Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
194 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 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

Comparison of Motion Encoding Frameworks on Human Manipulation Actions (2211.13024v3)

Published 23 Nov 2022 in cs.RO

Abstract: Movement generation, and especially generalisation to unseen situations, plays an important role in robotics. Different types of movement generation methods exist such as spline based methods, dynamical system based methods, and methods based on Gaussian mixture models (GMMs). Using a large, new dataset on human manipulations, in this paper we provide a highly detailed comparison of five fundamentally different and widely used movement encoding and generation frameworks: dynamic movement primitives (DMPs), time based Gaussian mixture regression (tbGMR), stable estimator of dynamical systems (SEDS), Probabilistic Movement Primitives (ProMP) and Optimal Control Primitives (OCP). We compare these frameworks with respect to their movement encoding efficiency, reconstruction accuracy, and movement generalisation capabilities. The new dataset consists of nine object manipulation actions performed by 12 humans: pick and place, put on top/take down, put inside/take out, hide/uncover, and push/pull with a total of 7,652 movement examples. Our analysis shows that for movement encoding and reconstruction DMPs and OCPs are the most efficient with respect to the number of parameters and reconstruction accuracy, if a sufficient number of kernels is used. In case of movement generalisation to new start- and end-point situations, DMPs, OCPs and task parameterized GMM (TP-GMM, movement generalisation framework based on tbGMR) lead to similar performance, which ProMPs only achieve when using many demonstrations for learning. All models outperform SEDS, which additionally proves to be difficult to fit. Furthermore we observe that TP-GMM and SEDS suffer from problems reaching the end-points of generalizations.These different quantitative results will help selecting the most appropriate models and designing trajectory representations in an improved task-dependent way in future robotic applications.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (46)
  1. Robotics: Modelling, Planning and Control. Springer Publishing Company, 2009.
  2. Robot Programming by Demonstration. In Bruno Siciliano and Oussama Khatib, editors, Springer Handbook of Robotics, pages 1371–1394. Springer Berlin Heidelberg, Berlin, Heidelberg, 2008.
  3. Recent Advances in Robot Learning from Demonstration. Annual Review of Control, Robotics, and Autonomous Systems, 3(1):297–330, May 2020.
  4. Dynamical Movement Primitives: Learning Attractor Models for Motor Behaviors. Neural Computation, 25(2):328–373, February 2013.
  5. Learning nonlinear dynamical system for movement primitives. In 2014 IEEE International Conference on Systems, Man, and Cybernetics (SMC), pages 3761–3766, October 2014.
  6. Novel learning from demonstration approach for repetitive teleoperation tasks. In 2017 IEEE World Haptics Conference (WHC), pages 60–65, Munich, Germany, June 2017. IEEE.
  7. Orientation in Cartesian space dynamic movement primitives. In 2014 IEEE International Conference on Robotics and Automation (ICRA), pages 2997–3004, May 2014.
  8. Adaptation of manipulation skills in physical contact with the environment to reference force profiles. Autonomous Robots, 39, May 2015.
  9. Geometry-aware Dynamic Movement Primitives. In 2020 IEEE International Conference on Robotics and Automation (ICRA), pages 4421–4426, May 2020.
  10. Biologically-inspired dynamical systems for movement generation: Automatic real-time goal adaptation and obstacle avoidance. In 2009 IEEE International Conference on Robotics and Automation, pages 2587–2592, May 2009.
  11. Safe and Efficient Human–Robot Collaboration Part II: Optimal Generalized Human-in-the-Loop Real-Time Motion Generation. IEEE Robotics and Automation Letters, 3(4):3781–3788, October 2018.
  12. Dynamic movement primitives in robotics: A tutorial survey. The International Journal of Robotics Research, 42(13):1133–1184, November 2023.
  13. Learning Via-Point Movement Primitives with Inter- and Extrapolation Capabilities. In 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 4301–4308, Macau, China, November 2019. IEEE.
  14. Generation of movements with boundary conditions based on optimal control theory. Robotics and Autonomous Systems, 94:1–11, August 2017.
  15. BM: An iterative algorithm to learn stable non-linear dynamical systems with Gaussian mixture models. In 2010 IEEE International Conference on Robotics and Automation, pages 2381–2388, Anchorage, AK, May 2010. IEEE.
  16. Learning Stable Nonlinear Dynamical Systems With Gaussian Mixture Models. IEEE Transactions on Robotics, 27(5):943–957, October 2011.
  17. On Learning, Representing, and Generalizing a Task in a Humanoid Robot. IEEE Transactions on Systems, Man and Cybernetics, Part B (Cybernetics), 37(2):286–298, April 2007.
  18. Sylvain Calinon. A tutorial on task-parameterized movement learning and retrieval. Intelligent Service Robotics, 9(1):1–29, January 2016.
  19. Movement primitive learning and generalization: Using mixture density networks. IEEE Robotics & Automation Magazine, 27(2):22–32, 2020.
  20. Transverse contraction criteria for existence, stability, and robustness of a limit cycle. Systems & Control Letters, 63:32–38, January 2014.
  21. Harish chaandar Ravichandar and Ashwin Dani. Learning position and orientation dynamics from demonstrations via contraction analysis. Autonomous Robots, 43(4):897–912, April 2019.
  22. Learning periodic motions from human demonstrations using transverse contraction analysis. In 2016 American Control Conference (ACC), pages 4853–4858, July 2016.
  23. Probabilistic Movement Primitives. Advances in Neural Information Processing Systems (NIPS), page 9, January 2013.
  24. Improved Generalization of Probabilistic Movement Primitives for Manipulation Trajectories. IEEE Robotics and Automation Letters, 9(1):287–294, January 2024.
  25. On improving the extrapolation capability of task-parameterized movement models. In 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems, pages 610–616, November 2013.
  26. Open-source benchmarking for learned reaching motion generation in robotics. Paladyn, Journal of Behavioral Robotics, 6(1), March 2015.
  27. Probabilistic movement primitives. In C.J. Burges, L. Bottou, M. Welling, Z. Ghahramani, and K.Q. Weinberger, editors, Advances in Neural Information Processing Systems, volume 26. Curran Associates, Inc., 2013.
  28. An approach to learn hand movements for robot actions from human demonstrations. In 2016 IEEE/SICE International Symposium on System Integration (SII), pages 711–716, December 2016.
  29. Task-Specific Generalization of Discrete and Periodic Dynamic Movement Primitives. Robotics, IEEE Transactions on, 26:800–815, November 2010.
  30. A probabilistic approach to robot trajectory generation. IEEE-RAS International Conference on Humanoid Robots, 2015:477–483, February 2015.
  31. Using probabilistic movement primitives in robotics. Autonomous Robots, 42, March 2018.
  32. A Simple Ontology of Manipulation Actions Based on Hand-Object Relations. IEEE Transactions on Autonomous Mental Development, 5(2):117–134, June 2013.
  33. DeepLabCut: Markerless pose estimation of user-defined body parts with deep learning. Nature Neuroscience, 21(9):1281–1289, September 2018.
  34. Using DeepLabCut for 3D markerless pose estimation across species and behaviors. Nature Protocols, 14(7):2152–2176, July 2019.
  35. DeepLabCut/DeepLabCut, 2018.
  36. Anipose: A toolkit for robust markerless 3D pose estimation. bioRxiv : the preprint server for biology, page 2020.05.26.117325, May 2020.
  37. Lambdaloop/anipose, 2019.
  38. Dynamics systems vs. optimal control — a unifying view. In Progress in Brain Research, volume 165, pages 425–445. Elsevier, 2007.
  39. Joining Movement Sequences: Modified Dynamic Movement Primitives for Robotics Applications Exemplified on Handwriting. IEEE Transactions on Robotics, 28(1):145–157, February 2012.
  40. Movement imitation with nonlinear dynamical systems in humanoid robots. In Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No.02CH37292), volume 2, pages 1398–1403 vol.2, May 2002.
  41. Dynamic optimality in real-time: A learning framework for near-optimal robot motions. In 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems, pages 5636–5643, Tokyo, November 2013. IEEE.
  42. Maximum Likelihood from Incomplete Data Via the EM Algorithm. Journal of the Royal Statistical Society: Series B (Methodological), 39(1):1–22, September 1977.
  43. Bayesian multi-task reinforcement learning. In Proceedings of the 27th International Conference on International Conference on Machine Learning, ICML’10, pages 599–606, Madison, WI, USA, June 2010. Omnipress.
  44. Robotics Innovation Center DFKI GmbH. Dfki-ric/movement_primitives, 2021.
  45. Constructive Incremental Learning from Only Local Information. Neural Computation, 10(8):2047–2084, November 1998.
  46. Locally weighted projection regression: An o (n) algorithm for incremental real time learning in high dimensional space. In Proceedings of the Seventeenth International Conference on Machine Learning (ICML 2000), volume 1, pages 288–293. Morgan Kaufmann, 2000.

Summary

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

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