and fabricated. A more recent article by project partner Prof Emma Hart appeared in New Scientist here: https://www.newscientist.com/article/mg25333751-700-meet-the-robots-that-can-reproduce-learn-and-evolve-all-by-themselves/ |
Exploitation Route | Other researchers in Evolutionary Robotics could make use of the RoboFab design as part of their own work. |
Sectors | Aerospace, Defence and Marine,Construction,Creative Economy,Education,Energy,Environment,Manufacturing, including Industrial Biotechology |
URL | https://www.york.ac.uk/robot-lab/are/ |
Description | ARE partnership |
Organisation | Edinburgh Napier University |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | UWE's primary contribution to the Autonomous Robot Evolution (ARE) project was to lead the design and development of a machine capable of both 3D printing the skeleton for a new evolved robot, but then autonomously selecting and adding pre fabricated organs (i.e. sense organs, wheel organs, or multi segmented limbs) to assemble a complete new robot - including wiring. We call this machine the Robot Fabricator, or RobFab. RoboFab is a unique machine that we have used to demonstrate, for the first time, the automated evolution of robots in real time and real space. UWE also contributed to other aspects of ARE, led by partners, both conceptually and practically. |
Collaborator Contribution | The University of York (Prof Andy Tyrrell and team) led, project managed and coordinated the work of the ARE project, in additional to contributing conceptually and practically to the work of the other partners. Edinburgh Napier University (Prof Emma Hart and team), led on the design of both learning algorithms and the ARE software architecture and its implementation, as well as contributing conceptually and practically to the work of the other partners. The Free University of Amsterdam (Prof Guszti Eiben), led on the design of the evolutionary algorithms, as well as contributing conceptually and practically to the work of the other partners. |
Impact | The ARE project has, to date, produced 1 book chapter, 6 journal papers, 6 conference papers and 1 poster. These are listed on the web site above. The outputs that UWE co-authored are: Le Goff et al (2022) Morpho-evolution with learning using a controller archive as an inheritance mechanism Eiben et al (2021) Towards Autonomous Robot Evolution Buchanan et al (2020) Evolution of Diverse, Manufacturable Robot Body Plans Buchanan et al (2020) Bootstrapping Artificial Evolution to Design Robots for Autonomous Fabrication Hale et al (2020) Hardware Design for Autonomous Robot Evolution Le Goff et al (2020) Sample and time efficient policy learning with CMA-ES and Bayesian Optimisation Hale et al (2019) The ARE Robot Fabricator: How to (Re)produce Robots that Can Evolve in the Real World The project covers a range of engineering disciplines, including mechanical (UWE), electronics (York and UWE) and software engineering (Napier), together with the fields of evolutionary robotics (York, UWE, Napier and Amsterdam) and robot learning (Napier). |
Start Year | 2018 |
Description | ARE partnership |
Organisation | Free University of Amsterdam |
Country | Netherlands |
Sector | Academic/University |
PI Contribution | UWE's primary contribution to the Autonomous Robot Evolution (ARE) project was to lead the design and development of a machine capable of both 3D printing the skeleton for a new evolved robot, but then autonomously selecting and adding pre fabricated organs (i.e. sense organs, wheel organs, or multi segmented limbs) to assemble a complete new robot - including wiring. We call this machine the Robot Fabricator, or RobFab. RoboFab is a unique machine that we have used to demonstrate, for the first time, the automated evolution of robots in real time and real space. UWE also contributed to other aspects of ARE, led by partners, both conceptually and practically. |
Collaborator Contribution | The University of York (Prof Andy Tyrrell and team) led, project managed and coordinated the work of the ARE project, in additional to contributing conceptually and practically to the work of the other partners. Edinburgh Napier University (Prof Emma Hart and team), led on the design of both learning algorithms and the ARE software architecture and its implementation, as well as contributing conceptually and practically to the work of the other partners. The Free University of Amsterdam (Prof Guszti Eiben), led on the design of the evolutionary algorithms, as well as contributing conceptually and practically to the work of the other partners. |
Impact | The ARE project has, to date, produced 1 book chapter, 6 journal papers, 6 conference papers and 1 poster. These are listed on the web site above. The outputs that UWE co-authored are: Le Goff et al (2022) Morpho-evolution with learning using a controller archive as an inheritance mechanism Eiben et al (2021) Towards Autonomous Robot Evolution Buchanan et al (2020) Evolution of Diverse, Manufacturable Robot Body Plans Buchanan et al (2020) Bootstrapping Artificial Evolution to Design Robots for Autonomous Fabrication Hale et al (2020) Hardware Design for Autonomous Robot Evolution Le Goff et al (2020) Sample and time efficient policy learning with CMA-ES and Bayesian Optimisation Hale et al (2019) The ARE Robot Fabricator: How to (Re)produce Robots that Can Evolve in the Real World The project covers a range of engineering disciplines, including mechanical (UWE), electronics (York and UWE) and software engineering (Napier), together with the fields of evolutionary robotics (York, UWE, Napier and Amsterdam) and robot learning (Napier). |
Start Year | 2018 |
Description | ARE partnership |
Organisation | University of York |
Department | Department of Electronics |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | UWE's primary contribution to the Autonomous Robot Evolution (ARE) project was to lead the design and development of a machine capable of both 3D printing the skeleton for a new evolved robot, but then autonomously selecting and adding pre fabricated organs (i.e. sense organs, wheel organs, or multi segmented limbs) to assemble a complete new robot - including wiring. We call this machine the Robot Fabricator, or RobFab. RoboFab is a unique machine that we have used to demonstrate, for the first time, the automated evolution of robots in real time and real space. UWE also contributed to other aspects of ARE, led by partners, both conceptually and practically. |
Collaborator Contribution | The University of York (Prof Andy Tyrrell and team) led, project managed and coordinated the work of the ARE project, in additional to contributing conceptually and practically to the work of the other partners. Edinburgh Napier University (Prof Emma Hart and team), led on the design of both learning algorithms and the ARE software architecture and its implementation, as well as contributing conceptually and practically to the work of the other partners. The Free University of Amsterdam (Prof Guszti Eiben), led on the design of the evolutionary algorithms, as well as contributing conceptually and practically to the work of the other partners. |
Impact | The ARE project has, to date, produced 1 book chapter, 6 journal papers, 6 conference papers and 1 poster. These are listed on the web site above. The outputs that UWE co-authored are: Le Goff et al (2022) Morpho-evolution with learning using a controller archive as an inheritance mechanism Eiben et al (2021) Towards Autonomous Robot Evolution Buchanan et al (2020) Evolution of Diverse, Manufacturable Robot Body Plans Buchanan et al (2020) Bootstrapping Artificial Evolution to Design Robots for Autonomous Fabrication Hale et al (2020) Hardware Design for Autonomous Robot Evolution Le Goff et al (2020) Sample and time efficient policy learning with CMA-ES and Bayesian Optimisation Hale et al (2019) The ARE Robot Fabricator: How to (Re)produce Robots that Can Evolve in the Real World The project covers a range of engineering disciplines, including mechanical (UWE), electronics (York and UWE) and software engineering (Napier), together with the fields of evolutionary robotics (York, UWE, Napier and Amsterdam) and robot learning (Napier). |
Start Year | 2018 |
Description | Robotics for Nuclear Environments (RNE) collaboration |
Organisation | University of Manchester |
Department | School of Electrical and Electronic Engineering |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | UWE is also a partner in the EPSRC funded project Robotics for Nuclear Environments (RNE), led by the University of Manchester, and we were able to introduce Manchester and the RNE project to ARE. |
Collaborator Contribution | The RNE project, led by Prof Barry Lennox at the University of Manchester, was able to bring examples of real-world extreme environments that might be strong candidates for the evolutionary robotics approach envisioned by the ARE project. Prof Lennox also served on the advisory board for ARE. |
Impact | One outcome was a deeper understanding within the ARE team of the severe challenges faced by robots in extreme environments. |
Start Year | 2018 |
Description | Article in EE news, July 2018 |
Form Of Engagement Activity | A magazine, newsletter or online publication |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Professional Practitioners |
Results and Impact | Journalist Nick Flaherty contacted Alan Winfield about the ARE project, resulting in the short online article entitled Researchers use 3D printing for autonomous robot evolution, published in July 2018. |
Year(s) Of Engagement Activity | 2018 |
URL | https://www.eenewseurope.com/news/researchers-use-3d-printing-autonomous-robot-evolution |
Description | Article in Technology Review, May 2019 |
Form Of Engagement Activity | A magazine, newsletter or online publication |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Industry/Business |
Results and Impact | Based on interviews with project investigators Emma Hart and Gusz Eiben then article Darwin's Machines appeared in Technology Review (in German), in May 2019. |
Year(s) Of Engagement Activity | 2019 |
URL | https://www.heise.de/tr/artikel/Darwins-Maschinen-4404076.html |
Description | Blog post reporting on this project (recurring) |
Form Of Engagement Activity | Engagement focused website, blog or social media channel |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Public/other audiences |
Results and Impact | Examples include blog posts resulting in (1) contact by journalists, seeking interviews and (2) contact by school children asking questions for term projects. |
Year(s) Of Engagement Activity | 2018,2019,2020 |
URL | https://alanwinfield.blogspot.com/search/label/Autonomous%20Robot%20Evolution |
Description | Invited public talk by Alan Winfield - Devonshire Association President's Symposium |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Public/other audiences |
Results and Impact | Prof. Alan Winfield gave a talk on "A long-term vision for truly sustainable Robotics and Artificial Intelligence", at Devonshire Association President's Symposium, Taunton, 14 May 2022 |
Year(s) Of Engagement Activity | 2022 |
URL | https://devonassoc.org.uk/presidents-symposium-may-2022/ |
Description | Invited talk for Bristol Model Engineering Society, Bristol February 2020 |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | Local |
Primary Audience | Public/other audiences |
Results and Impact | Alan Winfield gave an invited talk to the Bristol Model Engineers Society (founded 1909) on 19 February 2020, entitled: Robot bodies and how to evolve them. The talk outlined the work of the Autonomous Robot Evolution project. |
Year(s) Of Engagement Activity | 2020 |
Description | Keynote lecture for the York Festival of Ideas |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | Regional |
Primary Audience | Public/other audiences |
Results and Impact | Alan Winfield was invited to give the keynote lecture for the theme 'Artificial Intelligence: Promises and Perils' hosted by the University of York as part of the York Festival of Ideas. The keynote was entitled What is Artificial Intelligence? and was chaired by Sir Malcolm Grant. Autonomous Robot Evolution was one of the projects used to illustrate the talk. |
Year(s) Of Engagement Activity | 2018 |
URL | http://yorkfestivalofideas.com/2018/talks/what-is-ai/ |
Description | Talk at Robosoft: Software Engineering for Robotics, RAEng, London, November 2019 |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Professional Practitioners |
Results and Impact | Alan Winfield gave an invited talk entitled 'Ethical Standards in robotics and AI: Responsible Robotics' at the 2 day meeting RoboSoft: Software Engineering for Robotics, organised by the University of York and hosted at the Royal Academy of Engineering. The talk presented RoboTIPS and the ethical black box; Autonomous Vehicles were also used as an illustrative example. |
Year(s) Of Engagement Activity | 2019 |
URL | https://www.cs.york.ac.uk/robostar/robosoft/ |