Designing effective practice schedules for high-dimensional motor learning tasks remains a challenge, especially when skill states are unobservable and task performance may not reflect the true learning. We propose an automated curriculum design framework that combines a human motor learning model and personalized real-time skill estimation with Stochastic Nonlinear Model Predictive Control in \emph{de-novo} (novel) motor learning paradigms. We validated our framework both through simulations and human-subject studies (N = 36) using a hand exoskeleton. Our proposed approach accelerates skill acquisition by ∼23%, and ∼17% when compared to a random curriculum and a performance heuristics-based curriculum, respectively. These significant gains in learning efficiency highlight the potential of model-based, individualized curricula for motor rehabilitation and complex skill training.
@article{arxiv.2605.14367,
title = {Automated Curriculum Design for High-dimensional Human Motor Learning},
author = {Ankur Kamboj and Rajiv Ranganathan and Xiaobo Tan and Vaibhav Srivastava},
journal= {arXiv preprint arXiv:2605.14367},
year = {2026}
}