Self-Supervised Meta-Learning for All-Layer DNN-Based Adaptive Control with Stability Guarantees

TL;DR

This paper presents a self-supervised meta-learning framework for adaptive control of DNNs that enables robots to adapt rapidly in dynamic environments with stability guarantees, outperforming traditional methods.

Contribution

It introduces a novel self-supervised meta-learning approach to pretrain DNNs for adaptive control, allowing full DNN online adaptation with stability guarantees.

Findings

Achieves 19-39% better performance than baseline methods.

Successfully handles large dynamic wind disturbances in quadrotor control.

Provides stability guarantees for the online adaptation law.

Abstract

A critical goal of adaptive control is enabling robots to rapidly adapt in dynamic environments. Recent studies have developed a meta-learning-based adaptive control scheme, which uses meta-learning to extract nonlinear features (represented by Deep Neural Networks (DNNs)) from offline data, and uses adaptive control to update linear coefficients online. However, such a scheme is fundamentally limited by the linear parameterization of uncertainties and does not fully unleash the capability of DNNs. This paper introduces a novel learning-based adaptive control framework that pretrains a DNN via self-supervised meta-learning (SSML) from offline trajectories and online adapts the full DNN via composite adaptation. In particular, the offline SSML stage leverages the time consistency in trajectory data to train the DNN to predict future disturbances from history, in a self-supervised manner without environment condition labels. The online stage carefully designs a control law and an adaptation law to update the full DNN with stability guarantees. Empirically, the proposed framework significantly outperforms (19-39%) various classic and learning-based adaptive control baselines, in challenging real-world quadrotor tracking problems under large dynamic wind disturbance.