TY - GEN
T1 - Unsupervised Coadaptation of an Assistive Interface to Facilitate Sensorimotor Learning of Redundant Control
AU - De Santis, Dalia
AU - Dzialecka, Patrycja
AU - Mussa-Ivaldi, Ferdinando A.
N1 - Funding Information:
*Research supported by NIDRR grant H133E120010 and NICHD grant 1R01HD072080. Results incorporated in this manuscript have received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Sklodowska-Curie, project REBoT, G.A. No 750464.
Publisher Copyright:
© 2018 IEEE.
PY - 2018/10/9
Y1 - 2018/10/9
N2 - Interfaces that exploit biological signals or movements to control the operation of lower-dimensional systems external to the body are at the frontier for augmenting human abilities, but also constitute a learning challenge for their users. We developed and tested an unsupervised coadaptive algorithm that changed the mapping of a body machine interface to match the natural movement distribution of the users. Users controlled a cursor on a computer monitor using arm and shoulder motions captured by a set of inertial sensors in either of three conditions: I) a constant body-to-cursor map obtained through Principal Component Analysis of calibration movements, ii) a map that was recomputed at specified points in time, iii) a map that adaptively changed over time. We used recursive online PCA to incrementally shift the projection space towards the 2-dimensional subspace capturing the greatest sensor signal variance. Results suggest that training with the coadaptive BMI allows for faster internalization of the control space while reducing user's reliance on visual feedback.
AB - Interfaces that exploit biological signals or movements to control the operation of lower-dimensional systems external to the body are at the frontier for augmenting human abilities, but also constitute a learning challenge for their users. We developed and tested an unsupervised coadaptive algorithm that changed the mapping of a body machine interface to match the natural movement distribution of the users. Users controlled a cursor on a computer monitor using arm and shoulder motions captured by a set of inertial sensors in either of three conditions: I) a constant body-to-cursor map obtained through Principal Component Analysis of calibration movements, ii) a map that was recomputed at specified points in time, iii) a map that adaptively changed over time. We used recursive online PCA to incrementally shift the projection space towards the 2-dimensional subspace capturing the greatest sensor signal variance. Results suggest that training with the coadaptive BMI allows for faster internalization of the control space while reducing user's reliance on visual feedback.
UR - http://www.scopus.com/inward/record.url?scp=85056579773&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85056579773&partnerID=8YFLogxK
U2 - 10.1109/BIOROB.2018.8487912
DO - 10.1109/BIOROB.2018.8487912
M3 - Conference contribution
AN - SCOPUS:85056579773
T3 - Proceedings of the IEEE RAS and EMBS International Conference on Biomedical Robotics and Biomechatronics
SP - 801
EP - 806
BT - BIOROB 2018 - 7th IEEE International Conference on Biomedical Robotics and Biomechatronics
PB - IEEE Computer Society
T2 - 7th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics, BIOROB 2018
Y2 - 26 August 2018 through 29 August 2018
ER -