Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
126 tokens/sec
GPT-4o
47 tokens/sec
Gemini 2.5 Pro Pro
43 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Adaptive Contact-Implicit Model Predictive Control with Online Residual Learning (2310.09893v2)

Published 15 Oct 2023 in cs.RO

Abstract: The hybrid nature of multi-contact robotic systems, due to making and breaking contact with the environment, creates significant challenges for high-quality control. Existing model-based methods typically rely on either good prior knowledge of the multi-contact model or require significant offline model tuning effort, thus resulting in low adaptability and robustness. In this paper, we propose a real-time adaptive multi-contact model predictive control framework, which enables online adaption of the hybrid multi-contact model and continuous improvement of the control performance for contact-rich tasks. This framework includes an adaption module, which continuously learns a residual of the hybrid model to minimize the gap between the prior model and reality, and a real-time multi-contact MPC controller. We demonstrated the effectiveness of the framework in synthetic examples, and applied it on hardware to solve contact-rich manipulation tasks, where a robot uses its end-effector to roll different unknown objects on a table to track given paths. The hardware experiments show that with a rough prior model, the multi-contact MPC controller adapts itself on-the-fly with an adaption rate around 20 Hz and successfully manipulates previously unknown objects with non-smooth surface geometries.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (48)
  1. Y. Tassa, T. Erez, and E. Todorov, “Synthesis and Stabilization of Complex Behaviors Through Online Trajectory Optimization,” in 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, pp. 4906–4913, IEEE, 2012.
  2. A. Aydinoglu, V. M. Preciado, and M. Posa, “Contact-Aware Controller Design for Complementarity Systems,” arXiv preprint arXiv:1909.11221, 2019.
  3. S. L. Cleac’h, T. Howell, M. Schwager, and Z. Manchester, “Fast contact-implicit model-predictive control,” arXiv preprint arXiv:2107.05616, 2021.
  4. W. Jin and M. Posa, “Task-driven hybrid model reduction for dexterous manipulation,” arXiv preprint arXiv:2211.16657, 2022.
  5. Y.-M. Chen and M. Posa, “Optimal reduced-order modeling of bipedal locomotion,” in 2020 IEEE International Conference on Robotics and Automation (ICRA), pp. 8753–8760, IEEE, 2020.
  6. W. Jin, A. Aydinoglu, M. Halm, and M. Posa, “Learning linear complementarity systems,” in Learning for Dynamics and Control Conference, pp. 1137–1149, PMLR, 2022.
  7. S. Pfrommer, M. Halm, and M. Posa, “Contactnets: Learning discontinuous contact dynamics with smooth, implicit representations,” in Conference on Robot Learning, pp. 2279–2291, PMLR, 2021.
  8. H. Qi, A. Kumar, R. Calandra, Y. Ma, and J. Malik, “In-hand object rotation via rapid motor adaptation,” in Conference on Robot Learning, pp. 1722–1732, PMLR, 2023.
  9. A. Kumar, Z. Fu, D. Pathak, and J. Malik, “Rma: Rapid motor adaptation for legged robots,” arXiv preprint arXiv:2107.04034, 2021.
  10. S. Sastry, M. Bodson, and J. F. Bartram, “Adaptive control: stability, convergence, and robustness,” 1990.
  11. A. Aydinoglu, A. Wei, and M. Posa, “Consensus complementarity control for multi-contact mpc,” arXiv preprint arXiv:2304.11259, 2023.
  12. M. Geilinger, D. Hahn, J. Zehnder, M. Bächer, B. Thomaszewski, and S. Coros, “Add: Analytically differentiable dynamics for multi-body systems with frictional contact,” ACM Transactions on Graphics (TOG), vol. 39, no. 6, pp. 1–15, 2020.
  13. E. Heiden, D. Millard, E. Coumans, and G. S. Sukhatme, “Augmenting differentiable simulators with neural networks to close the sim2real gap,” arXiv preprint arXiv:2007.06045, 2020.
  14. T. A. Howell, S. L. Cleac’h, J. Brüdigam, J. Z. Kolter, M. Schwager, and Z. Manchester, “Dojo: A differentiable physics engine for robotics,” arXiv preprint arXiv:2203.00806, 2022.
  15. M. Parmar, M. Halm, and M. Posa, “Fundamental challenges in deep learning for stiff contact dynamics,” in 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 5181–5188, IEEE, 2021.
  16. B. Bianchini, M. Halm, N. Matni, and M. Posa, “Generalization bounded implicit learning of nearly discontinuous functions,” in Learning for Dynamics and Control Conference, pp. 1112–1124, PMLR, 2022.
  17. J.-P. Sleiman, F. Farshidian, M. V. Minniti, and M. Hutter, “A unified mpc framework for whole-body dynamic locomotion and manipulation,” IEEE Robotics and Automation Letters, vol. 6, no. 3, pp. 4688–4695, 2021.
  18. C. Mastalli, R. Budhiraja, W. Merkt, G. Saurel, B. Hammoud, M. Naveau, J. Carpentier, L. Righetti, S. Vijayakumar, and N. Mansard, “Crocoddyl: An efficient and versatile framework for multi-contact optimal control,” in 2020 IEEE International Conference on Robotics and Automation (ICRA), pp. 2536–2542, IEEE, 2020.
  19. A. W. Winkler, C. D. Bellicoso, M. Hutter, and J. Buchli, “Gait and trajectory optimization for legged systems through phase-based end-effector parameterization,” IEEE Robotics and Automation Letters, vol. 3, no. 3, pp. 1560–1567, 2018.
  20. F. R. Hogan and A. Rodriguez, “Reactive planar non-prehensile manipulation with hybrid model predictive control,” The International Journal of Robotics Research, vol. 39, no. 7, pp. 755–773, 2020.
  21. A. Nagabandi, K. Konolige, S. Levine, and V. Kumar, “Deep dynamics models for learning dexterous manipulation,” in Conference on Robot Learning, pp. 1101–1112, PMLR, 2020.
  22. A. M. Annaswamy and A. L. Fradkov, “A historical perspective of adaptive control and learning,” Annual Reviews in Control, vol. 52, pp. 18–41, 2021.
  23. A. Becker, P. Kumar, and C.-Z. Wei, “Adaptive control with the stochastic approximation algorithm: Geometry and convergence,” IEEE Transactions on Automatic Control, vol. 30, no. 4, pp. 330–338, 1985.
  24. M. Tanaskovic, L. Fagiano, R. Smith, and M. Morari, “Adaptive receding horizon control for constrained mimo systems,” Automatica, vol. 50, no. 12, pp. 3019–3029, 2014.
  25. H. Fukushima, T.-H. Kim, and T. Sugie, “Adaptive model predictive control for a class of constrained linear systems based on the comparison model,” Automatica, vol. 43, no. 2, pp. 301–308, 2007.
  26. A. Weiss and S. Di Cairano, “Robust dual control mpc with guaranteed constraint satisfaction,” in 53rd IEEE Conference on Decision and Control, pp. 6713–6718, IEEE, 2014.
  27. V. R. Desaraju, A. Spitzer, and N. Michael, “Experience-driven predictive control with robust constraint satisfaction under time-varying state uncertainty.,” in Robotics: Science and Systems, 2017.
  28. S. Wang, K. Pei, J. Whitehouse, J. Yang, and S. Jana, “Efficient formal safety analysis of neural networks,” in Advances in Neural Information Processing Systems, pp. 6367–6377, 2018.
  29. Y. Ma, S. Vichik, and F. Borrelli, “Fast stochastic mpc with optimal risk allocation applied to building control systems,” in 2012 IEEE 51st IEEE Conference on Decision and Control (CDC), pp. 7559–7564, IEEE, 2012.
  30. K. Pereida and A. P. Schoellig, “Adaptive model predictive control for high-accuracy trajectory tracking in changing conditions,” in 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 7831–7837, IEEE, 2018.
  31. C. Stamouli, A. Tsiamis, M. Morari, and G. J. Pappas, “Adaptive stochastic mpc under unknown noise distribution,” in Learning for Dynamics and Control Conference, pp. 596–607, PMLR, 2022.
  32. K. Y. Chee, T. C. Silva, M. A. Hsieh, and G. J. Pappas, “Enhancing sample efficiency and uncertainty compensation in learning-based model predictive control for aerial robots,” arXiv preprint arXiv:2308.00570, 2023.
  33. R. Guzman, R. Oliveira, and F. Ramos, “Adaptive model predictive control by learning classifiers,” in Learning for Dynamics and Control Conference, pp. 480–491, PMLR, 2022.
  34. Springer, 2016.
  35. D. Stewart and J. C. Trinkle, “An Implicit Time-stepping Scheme for Rigid Body Dynamics with Coulomb Friction,” in Proceedings 2000 ICRA. Millennium Conference. IEEE International Conference on Robotics and Automation. Symposia Proceedings (Cat. No. 00CH37065), vol. 1, pp. 162–169, IEEE, 2000.
  36. M. Anitescu, “Optimization-based simulation of nonsmooth rigid multibody dynamics,” Mathematical Programming, vol. 105, pp. 113–143, 2006.
  37. SIAM, 2009.
  38. A. Aydinoglu, P. Sieg, V. M. Preciado, and M. Posa, “Stabilization of complementarity systems via contact-aware controllers,” IEEE Transactions on Robotics, vol. 38, no. 3, pp. 1735–1754, 2021.
  39. W. Heemels, J. M. Schumacher, and S. Weiland, “Linear Complementarity Systems,” SIAM Journal on Applied Mathematics, vol. 60, no. 4, pp. 1234–1269, 2000.
  40. D. E. Stewart, “Rigid-body Dynamics with Friction and Impact,” SIAM review, vol. 42, no. 1, pp. 3–39, 2000.
  41. D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” arXiv preprint arXiv:1412.6980, 2014.
  42. T. Koller, F. Berkenkamp, M. Turchetta, and A. Krause, “Learning-based model predictive control for safe exploration,” in 2018 IEEE conference on decision and control (CDC), pp. 6059–6066, IEEE, 2018.
  43. G. Kulathunga, H. Hamed, and A. Klimchik, “Residual dynamics learning for trajectory tracking for multi-rotor aerial vehicles,” arXiv preprint arXiv:2305.15791, 2023.
  44. G. Torrente, E. Kaufmann, P. Föhn, and D. Scaramuzza, “Data-driven mpc for quadrotors,” IEEE Robotics and Automation Letters, vol. 6, no. 2, pp. 3769–3776, 2021.
  45. B. Bianchini, M. Halm, and M. Posa, “Simultaneous learning of contact and continuous dynamics,” in Conference on Robot Learning (CoRL), Nov. 2023.
  46. O. Khatib, “A unified approach for motion and force control of robot manipulators: The operational space formulation,” IEEE Journal on Robotics and Automation, vol. 3, no. 1, pp. 43–53, 1987.
  47. N. Hogan, “Impedance control: An approach to manipulation,” in 1984 American control conference, pp. 304–313, IEEE, 1984.
  48. R. O. Duda and P. E. Hart, “Use of the hough transformation to detect lines and curves in pictures,” Communications of the ACM, vol. 15, no. 1, pp. 11–15, 1972.
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