{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/robots-that-can-adapt-like-animals","title":"Robots that can adapt like animals","arxiv_id":"1407.3501","date":"2014-07-13","proceeding":null,"authors":["Antoine Cully","Jeff Clune","Danesh Tarapore","Jean-Baptiste Mouret"],"abstract":"As robots leave the controlled environments of factories to autonomously\nfunction in more complex, natural environments, they will have to respond to\nthe inevitable fact that they will become damaged. However, while animals can\nquickly adapt to a wide variety of injuries, current robots cannot \"think\noutside the box\" to find a compensatory behavior when damaged: they are limited\nto their pre-specified self-sensing abilities, can diagnose only anticipated\nfailure modes, and require a pre-programmed contingency plan for every type of\npotential damage, an impracticality for complex robots. Here we introduce an\nintelligent trial and error algorithm that allows robots to adapt to damage in\nless than two minutes, without requiring self-diagnosis or pre-specified\ncontingency plans. Before deployment, a robot exploits a novel algorithm to\ncreate a detailed map of the space of high-performing behaviors: This map\nrepresents the robot's intuitions about what behaviors it can perform and their\nvalue. If the robot is damaged, it uses these intuitions to guide a\ntrial-and-error learning algorithm that conducts intelligent experiments to\nrapidly discover a compensatory behavior that works in spite of the damage.\nExperiments reveal successful adaptations for a legged robot injured in five\ndifferent ways, including damaged, broken, and missing legs, and for a robotic\narm with joints broken in 14 different ways. This new technique will enable\nmore robust, effective, autonomous robots, and suggests principles that animals\nmay use to adapt to injury.","url_abs":"http://arxiv.org/abs/1407.3501v4","url_pdf":"http://arxiv.org/pdf/1407.3501v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"robots-that-can-adapt-like-animals","repo_url":"https://github.com/jbmouret/limbo","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"robots-that-can-adapt-like-animals","repo_url":"https://github.com/uber-research/Map-Elites-Evolutionary","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1407.3501","atlas_url":"https://app.syntology.ai/?focus=1407.3501","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}