TY - GEN
T1 - Automatically customizing a powered knee prosthesis with human in the loop using adaptive dynamic programming
AU - Wen, Yue
AU - Brandt, Andrea
AU - Si, Jennie
AU - Huang, He Helen
N1 - Funding Information:
This work was supported by NSF # 1406750 and #1361549. Y. Wen, A. Brandt, and H. Huang are with the NCSU/UNC Department of Biomedical Engineering, NC State University, Raleigh, NC, 27695-7115; University of North Carolina at Chapel Hill, Chapel Hill, NC, 27599. J. Si is with the Department of Electrical, Computer, and Energy Engineering, Arizona State University, Tempe, AZ, 85281. Corresponding author: Yue Wen (E-mail: ywen3@ncsu.edu).
Publisher Copyright:
© 2017 IEEE.
PY - 2018/6/12
Y1 - 2018/6/12
N2 - In this study, we validated a human-in-the-loop auto-tuner using machine learning to automatically customize powered knee prosthesis control parameters for an amputee subject in real time. The experimental powered knee prosthesis was controlled by a finite state impedance controller, which had 12 configurable impedance control parameters. Using adaptive dynamic programming with reinforcement learning while one transfemoral amputee subject walked with the powered knee prosthesis, the auto-tuner would interact with the human-prosthesis system and learn to configure the high dimension of control parameters. We tested 4 different initial conditions with one unilateral transfemoral amputee subject. The results showed that the auto-tuner discovered the control parameters that allowed amputee subject to generate normative knee kinematics in 3 out of 4 tuning sessions. For all test sessions, the averaged root-mean-square error of the knee kinematics relative to the normative knee kinematics decreased from 6.6 degrees to 4.6 degrees after tuning procedure.
AB - In this study, we validated a human-in-the-loop auto-tuner using machine learning to automatically customize powered knee prosthesis control parameters for an amputee subject in real time. The experimental powered knee prosthesis was controlled by a finite state impedance controller, which had 12 configurable impedance control parameters. Using adaptive dynamic programming with reinforcement learning while one transfemoral amputee subject walked with the powered knee prosthesis, the auto-tuner would interact with the human-prosthesis system and learn to configure the high dimension of control parameters. We tested 4 different initial conditions with one unilateral transfemoral amputee subject. The results showed that the auto-tuner discovered the control parameters that allowed amputee subject to generate normative knee kinematics in 3 out of 4 tuning sessions. For all test sessions, the averaged root-mean-square error of the knee kinematics relative to the normative knee kinematics decreased from 6.6 degrees to 4.6 degrees after tuning procedure.
UR - http://www.scopus.com/inward/record.url?scp=85049991814&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85049991814&partnerID=8YFLogxK
U2 - 10.1109/WEROB.2017.8383835
DO - 10.1109/WEROB.2017.8383835
M3 - Conference contribution
AN - SCOPUS:85049991814
T3 - 2017 International Symposium on Wearable Robotics and Rehabilitation, WeRob 2017
SP - 1
EP - 2
BT - 2017 International Symposium on Wearable Robotics and Rehabilitation, WeRob 2017
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2017 International Symposium on Wearable Robotics and Rehabilitation, WeRob 2017
Y2 - 5 November 2017 through 8 November 2017
ER -