Active Learning of Discrete-Time Dynamics for Uncertainty-Aware Model Predictive Control

10/23/2022
by   Alessandro Saviolo, et al.
0

Model-based control requires an accurate model of the system dynamics for precisely and safely controlling the robot in complex and dynamic environments. Moreover, in presence of variations in the operating conditions, the model should be continuously refined to compensate for dynamics changes. In this paper, we propose a self-supervised learning approach to actively model robot discrete-time dynamics. We combine offline learning from past experience and online learning from present robot interaction with the unknown environment. These two ingredients enable highly sample-efficient and adaptive learning for accurate inference of the model dynamics in real-time even in operating regimes significantly different from the training distribution. Moreover, we design an uncertainty-aware model predictive controller that is conditioned to the aleatoric (data) uncertainty of the learned dynamics. The controller actively selects the optimal control actions that (i) optimize the control performance and (ii) boost the online learning sample efficiency. We apply the proposed method to a quadrotor system in multiple challenging real-world experiments. Our approach exhibits high flexibility and generalization capabilities by consistently adapting to unseen flight conditions, while it significantly outperforms classical and adaptive control baselines.

READ FULL TEXT

page 1

page 5

page 9

page 14

research
10/23/2020

TAMPC: A Controller for Escaping Traps in Novel Environments

We propose an approach to online model adaptation and control in the cha...
research
07/10/2019

DOB-Net: Actively Rejecting Unknown Excessive Time-Varying Disturbances

This paper presents an observer-integrated Reinforcement Learning (RL) a...
research
03/11/2018

Experience Recommendation for Long Term Safe Learning-based Model Predictive Control in Changing Operating Conditions

Learning has propelled the cutting edge of performance in robotic contro...
research
12/17/2019

When Your Robot Breaks: Active Learning During Plant Failure

Detecting and adapting to catastrophic failures in robotic systems requi...
research
06/05/2020

An Ergodic Measure for Active Learning From Equilibrium

This paper develops KL-Ergodic Exploration from Equilibrium (KL-E^3), a ...
research
03/18/2022

Time-Optimal Online Replanning for Agile Quadrotor Flight

In this paper, we tackle the problem of flying a quadrotor using time-op...
research
03/27/2021

Minimum directed information: A design principle for compliant robots

A robot's dynamics – especially the degree and location of compliance – ...

Please sign up or login with your details

Forgot password? Click here to reset