{"id":90,"date":"2015-06-02T03:32:20","date_gmt":"2015-06-02T03:32:20","guid":{"rendered":"http:\/\/www.kurzweilai.net\/?p=253362"},"modified":"2015-06-03T02:49:51","modified_gmt":"2015-06-03T02:49:51","slug":"emulating-animals-these-robots-can-recover-from-damage-in-two-minutes","status":"publish","type":"post","link":"https:\/\/hoo.central12.com\/fugic\/2015\/06\/02\/emulating-animals-these-robots-can-recover-from-damage-in-two-minutes\/","title":{"rendered":"Emulating animals, these robots can recover from damage in two minutes"},"content":{"rendered":"<p><iframe frameborder=\"0\" height=\"360\" src=\"https:\/\/www.youtube.com\/embed\/T-c17RKh3uE?rel=0\" width=\"640\"><\/iframe><\/p>\n<p>Researchers in France and the U.S. have developed a new technology that enables robots to quickly recover from an injury in less than two minutes, similar to how injured animals adapt. Such autonomous mobile robots would be useful in remote or hostile environments such as disaster areas, space, and deep oceans.<\/p>\n<p>The video above shows a six-legged robot that adapts to keep walking even if two of its legs are\u00a0<a href=\"https:\/\/www.youtube.com\/watch?v=T-c17RKh3uE&amp;feature=youtu.be\" >broken<\/a>. It also shows a robotic arm that learned how to correctly place an object even with several broken motors.<\/p>\n<p>&#8220;When injured, animals do not start learning from scratch,&#8221; says Jean-Baptiste Mouret from Pierre and Marie Curie University. &#8220;Instead, they have intuitions about different ways to behave. These intuitions allow them to intelligently select a few, different behaviors to try out and, after these tests, they choose one that works in spite of the injury. We made robots that can do the same.&#8221;<\/p>\n<p>The researchers developed an &#8220;Intelligent Trial and Error&#8221; algorithm that allows robots to emulate animals: the robots conduct experiments to rapidly discover a compensatory behavior that works despite the damage.<\/p>\n<p>&#8220;For example, if walking, mostly on its hind legs, does not work well, it will next try walking mostly on its front legs,&#8221; \u00a0explains Antoine Cully, lead author of a May 28 cover article on this research in the journal <em>Nature. &#8220;<\/em>What&#8217;s surprising is how quickly it can learn a new way to walk. It&#8217;s amazing to watch a robot go from crippled and flailing around to efficiently limping away in about two minutes.&#8221;<\/p>\n<hr \/>\n<p><strong>Abstract of <em>Robots that can adapt like animals. <\/em><\/strong><\/p>\n<p>Robots have transformed many industries, most notably manufacturing<sup><a id=\"ref-link-6\" title=\"Siciliano, B. &amp; Khatib, O. Springer Handbook of Robotics (Springer, 2008).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref1\">1<\/a><\/sup>, and have the power to deliver tremendous benefits to society, such as in search and rescue<sup><a id=\"ref-link-7\" title=\"Murphy, R. R. Trial by fire. Robot. Automat. Mag. 11, 50-61 (2004).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref2\">2<\/a><\/sup>, disaster response<sup><a id=\"ref-link-8\" title=\"Nagatani, K. et al. Emergency response to the nuclear accident at the Fukushima Daiichi nuclear power plants using mobile rescue robots. J. Field Robot. 30, 44-63 (2013).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref3\">3<\/a><\/sup>, health care<sup><a id=\"ref-link-9\" title=\"Broadbent, E., Stafford, R. &amp; MacDonald, B. Acceptance of healthcare robots for the older population: review and future directions. Int. J. Social Robot. 1, 319-330 (2009).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref4\">4<\/a><\/sup>\u00a0and transportation<sup><a id=\"ref-link-10\" title=\"Thrun, S. et al. Stanley: the robot that won the DARPA grand challenge. J. Field Robot. 23, 661-692 (2006).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref5\">5<\/a><\/sup>. They are also invaluable tools for scientific exploration in environments inaccessible to humans, from distant planets<sup><a id=\"ref-link-11\" title=\"Sanderson, K. Mars rover Spirit (2003-10). Nature 463, 600 (2010).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref6\">6<\/a><\/sup>\u00a0to deep oceans<sup><a id=\"ref-link-12\" title=\"Antonelli, G. Fossen, T. I. &amp; Yoerger, D. R. in Springer Handbook of Robotics (eds Siciliano, B. &amp; Khatib, O.) 987-1008 (Springer, 2008).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref7\">7<\/a><\/sup>. A major obstacle to their widespread adoption in more complex environments outside factories is their fragility<sup><a id=\"ref-link-13\" title=\"Sanderson, K. Mars rover Spirit (2003-10). Nature 463, 600 (2010).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref6\">6<\/a>,\u00a0<a id=\"ref-link-14\" title=\"Carlson, J. &amp; Murphy, R. R. How UGVs physically fail in the field. IEEE Trans. Robot. 21, 423-437 (2005).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref8\">8<\/a><\/sup>. Whereas animals can quickly adapt to injuries, current robots cannot \u2018think outside the box\u2019 to find a compensatory behaviour when they are damaged: they are limited to their pre-specified self-sensing abilities, can diagnose only anticipated failure modes<sup><a id=\"ref-link-15\" title=\"Blanke, M. Kinnaert, M., Lunze, J. &amp; Staroswiecki, M. Diagnosis and Fault-Tolerant Control (Springer, 2006).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref9\">9<\/a><\/sup>, and require a pre-programmed contingency plan for every type of potential damage, an impracticality for complex robots<sup><a id=\"ref-link-16\" title=\"Sanderson, K. Mars rover Spirit (2003-10). Nature 463, 600 (2010).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref6\">6<\/a>,\u00a0<a id=\"ref-link-17\" title=\"Carlson, J. &amp; Murphy, R. R. How UGVs physically fail in the field. IEEE Trans. Robot. 21, 423-437 (2005).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref8\">8<\/a><\/sup>. A promising approach to reducing robot fragility involves having robots learn appropriate behaviours in response to damage<sup><a id=\"ref-link-18\" title=\"Sproewitz, A., Moeckel, R., Maye, J. &amp; Ijspeert, A. Learning to move in modular robots using central pattern generators and online optimization. Int. J. Robot. Res. 27, 423-443 (2008).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref10\">10<\/a>,\u00a0<a id=\"ref-link-19\" title=\"Christensen, D. J., Schultz, U. P. &amp; Stoy, K. A distributed and morphology-independent strategy for adaptive locomotion in self-reconfigurable modular robots. Robot. Auton. Syst. 61, 1021-1035 (2013).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref11\">11<\/a><\/sup>, but current techniques are slow even with small, constrained search spaces<sup><a id=\"ref-link-20\" title=\"Kober, J., Bagnell, J. A. &amp; Peters, J. Reinforcement learning in robotics: a survey. Int. J. Robot. Res. 32, 1238-1274 (2013).\" href=\"http:\/\/www.nature.com\/nature\/journal\/v521\/n7553\/full\/nature14422.html#ref12\">12<\/a><\/sup>. Here we introduce an intelligent trial-and-error algorithm that allows robots to adapt to damage in less than two minutes in large search spaces without requiring self-diagnosis or pre-specified contingency plans. Before the robot is deployed, it uses a novel technique to create a detailed map of the space of high-performing behaviours. This map represents the robot\u2019s prior knowledge about what behaviours it can perform and their value. When the robot is damaged, it uses this prior knowledge to guide a trial-and-error learning algorithm that conducts intelligent experiments to rapidly discover a behaviour that compensates for the damage. Experiments reveal successful adaptations for a legged robot injured in five different ways, including damaged, broken, and missing legs, and for a robotic arm with joints broken in 14 different ways. This new algorithm will enable more robust, effective, autonomous robots, and may shed light on the principles that animals use to adapt to injury.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Researchers in France and the U.S. have developed a new technology that enables robots to quickly recover from an injury in less than two minutes, similar to how injured animals adapt. Such autonomous mobile robots would be useful in remote or hostile environments such as disaster areas, space, and deep oceans. The video above shows [&#8230;]<\/p>\n","protected":false},"author":13,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[46,43],"tags":[],"class_list":["post-90","post","type-post","status-publish","format-standard","hentry","category-airobotics","category-news"],"_links":{"self":[{"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/posts\/90"}],"collection":[{"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/users\/13"}],"replies":[{"embeddable":true,"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/comments?post=90"}],"version-history":[{"count":1,"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/posts\/90\/revisions"}],"predecessor-version":[{"id":91,"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/posts\/90\/revisions\/91"}],"wp:attachment":[{"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/media?parent=90"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/categories?post=90"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/hoo.central12.com\/fugic\/wp-json\/wp\/v2\/tags?post=90"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}