Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
144 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
45 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Optimal Potential Shaping on SE(3) via Neural ODEs on Lie Groups (2401.15107v2)

Published 25 Jan 2024 in math.OC and cs.LG

Abstract: This work presents a novel approach for the optimization of dynamic systems on finite-dimensional Lie groups. We rephrase dynamic systems as so-called neural ordinary differential equations (neural ODEs), and formulate the optimization problem on Lie groups. A gradient descent optimization algorithm is presented to tackle the optimization numerically. Our algorithm is scalable, and applicable to any finite dimensional Lie group, including matrix Lie groups. By representing the system at the Lie algebra level, we reduce the computational cost of the gradient computation. In an extensive example, optimal potential energy shaping for control of a rigid body is treated. The optimal control problem is phrased as an optimization of a neural ODE on the Lie group SE(3), and the controller is iteratively optimized. The final controller is validated on a state-regulation task.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (54)
  1. Murray RM, Li Z and Sastry SS. A mathematical introduction to robotic manipulation. 1994. ISBN 9781351469791. 10.1201/9781315136370.
  2. Schmid R. Infinite-dimensional lie groups and algebras in mathematical physics. Advances in Mathematical Physics 2010; 2010: 280362. 10.1155/2010/280362.
  3. Brockett RW. Lie Algebras and Lie Groups in Control Theory. Geometric Methods in System Theory 1973; : 43–8210.1007/978-94-010-2675-8_2.
  4. Jurdjevic V. Geometric Control Theory. Geometric Control Theory 1996; 10.1017/CBO9780511530036.
  5. Introduction to Mechanics and Symmetry, volume 17. Springer New York, 1999. ISBN 978-1-4419-3143-6. 10.1007/978-0-387-21792-5.
  6. Proportional Derivative (PD) Control on the Euclidean Group. 1995.
  7. Lee T, Leok M and McClamroch NH. Geometric tracking control of a quadrotor uav on se(3). In 49th IEEE Conference on Decision and Control (CDC). pp. 5420–5425. 10.1109/CDC.2010.5717652.
  8. Goodarzi F, Lee D and Lee T. Geometric nonlinear pid control of a quadrotor uav on se(3). In 2013 European Control Conference (ECC). pp. 3845–3850. 10.23919/ECC.2013.6669644.
  9. Rashad R, Califano F and Stramigioli S. Port-Hamiltonian Passivity-Based Control on SE(3) of a Fully Actuated UAV for Aerial Physical Interaction Near-Hovering. IEEE Robotics and Automation Letters 2019; 4(4): 4378–4385. 10.1109/LRA.2019.2932864.
  10. Time optimal control for linear systems on Lie groups. Systems & Control Letters 2021; 153: 104956. 10.1016/J.SYSCONLE.2021.104956.
  11. Spindler K. Optimal control on lie groups with applications to attitude control. Mathematics of Control, Signals and Systems 1998; 11(3): 197–219. 10.1007/BF02741891.
  12. Discrete geometric optimal control on lie groups. IEEE Transactions on Robotics 2011; 27(4): 641–655. 10.1109/TRO.2011.2139130.
  13. Saccon A, Hauser J and Aguiar AP. Optimal control on lie groups: The projection operator approach. IEEE Transactions on Automatic Control 2013; 58(9): 2230–2245. 10.1109/TAC.2013.2258817.
  14. Lie algebraic cost function design for control on lie groups. In 2022 IEEE 61st Conference on Decision and Control (CDC). pp. 1867–1874. 10.1109/CDC51059.2022.9993143.
  15. Machine learning and its impact on control systems: A review. Materials Today: Proceedings 2021; 47: 3744–3749. https://doi.org/10.1016/j.matpr.2021.02.281. 3rd International Conference on Computational and Experimental Methods in Mechanical Engineering.
  16. Taylor AT, Berrueta TA and Murphey TD. Active learning in robotics: A review of control principles. Mechatronics 2021; 77(May): 102576. 10.1016/j.mechatronics.2021.102576. 2106.13697.
  17. How to train your robot with deep reinforcement learning: lessons we have learned. International Journal of Robotics Research 2021; 40(4-5): 698–721. 10.1177/0278364920987859. 2102.02915.
  18. Soori M, Arezoo B and Dastres R. Artificial intelligence, machine learning and deep learning in advanced robotics, a review. Cognitive Robotics 2023; 3: 54–70. https://doi.org/10.1016/j.cogr.2023.04.001.
  19. Review of machine learning methods in soft robotics. PLoS One 2021; 16(2): e0246102.
  20. Paris R, Beneddine S and Dandois J. Robust flow control and optimal sensor placement using deep reinforcement learning. Journal of Fluid Mechanics 2021; 913: A25. 10.1017/jfm.2020.1170.
  21. Learning-Based Model Predictive Control: Toward Safe Learning in Control. Annual Review of Control, Robotics, and Autonomous Systems 2020; 3: 269–296. 10.1146/annurev-control-090419-075625.
  22. Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learning. Annual Review of Control, Robotics, and Autonomous Systems 2022; 5: 411–444. 10.1146/annurev-control-042920-020211. 2108.06266.
  23. Geometric Deep Learning. 2021. arXiv:2104.13478v1.
  24. Fanzhang L, Li Z and Zhao Z. Lie group machine learning. De Gruyter, 2019. ISBN 9783110499506. 10.1515/9783110499506/EPUB.
  25. Survey on lie group machine learning. Big Data Mining and Analytics 2020; 3(4): 235–258. 10.26599/BDMA.2020.9020011.
  26. Deep learning on lie groups for skeleton-based action recognition. In 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Los Alamitos, CA, USA: IEEE Computer Society, pp. 1243–1252. 10.1109/CVPR.2017.137.
  27. Machine learning Lie structures & applications to physics. Physics Letters B 2021; 817: 136297. 10.1016/J.PHYSLETB.2021.136297. 2011.00871.
  28. Deep Learning Symmetries and Their Lie Groups, Algebras, and Subalgebras from First Principles. Machine Learning: Science and Technology 2023; 10.1088/2632-2153/acd989.
  29. Hamiltonian-based Neural ODE Networks on the SE(3) Manifold For Dynamics Learning and Control. 2106.12782v3.
  30. Adaptive Control of SE(3) Hamiltonian Dynamics with Learned Disturbance Features. IEEE Control Systems Letters 2021; 6: 2773–2778. 10.1109/LCSYS.2022.3177156. 2109.09974.
  31. Neural ordinary differential equations. CoRR 2018; abs/1806.07366. URL http://arxiv.org/abs/1806.07366. 1806.07366.
  32. Optimal energy shaping via neural approximators. SIAM Journal on Applied Dynamical Systems 2022; 21(3): 2126–2147. 10.1137/21M1414279. URL https://doi.org/10.1137/21M1414279. https://doi.org/10.1137/21M1414279.
  33. Neural Ordinary Differential Equations on Manifolds, 2020. 2006.06663.
  34. Neural Manifold Ordinary Differential Equations. Advances in Neural Information Processing Systems 2020; 2020-Decem. 2006.10254.
  35. Dissecting neural ODEs. Advances in Neural Information Processing Systems 2020; 2020-Decem(NeurIPS). 2002.08071.
  36. Isham CJ. Modern Differential Geometry for Physicists. 1999. ISBN ISBN: 981-02-3555-0. https://doi.org/10.1142/3867.
  37. Hall BC. Lie Groups, Lie Algebras, and Representations: An Elementary Introduction. Graduate Texts in Mathematics (GTM, volume 222), Springer, 2015. ISBN ISBN 978-3319134666.
  38. Solà J, Deray J and Atchuthan D. A micro lie theory for state estimation in robotics, 2021. 1812.01537.
  39. A Stochastic Approximation Method. The Annals of Mathematical Statistics 1951; 22(3): 400 – 407. 10.1214/aoms/1177729586.
  40. Ruder S. An overview of gradient descent optimization algorithms, 2017. 1609.04747.
  41. Xie Z, Sato I and Sugiyama M. A diffusion theory for deep learning dynamics: Stochastic gradient descent exponentially favors flat minima. 2002.03495.
  42. Visser M, Stramigioli S and Heemskerk C. Cayley-Hamilton for roboticists. IEEE International Conference on Intelligent Robots and Systems 2006; 1: 4187–4192. 10.1109/IROS.2006.281911.
  43. Oprea J. Applications of Lusternik-Schnirelmann Category and its Generalizations. https://doiorg/107546/jgsp-36-2014-59-97 2014; 36(none): 59–97. 10.7546/JGSP-36-2014-59-97.
  44. A minimal atlas for the rotation group SO(3). GEM - International Journal on Geomathematics 2011 2:1 2011; 2(1): 113–122. 10.1007/S13137-011-0018-X.
  45. Rossmann W. Lie Groups: An Introduction Through Linear Groups. Oxford graduate texts in mathematics, Oxford University Press, 2006. ISBN 9780199202515.
  46. Munthe-Kaas H. High order runge-kutta methods on manifolds. Applied Numerical Mathematics 1999; 29(1): 115–127. https://doi.org/10.1016/S0168-9274(98)00030-0. Proceedings of the NSF/CBMS Regional Conference on Numerical Analysis of Hamiltonian Differential Equations.
  47. Lie-group methods. Acta Numerica 2000; 9: 215–365. 10.1017/S0962492900002154.
  48. Celledoni E, Marthinsen H and Owren B. An introduction to lie group integrators – basics, new developments and applications. Journal of Computational Physics 2014; 257: 1040–1061. 10.1016/j.jcp.2012.12.031.
  49. Lie group methods for rigid body dynamics and time integration on manifolds. Computer Methods in Applied Mechanics and Engineering 2003; 192(3): 421–438. https://doi.org/10.1016/S0045-7825(02)00520-0.
  50. Energy-balancing passivity-based control. Proceedings of the American Control Conference 2000; 2: 1265–1270. 10.1109/ACC.2000.876703.
  51. Control by interconnection and standard passivity-based control of port-hamiltonian systems. IEEE Transactions on Automatic Control 2008; 53: 2527–2542. 10.1109/TAC.2008.2006930.
  52. Discovering efficient periodic behaviors in mechanical systems via neural approximators. Optimal Control Applications and Methods 2023; 44(6): 3052–3079. https://doi.org/10.1002/oca.3025.
  53. Foundations of Mechanics, Second Edition. Addison-Wesley Publishing Company, Inc., 1987.
  54. Supplementary chapters for Geometric Control of Mechanical Systems FB/ADL:04, 2005.
Citations (1)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets