Data-Driven Robust Barrier Functions for Safe, Long-Term Operation

TL;DR

This paper presents a modular control framework using data-driven robust barrier functions to ensure safe, long-term operation of multi-robot systems in unpredictable environments, accommodating various disturbance models.

Contribution

It introduces a novel controller-synthesis method that models disturbances as convex hulls, enabling compatibility with different disturbance estimation techniques like Gaussian processes.

Findings

Successfully applied to multi-robot scenarios

Ensures constraint satisfaction under disturbances

Compatible with various disturbance estimation methods

Abstract

Applications that require multi-robot systems to operate independently for extended periods of time in unknown or unstructured environments face a broad set of challenges, such as hardware degradation, changing weather patterns, or unfamiliar terrain. To operate effectively under these changing conditions, algorithms developed for long-term autonomy applications require a stronger focus on robustness. Consequently, this work considers the ability to satisfy the operation-critical constraints of a disturbed system in a modular fashion, which means compatibility with different system objectives and disturbance representations. Toward this end, this paper introduces a controller-synthesis approach to constraint satisfaction for disturbed control-affine dynamical systems by utilizing Control Barrier Functions (CBFs). The aforementioned framework is constructed by modelling the disturbance as a union of convex hulls and leveraging previous work on CBFs for differential inclusions. This method of disturbance modeling grants compatibility with different disturbance-estimation methods. For example, this work demonstrates how a disturbance learned via a Gaussian process may be utilized in the proposed framework. These estimated disturbances are incorporated into the proposed controller-synthesis framework which is then tested on a fleet of robots in different scenarios.