arXivRobotics
Learning Contractive Dynamical Representations for Composite Adaptive Control
基於收縮動力學表徵學習的複合自適應控制框架
This paper introduces a representation-learning framework that combines statistical learning with composite adaptive control, utilizing hard-EM and Kalman smoothing to learn contractive latent disturbance representations with provable exponential convergence.
2 min read