Open Access Open Access  Restricted Access Subscription Access

Reinforcement Learning of 3D Musculoskeletal Model for Walking or Running with Minimum Efforts


Affiliations
1 Data Science Student, Manipal University, B-80 Aakriti Gardens, Nehru Nagar, Bhopal - 462 003, India
2 Senior Faculty, Data Science and Machine Learning, with Manipal ProLearn (Manipal Academy of Higher Education – South Bangalore Campus), 3rd Floor, Salarpuria Symphony, 7, Service Road, Pragathi Nagar, Electronics City Post, Bengaluru – 560 100, India

   Subscribe/Renew Journal


In this work we are trying to build a controller for a musculoskeletal model that has the goal of matching a given time-varying velocity vector. The major objective is building a musculoskeletal model which is fully comprehensive and reproduces realistic human movements driven by muscle contraction dynamics. Spectrum of human movement will be generated through variations in the anatomic model.

Keywords

3D Musculoskeletal Model, Artificial Intelligence, Reinforcement Learning.

Manuscript Received: April 3, 2020; Revised: April 25, 2020; Accepted: May 2, 2020.

User
Subscription Login to verify subscription
Notifications
Font Size

  • K. Arulkumaran, N. Dilokthanakul, M. Shanahan, and A.A. Bharath, "Classifying options for Deep Reinforcement Learning,” 2016. Imperial College, London. [Online]. Available: https://arxiv.org/pdf/1604.08153.pdf
  • P-L. Bacon, J. Harb, and D. Precup, "The option-critic architecture," Reasoning and Learning Lab, McGill University, 2016. [Online]. Available: https://arxiv.org/pdf/1609.05140.pdf
  • T. Haarnoja, A. Zhou, P. Abbeel, and S. Levine, “Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor,” 2018. [Online]. Available: https://arxiv.org/pdf/1801.01290.pd
  • D. Horgan, J. Quan, D. Budden, G. Barth-Maron, M. Hessel, H. van Hasselt, and D. Silver, “Distributed prioritized experience replay,” 2018. [Online]. Available: https://arxiv.org/pdf/1803.00933.pdf
  • G. Huang, Y. Li, G. Pleiss, Z. Liu, J. E. Hopcroft, and K. Q. Weinberger, Snapshot ensembles: Train 1, get m for free,” 2017. [Online]. Available: https://arxiv.org/abs/1704.00109
  • Z. Huang, S. Zhou, B. Zhuang, and X. Zhou, “Learning to run with actor-critic ensemble,” 2017. [Online]. Available: https://arXiv:1712.08987
  • I. Osband and C. Blundell, “A.P.B.V.R.: Deep exploration via bootstrapped dqn,” 2016.
  • W. Jas´kowski, O. R. Lykkebø, N. E. Toklu, N.E., F. Trifterer, Z. Buk, J. Koutník, and F. Gomez, Reinforcement Learning to Run... Fast. In: S. Escalera, M. Weimer (eds.) NIPS 2017 Competition Book. Springer, 2018.
  • C. T. John, F. C. Anderson, J. S. Higginson, and S. L. Delp, “Stabilisation of walking by intrinsic muscle properties revealed in a three-dimensional muscle-driven simulation,” Comput. Methods in Biomechanics and Biomedical Eng., vol. 16, no. 4, 2013. https://dx.doi.org/10.1080/10255842.2011.627560
  • L. Kidzin´ski, S. P. Mohanty, C. Ong, Z. Huang, S. Zhou, A. Pechenko, A. Stelmaszczyk, P. Jarosik, M. Pavlov, and S. Kolesnikov et al., “Learning to run challenge solutions: Adapting reinforcement learning methods for neuromusculoskeletal environments,” 2018. [Online]. Available: https://arxiv.org/pdf/1804.00361.pdf
  • L. Kidzin´ski, S. P. Mohanty, C. Ong, J. Hicks, S. F. Carroll, S. Levine, M. Salathé, and S. L. Delp, “Learning to run challenge: Synthesizing physiologically accurate motion using deep reinforcement learning,” in S. Escalera.
  • M. Weimer (eds.) NIPS 2017 Competition Book. Springer, Springer 2018. [Online]. Available: https://arxiv.org/abs/1804.00198

Abstract Views: 248

PDF Views: 0




  • Reinforcement Learning of 3D Musculoskeletal Model for Walking or Running with Minimum Efforts

Abstract Views: 248  |  PDF Views: 0

Authors

Vikram Singh Chandel
Data Science Student, Manipal University, B-80 Aakriti Gardens, Nehru Nagar, Bhopal - 462 003, India
Subhabaha Pal
Senior Faculty, Data Science and Machine Learning, with Manipal ProLearn (Manipal Academy of Higher Education – South Bangalore Campus), 3rd Floor, Salarpuria Symphony, 7, Service Road, Pragathi Nagar, Electronics City Post, Bengaluru – 560 100, India

Abstract


In this work we are trying to build a controller for a musculoskeletal model that has the goal of matching a given time-varying velocity vector. The major objective is building a musculoskeletal model which is fully comprehensive and reproduces realistic human movements driven by muscle contraction dynamics. Spectrum of human movement will be generated through variations in the anatomic model.

Keywords


3D Musculoskeletal Model, Artificial Intelligence, Reinforcement Learning.

Manuscript Received: April 3, 2020; Revised: April 25, 2020; Accepted: May 2, 2020.


References





DOI: https://doi.org/10.17010/ijcs%2F2020%2Fv5%2Fi2-3%2F152870