{"id":2185,"date":"2020-01-12T12:26:18","date_gmt":"2020-01-12T02:26:18","guid":{"rendered":"https:\/\/www.cognav.net\/?p=2185"},"modified":"2020-01-12T12:26:18","modified_gmt":"2020-01-12T02:26:18","slug":"learning-from-animals-how-to-navigate-complex-terrains","status":"publish","type":"post","link":"https:\/\/braininspirednavigation.com\/?p=2185","title":{"rendered":"Learning from animals: How to Navigate Complex Terrains?"},"content":{"rendered":"<p style=\"text-align: justify;\">Zhu H, Liu H, Ataei A, Munk Y, Daniel T, Paschalidis IC (2020) <a href=\"https:\/\/journals.plos.org\/ploscompbiol\/article?id=10.1371\/journal.pcbi.1007452\"><strong>Learning from animals: How to Navigate Complex Terrains<\/strong><\/a>. PLoS Comput Biol 16(1): e1007452. https:\/\/doi.org\/10.1371\/journal.pcbi.1007452<\/p>\n<p style=\"text-align: justify;\">Abstract<br \/>\n&#8220;<strong><span style=\"color: #ff0000;\">We develop a method to learn a bio-inspired motion control policy using data collected from hawkmoths navigating in a virtual forest<\/span><\/strong>. A Markov Decision Process (MDP) framework is introduced to model the dynamics of moths and sparse logistic regression is used to learn control policy parameters from the data. <strong><span style=\"color: #ff0000;\">The results show that moths do not favor detailed obstacle location information in navigation, but rely heavily on optical flow<\/span><\/strong>. Using the policy learned from the moth data as a starting point, we propose an actor-critic learning algorithm to refine policy parameters and obtain a policy that can be used by an autonomous aerial vehicle operating in a cluttered environment. <strong><span style=\"color: #ff0000;\">Compared with the moths\u2019 policy, the policy we obtain integrates both obstacle location and optical flow<\/span><\/strong>. We compare the performance of these two policies in terms of their ability to navigate in artificial forest areas. While the optimized policy can adjust its parameters to outperform the moth\u2019s policy in each different terrain, the moth\u2019s policy exhibits a high level of robustness across terrains.&#8221;<\/p>\n<p style=\"text-align: justify;\">Author summary<br \/>\n&#8220;<strong><span style=\"color: #ff0000;\">Many animals exhibit a remarkable ability to navigate in complex forest terrains. Can we learn their navigation strategy from observed flying trajectories?<\/span><\/strong> Further, <strong><span style=\"color: #ff0000;\">can we refine these strategies to design UAV\/drone navigation policies in dense cluttered terrains?<\/span> <\/strong>To that end, we propose a method to analyze data from hawkmoth flight trajectories in a closed-loop virtual forest and extract the navigation control policy. <strong><span style=\"color: #ff0000;\">We find that moths rely heavily on optical flow rather than detailed information on the location of obstacles around them<\/span><\/strong>. We also develop a method to refine the hawkmoth control policy to be used by autonomous aerial vehicles in a cluttered environment. <strong><span style=\"color: #ff0000;\">We find that integrating both obstacle location information and optical flow improves navigation performance<\/span><\/strong>.&#8221;<\/p>\n<p style=\"text-align: justify;\">Zhu H, Liu H, Ataei A, Munk Y, Daniel T, Paschalidis IC (2020) <a href=\"https:\/\/journals.plos.org\/ploscompbiol\/article?id=10.1371\/journal.pcbi.1007452\"><strong>Learning from animals: How to Navigate Complex Terrains<\/strong><\/a>. PLoS Comput Biol 16(1): e1007452. https:\/\/doi.org\/10.1371\/journal.pcbi.1007452<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Zhu H, Liu H, Ataei A, Munk Y, Daniel T, Paschalidis IC (2020) Learning from animals: How to Navigate Complex Terrains. PLoS Comput Biol 16(1): e1007452. https:\/\/doi.org\/10.1371\/journal.pcbi.1007452 Abstract &#8220;We develop a method to learn a bio-inspired motion control policy using data collected from hawkmoths navigating in a virtual forest. A Markov Decision Process (MDP) framework [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[96,468,376],"tags":[698,700,695,699,701,696,688,697],"_links":{"self":[{"href":"https:\/\/braininspirednavigation.com\/index.php?rest_route=\/wp\/v2\/posts\/2185"}],"collection":[{"href":"https:\/\/braininspirednavigation.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/braininspirednavigation.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/braininspirednavigation.com\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/braininspirednavigation.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=2185"}],"version-history":[{"count":1,"href":"https:\/\/braininspirednavigation.com\/index.php?rest_route=\/wp\/v2\/posts\/2185\/revisions"}],"predecessor-version":[{"id":2186,"href":"https:\/\/braininspirednavigation.com\/index.php?rest_route=\/wp\/v2\/posts\/2185\/revisions\/2186"}],"wp:attachment":[{"href":"https:\/\/braininspirednavigation.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=2185"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/braininspirednavigation.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=2185"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/braininspirednavigation.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=2185"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}