Deep Equivariant Multi-Agent Control Barrier Functions

TL;DR

This paper introduces symmetry-aware distributed Control Barrier Functions for multi-agent systems, improving safety, scalability, and generalization in complex environments through equivariant neural network design.

Contribution

It proposes a novel symmetry-infused approach for learning safety certificates that generalize zero-shot to larger multi-agent systems, addressing scalability and efficiency issues.

Findings

Outperforms state-of-the-art baselines in safety and success rates

Enables zero-shot generalization to larger swarms

Improves data efficiency and scalability of safety enforcement

Abstract

With multi-agent systems increasingly deployed autonomously at scale in complex environments, ensuring safety of the data-driven policies is critical. Control Barrier Functions have emerged as an effective tool for enforcing safety constraints, yet existing learning-based methods often lack in scalability, generalization and sampling efficiency as they overlook inherent geometric structures of the system. To address this gap, we introduce symmetries-infused distributed Control Barrier Functions, enforcing the satisfaction of intrinsic symmetries on learnable graph-based safety certificates. We theoretically motivate the need for equivariant parametrization of CBFs and policies, and propose a simple, yet efficient and adaptable methodology for constructing such equivariant group-modular networks via the compatible group actions. This approach encodes safety constraints in a distributed data-efficient manner, enabling zero-shot generalization to larger and denser swarms. Through extensive simulations on multi-robot navigation tasks, we demonstrate that our method outperforms state-of-the-art baselines in terms of safety, scalability, and task success rates, highlighting the importance of embedding symmetries in safe distributed neural policies.