First, Learn What You Don't Know: Active Information Gathering for Driving at the Limits of Handling

TL;DR

This paper introduces an active information gathering framework using Bayesian meta-learning to rapidly identify vehicle dynamics, enabling safe and reliable control of vehicles at the limits of handling during dynamic maneuvers.

Contribution

It presents a novel active data collection approach with Bayesian last-layer meta-learning for fast online vehicle dynamics adaptation at the limits of handling.

Findings

Framework enables reliable control at the edge of stability.

Online adaptation alone may cause transient errors or spin-outs.

Active data collection improves control reliability.

Abstract

Combining data-driven models that adapt online and model predictive control (MPC) has enabled effective control of nonlinear systems. However, when deployed on unstable systems, online adaptation may not be fast enough to ensure reliable simultaneous learning and control. For example, a controller on a vehicle executing highly dynamic maneuvers--such as drifting to avoid an obstacle--may push the vehicle's tires to their friction limits, destabilizing the vehicle and allowing modeling errors to quickly compound and cause a loss of control. To address this challenge, we present an active information gathering framework for identifying vehicle dynamics as quickly as possible. We propose an expressive vehicle dynamics model that leverages Bayesian last-layer meta-learning to enable rapid online adaptation. The model's uncertainty estimates are used to guide informative data collection and quickly improve the model prior to deployment. Dynamic drifting experiments on a Toyota Supra show that (i) the framework enables reliable control of a vehicle at the edge of stability, (ii) online adaptation alone may not suffice for zero-shot control and can lead to undesirable transient errors or spin-outs, and (iii) active data collection helps achieve reliable performance.