Safe Learning of Uncertain Environments

TL;DR

This paper presents a method for ensuring safety in nonlinear control systems during simultaneous learning and control by modeling uncertainties as Gaussian noise and adjusting control inputs with time-varying safety constraints.

Contribution

It introduces a novel approach to guarantee safety with high probability while learning the environment's uncertainty in real-time, including extensions for non-Gaussian noise.

Findings

Guarantees safety with high probability during learning and control.

Provides computationally efficient optimization formulations.

Extends to non-Gaussian and state-dependent uncertainties.

Abstract

In many learning based control methodologies, learning the unknown dynamic model precedes the control phase, while the aim is to control the system such that it remains in some safe region of the state space. In this work, our aim is to guarantee safety while learning and control proceed simultaneously. Specifically, we consider the problem of safe learning in nonlinear control-affine systems subject to unknown additive uncertainty. We first model the uncertainty as a Gaussian noise and use state measurements to learn its mean and covariance. We provide rigorous time-varying bounds on the mean and covariance of the uncertainty and employ them to modify the control input via an optimization program with potentially time-varying safety constraints. We show that with an arbitrarily large probability we can guarantee that the state will remain in the safe set, while learning and control are carried out simultaneously, provided that a feasible solution exists for the optimization problem. We provide a secondary formulation of this optimization that is computationally more efficient. This is based on tightening the safety constraints to counter the uncertainty about the learned mean and covariance. The magnitude of the tightening can be decreased as our confidence in the learned mean and covariance increases (i.e., as we gather more measurements about the environment). Extensions of the method are provided for non-Gaussian process noise with unknown mean and covariance as well as Gaussian uncertainties with state-dependent mean and covariance to accommodate more general environments.