Meta-Adaptive Nonlinear Control: Theory and Algorithms
This work addresses the challenge of rapid adaptation to new environmental conditions in robot control, offering a unified framework with theoretical guarantees, though it is incremental in combining existing methods from control and learning.
The paper tackles the problem of controlling nonlinear systems with adversarial disturbances and unknown environment-dependent dynamics by introducing Online Meta-Adaptive Control (OMAC), which integrates online representation learning with control theory to achieve non-asymptotic convergence guarantees and outperforms conventional adaptive control methods in tasks like inverted pendulum and drone control under varying wind conditions.
We present an online multi-task learning approach for adaptive nonlinear control, which we call Online Meta-Adaptive Control (OMAC). The goal is to control a nonlinear system subject to adversarial disturbance and unknown $\textit{environment-dependent}$ nonlinear dynamics, under the assumption that the environment-dependent dynamics can be well captured with some shared representation. Our approach is motivated by robot control, where a robotic system encounters a sequence of new environmental conditions that it must quickly adapt to. A key emphasis is to integrate online representation learning with established methods from control theory, in order to arrive at a unified framework that yields both control-theoretic and learning-theoretic guarantees. We provide instantiations of our approach under varying conditions, leading to the first non-asymptotic end-to-end convergence guarantee for multi-task nonlinear control. OMAC can also be integrated with deep representation learning. Experiments show that OMAC significantly outperforms conventional adaptive control approaches which do not learn the shared representation, in inverted pendulum and 6-DoF drone control tasks under varying wind conditions.