Muscle Activation Analysis from Gait Kinematics and Reinforcement Learning
2
Issued Date
2022-01-01
Resource Type
Scopus ID
2-s2.0-85133392260
Journal Title
19th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, ECTI-CON 2022
Rights Holder(s)
SCOPUS
Bibliographic Citation
19th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, ECTI-CON 2022 (2022)
Suggested Citation
Jatesiktat P., Anopas D., Kwong W.H., Sidarta A., Liang P., Ang W.T. Muscle Activation Analysis from Gait Kinematics and Reinforcement Learning. 19th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology, ECTI-CON 2022 (2022). doi:10.1109/ECTI-CON54298.2022.9795606 Retrieved from: https://repository.li.mahidol.ac.th/handle/123456789/84632
Title
Muscle Activation Analysis from Gait Kinematics and Reinforcement Learning
Author(s)
Other Contributor(s)
Abstract
We propose the use of reinforcement learning with imitation reward to estimate muscle activation from a purely kinematic motion capture sequence without the use of any force plate or electromyography (EMG) sensors. We also demonstrate the use of this method by comparing muscle activation between normal walking and U-Turning. Our simulation demonstrated a higher level of activation during U-Turning in the biceps femoris in the swing phase and the gluteus medius during the stance phase, which is consistent with the previous studies with EMG sensors on human subjects. Activation of ankle muscles generated from the simulation, however, did not match the conventional activation patterns. The source code and the data are made publicly available for research purposes.