{"id":25574,"date":"2026-04-08T09:15:44","date_gmt":"2026-04-08T09:15:44","guid":{"rendered":"https:\/\/aerospacerepository.org\/?p=25574"},"modified":"2026-04-08T09:15:46","modified_gmt":"2026-04-08T09:15:46","slug":"embedding-pontryagins-principle-in-neural-networks-for-optimal-asteroid-landing","status":"publish","type":"post","link":"https:\/\/aerospacerepository.org\/index.php\/2026\/04\/08\/embedding-pontryagins-principle-in-neural-networks-for-optimal-asteroid-landing\/","title":{"rendered":"Embedding Pontryagin\u2019s Principle in Neural Networks for Optimal Asteroid Landing"},"content":{"rendered":"\n<p><strong>SERGIO CUEVAS DEL VALLE; PABLO SOLANO-L\u00d3PEZ; HODEI URRUTXUA<\/strong><\/p>\n\n\n\n<p><strong>DOI Number: 10.13009\/EUCASS2023-056<\/strong><\/p>\n\n\n\n<p>This work proposes novel Neural Network (NN) training algorithms and architectures to solve with low\u00adcost general Optimal Control problems: regression of the optimal control policy for a given state trajectory. This is achieved via two novelties: first, a novel cost-effective optimal control solver is used for low-cost data augmentation of optimal state-control trajectories, and then combined with Feedforward and General Regression Neural Networks to solve the so-called direct regression problem. Secondly, Pontryagin&#8217;s Maximum Principle is leveraged to modify physics-informed NN to embed the Hamiltonian structure of Optimal Control within the training algorithm for enhanced robustness and generalized performance.<\/p>\n\n\n\n<p><a href=\"https:\/\/www.eucass.eu\/doi\/EUCASS2023-056.pdf\">Read the full paper here<\/a><\/p>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p><b>SERGIO CUEVAS DEL VALLE; PABLO SOLANO-L\u00d3PEZ; HODEI URRUTXUA<\/b><\/p>\n<p>DOI Number: 10.13009\/EUCASS2023-056<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[3766,993,3771],"tags":[3917,3918,2605,2799],"class_list":["post-25574","post","type-post","status-publish","format-standard","hentry","category-1-aerospace-europe-conference-2023","category-events","category-5-flight-dynamics-gnc-and-avionics","tag-ai","tag-asteroid-landing","tag-neural-networks","tag-optimal-control","category-3766","category-993","category-3771","description-off"],"_links":{"self":[{"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/posts\/25574","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/comments?post=25574"}],"version-history":[{"count":1,"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/posts\/25574\/revisions"}],"predecessor-version":[{"id":25575,"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/posts\/25574\/revisions\/25575"}],"wp:attachment":[{"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/media?parent=25574"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/categories?post=25574"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/tags?post=25574"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}