Safe Decentralized Multi-Agent Control using Black-Box Predictors, Conformal Decision Policies, and Control Barrier Functions

TL;DR

This paper presents a decentralized control framework for multi-agent robotic systems that ensures safety by adaptively balancing safety constraints and task goals using black-box trajectory predictors and conformal decision policies.

Contribution

It introduces a novel method combining conformal decision theory with control barrier functions to handle prediction uncertainty in decentralized multi-agent safety control.

Findings

The proposed controllers effectively balance safety and task objectives.

Experimental results demonstrate successful navigation in multi-agent scenarios.

The approach provides an upper bound on safety constraint violations over time.

Abstract

We address the challenge of safe control in decentralized multi-agent robotic settings, where agents use uncertain black-box models to predict other agents' trajectories. We use the recently proposed conformal decision theory to adapt the restrictiveness of control barrier functions-based safety constraints based on observed prediction errors. We use these constraints to synthesize controllers that balance between the objectives of safety and task accomplishment, despite the prediction errors. We provide an upper bound on the average over time of the value of a monotonic function of the difference between the safety constraint based on the predicted trajectories and the constraint based on the ground truth ones. We validate our theory through experimental results showing the performance of our controllers when navigating a robot in the multi-agent scenes in the Stanford Drone Dataset.