{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,10]],"date-time":"2026-02-10T22:04:04Z","timestamp":1770761044048,"version":"3.50.0"},"reference-count":91,"publisher":"SAGE Publications","issue":"10","license":[{"start":{"date-parts":[[2023,6,7]],"date-time":"2023-06-07T00:00:00Z","timestamp":1686096000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"funder":[{"name":"National Science Foundation (NSF), Cyber-Physical Systems","award":["1931815"],"award-info":[{"award-number":["1931815"]}]},{"name":"Energy, Power, Control, and Networks","award":["1809314"],"award-info":[{"award-number":["1809314"]}]},{"DOI":"10.13039\/501100000038","name":"Natural Sciences and Engineering Research Council of Canada","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000038","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["The International Journal of Robotics Research"],"published-print":{"date-parts":[[2023,9]]},"abstract":"<jats:p> Real-time adaptation is imperative to the control of robots operating in complex, dynamic environments. Adaptive control laws can endow even nonlinear systems with good trajectory tracking performance, provided that any uncertain dynamics terms are linearly parameterizable with known nonlinear features. However, it is often difficult to specify such features a priori, such as for aerodynamic disturbances on rotorcraft or interaction forces between a manipulator arm and various objects. In this paper, we turn to data-driven modeling with neural networks to learn, offline from past data, an adaptive controller with an internal parametric model of these nonlinear features. Our key insight is that we can better prepare the controller for deployment with control-oriented meta-learning of features in closed-loop simulation, rather than regression-oriented meta-learning of features to fit input-output data. Specifically, we meta-learn the adaptive controller with closed-loop tracking simulation as the base-learner and the average tracking error as the meta-objective. With both fully actuated and underactuated nonlinear planar rotorcraft subject to wind, we demonstrate that our adaptive controller outperforms other controllers trained with regression-oriented meta-learning when deployed in closed-loop for trajectory tracking control. <\/jats:p>","DOI":"10.1177\/02783649231165085","type":"journal-article","created":{"date-parts":[[2023,6,7]],"date-time":"2023-06-07T17:55:36Z","timestamp":1686160536000},"page":"777-797","update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":18,"title":["Control-oriented meta-learning"],"prefix":"10.1177","volume":"42","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7433-8774","authenticated-orcid":false,"given":"Spencer M.","family":"Richards","sequence":"first","affiliation":[{"name":"Department of Aeronautics & Astronautics, Stanford University, Stanford, CA, USA"}]},{"given":"Navid","family":"Azizan","sequence":"additional","affiliation":[{"name":"Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA"}]},{"given":"Jean-Jacques","family":"Slotine","sequence":"additional","affiliation":[{"name":"Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA"}]},{"given":"Marco","family":"Pavone","sequence":"additional","affiliation":[{"name":"Department of Aeronautics & Astronautics, Stanford University, Stanford, CA, USA"}]}],"member":"179","published-online":{"date-parts":[[2023,6,7]]},"reference":[{"key":"bibr1-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1002\/acs.1193"},{"issue":"2","key":"bibr2-02783649231165085","first-page":"107","volume":"1","author":"Agrawal A","year":"2019","journal-title":"Journal of Applied and Numerical Optimization"},{"key":"bibr3-02783649231165085","unstructured":"Agrawal A, Barratt S, Boyd S, et al. (2020) Learning convex optimization control policies. Learning for Dynamics & Control."},{"key":"bibr4-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2010.2040493"},{"key":"bibr5-02783649231165085","unstructured":"Amos B, Rodriguez IDJ, Sacks J, et al. (2018) Differentiable MPC for end-to-end planning and control. Conference on Neural Information Processing Systems."},{"key":"bibr6-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1016\/0005-1098(82)90021-8"},{"key":"bibr7-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1007\/s12532-018-0139-4"},{"key":"bibr8-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/9.989067"},{"key":"bibr9-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.1958.1105168"},{"key":"bibr10-02783649231165085","volume-title":"Feedback Systems: An Introduction for Scientists and Engineers","author":"\u00c5str\u00f6m KJ","year":"2020","edition":"2"},{"key":"bibr11-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1016\/0022-247X(71)90161-2"},{"key":"bibr12-02783649231165085","unstructured":"Azizan N, Hassibi B (2019) Stochastic gradient\/mirror descent: Minimax optimality and implicit regularization. International Conference on Learning Representations."},{"key":"bibr13-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2021.3087480"},{"key":"bibr14-02783649231165085","first-page":"267","volume":"4","author":"Barb\u0103lat I","year":"1959","journal-title":"Revue Roumaine de Math\u00e9matiques Pures et Appliqu\u00e9es"},{"key":"bibr15-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2021.3057046"},{"key":"bibr16-02783649231165085","unstructured":"Bertinetto L, Henriques J, Torr PHS, et al. (2019) Meta-learning with differentiable closed-form solvers. International Conference on Learning Representations."},{"key":"bibr17-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1162\/neco_a_01360"},{"key":"bibr18-02783649231165085","unstructured":"Boffi NM, Tu S, Matni N, et al. (2020) Learning stability certificates from data. Conference on Robot Learning."},{"key":"bibr19-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/CDC45484.2021.9682907"},{"key":"bibr20-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1016\/0005-1098(86)90002-6"},{"key":"bibr21-02783649231165085","volume-title":"Composable Transformations of Python+NumPy Programs","author":"Bradbury J","year":"2018"},{"key":"bibr22-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/TRO.2018.2813373"},{"key":"bibr23-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/CDC.2018.8618694"},{"key":"bibr24-02783649231165085","unstructured":"Chang Y-C, Roohi N, Gao S (2019) Neural Lyapunov control. Conference on Neural Information Processing Systems."},{"key":"bibr25-02783649231165085","unstructured":"Chen RTQ, Rubanova Y, Bettencourt J, et al. (2018) Neural ordinary differential equations. Conference on Neural Information Processing Systems."},{"key":"bibr26-02783649231165085","unstructured":"Clavera I, Rothfuss J, Schulman J, et al. (2018) Model-based reinforcement learning via meta-policy optimization. Conference on Robot Learning."},{"key":"bibr27-02783649231165085","unstructured":"Finn C, Abbeel P, Levine S (2017) Model-agnostic meta-learning for fast adaptation of deep networks. International Conference on Machine Learning."},{"key":"bibr28-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1016\/S0005-1098(99)00205-8"},{"key":"bibr29-02783649231165085","unstructured":"Gahlawat A, Zhao P, Patterson A, et al. 2020 L1-GP: L1adaptive control with Bayesian learning. Learning for Dynamics & Control."},{"key":"bibr30-02783649231165085","doi-asserted-by":"publisher","DOI":"10.3166\/ejc.11.335-352"},{"key":"bibr31-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/CDC.2013.6759990"},{"key":"bibr32-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-020-2649-2"},{"key":"bibr33-02783649231165085","unstructured":"Harrison J, Sharma A, Calandra R, et al. (2018a) Control adaptation via meta-learning dynamics. Conference on Neural Information Processing Systems - Workshop on Meta-Learning."},{"key":"bibr34-02783649231165085","unstructured":"Harrison J, Sharma A, Pavone M (2018b) Meta-learning priors for efficient online bayesian regression. Workshop on Algorithmic Foundations of Robotics."},{"key":"bibr35-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1016\/0005-1098(92)90029-F"},{"key":"bibr36-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1016\/S0005-1098(96)80003-3"},{"key":"bibr37-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3079209"},{"key":"bibr38-02783649231165085","volume-title":"Robust Adaptive Control","author":"Ioannou P","year":"2012"},{"key":"bibr39-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/CDC40024.2019.9029173"},{"key":"bibr40-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/CDC40024.2019.9029173"},{"key":"bibr41-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/MRA.2018.2866758"},{"key":"bibr42-02783649231165085","volume-title":"Nonlinear Systems","author":"Khalil HK","year":"2002"},{"key":"bibr43-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/TRO.2011.2159412"},{"key":"bibr44-02783649231165085","unstructured":"Kingma DP, Ba JL (2015) Adam: a method for stochastic optimization. International Conference on Learning Representations."},{"key":"bibr45-02783649231165085","author":"K\u00f6hler J","year":"2020","journal-title":"International Journal of Robust and Nonlinear Control"},{"key":"bibr46-02783649231165085","unstructured":"Kurutach T, Clavera I, Duan Y, et al. (2018) Model-ensemble trust-region policy optimization. International Conference on Learning Representations."},{"key":"bibr47-02783649231165085","unstructured":"Lakshminarayanan B, Pritzel A, Blundell C (2017) Simple and scalable predictive uncertainty estimation using deep ensembles. Conference on Neural Information Processing Systems."},{"key":"bibr48-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1007\/978-0-85729-664-1"},{"key":"bibr49-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/TCT.1960.1086720"},{"key":"bibr50-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4471-4396-3"},{"key":"bibr51-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01091"},{"key":"bibr52-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1002\/047134608X.W1046"},{"key":"bibr53-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1016\/S0005-1098(98)00019-3"},{"key":"bibr54-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/LCSYS.2020.3000190"},{"key":"bibr55-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/LCSYS.2021.3133359"},{"key":"bibr56-02783649231165085","volume-title":"Obshchaya Zadacha Ob Ustoichivosti Dvizheniya (The General Problem of the Stability of Motion)","author":"Lyapunov AM","year":"1892"},{"key":"bibr57-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2017.2668380"},{"key":"bibr58-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1016\/B978-0-08-034085-2.50031-6"},{"key":"bibr59-02783649231165085","unstructured":"Medina JR, Billard A (2017) Learning stable task sequences from demonstration with linear parameter varying systems and hidden Markov models. Conference on Robot Learning."},{"key":"bibr60-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2011.5980409"},{"key":"bibr61-02783649231165085","unstructured":"Millard D, Heiden E, Agrawal S, et al. (2020) Automatic differentiation and continuous sensitivity analysis of rigid body dynamics. Available at https:\/\/arxiv.org\/abs\/2001.08539."},{"key":"bibr62-02783649231165085","unstructured":"Nagabandi A, Clavera I, Liu S, et al. (2019) Learning to adapt in dynamic, real-world environments through meta-reinforcement learning. International Conference on Learning Representations."},{"key":"bibr63-02783649231165085","volume-title":"Stable Adaptive Systems","author":"Narendra KS","year":"2005"},{"key":"bibr64-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.1980.1102362"},{"key":"bibr65-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1177\/027836499101000206"},{"key":"bibr66-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.1978.1101823"},{"key":"bibr67-02783649231165085","unstructured":"O\u2019Connell M, Shi G, Shi X, et al. (2021) Meta-learning-based robust adaptive flight control under uncertain wind conditions. Available at https:\/\/arxiv.org\/abs\/2103.01932."},{"key":"bibr68-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1007\/BF00140873"},{"key":"bibr69-02783649231165085","volume-title":"The Mathematical Theory of Optimal Processes","author":"Pontryagin LS","year":"1962"},{"key":"bibr70-02783649231165085","unstructured":"Rajeswaran A, Ghotra S, Ravindran B, et al. (2017) EPOpt: learning robust neural network policies using model ensembles. International Conference on Learning Representations."},{"key":"bibr71-02783649231165085","unstructured":"Rajeswaran A, Finn C, Kakade S, et al. (2019) Meta-learning with implicit gradients. Conference on Neural Information Processing Systems."},{"key":"bibr72-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/MRA.2019.2955964"},{"key":"bibr73-02783649231165085","unstructured":"Richards SM, Berkenkamp F, Krause A (2018) The Lyapunov neural network: Adaptive stability certification for safe learning of dynamical systems. Conference on Robot Learning."},{"key":"bibr74-02783649231165085","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2021.XVII.056"},{"key":"bibr75-02783649231165085","unstructured":"Richter C, Bry A, Roy N (2013) Polynomial trajectory planning for aggressive quadrotor flight in dense indoor environments. Proceedings of the International Symposium of Robotics Research."},{"key":"bibr76-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2017.7989608"},{"key":"bibr77-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1995.7.4.753"},{"key":"bibr78-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1177\/0278364920949931"},{"key":"bibr79-02783649231165085","doi-asserted-by":"publisher","DOI":"10.23919\/ACC53348.2022.9867457"},{"key":"bibr80-02783649231165085","doi-asserted-by":"publisher","DOI":"10.23919\/ACC53348.2022.9867457"},{"key":"bibr81-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1080\/00207178908559735"},{"key":"bibr82-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1177\/027836498700600303"},{"key":"bibr83-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1016\/0005-1098(89)90094-0"},{"key":"bibr84-02783649231165085","volume-title":"Applied Nonlinear Control","author":"Slotine J-JE","year":"1991"},{"key":"bibr85-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/CDC40024.2019.9029693"},{"key":"bibr86-02783649231165085","unstructured":"Sun D, Jha S, Fan C (2020) Learning certified control using contraction metric. Conference on Robot Learning."},{"key":"bibr87-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/LCSYS.2020.3046529"},{"issue":"12","key":"bibr88-02783649231165085","doi-asserted-by":"crossref","first-page":"e0236661","DOI":"10.1371\/journal.pone.0236661","volume":"15","author":"Wensing PM","year":"2020","journal-title":"PLoS ONE"},{"key":"bibr89-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2017.2729659"},{"key":"bibr90-02783649231165085","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2020.3010460"},{"key":"bibr91-02783649231165085","unstructured":"Zhuang J, Dvornek N, Li X, et al. (2020) Adaptive checkpoint adjoint method for gradient estimation in neural ODE. International Conference on Machine Learning."}],"container-title":["The International Journal of Robotics Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/02783649231165085","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/02783649231165085","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/02783649231165085","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,3]],"date-time":"2025-03-03T16:03:22Z","timestamp":1741017802000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/02783649231165085"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2023,6,7]]},"references-count":91,"journal-issue":{"issue":"10","published-print":{"date-parts":[[2023,9]]}},"alternative-id":["10.1177\/02783649231165085"],"URL":"https:\/\/doi.org\/10.1177\/02783649231165085","relation":{},"ISSN":["0278-3649","1741-3176"],"issn-type":[{"value":"0278-3649","type":"print"},{"value":"1741-3176","type":"electronic"}],"subject":[],"published":{"date-parts":[[2023,6,7]]}}}