Learning responsibility allocations for multi-agent interactions: A differentiable optimization approach with control barrier functions

Isaac Remy1, David Fridovich-Keil2, Karen Leung1,3
1University of Washington 2UT Austin 3NVIDIA
American Control Conference 2025
Paper Code arXiv

When multiple agents must deviate from their ideal trajectories due to safety constraints, how do we infer who is more responsible for ensuring safety?

Abstract

From autonomous driving to package delivery, ensuring safe yet efficient multi-agent interaction is challenging as the interaction dynamics are influenced by hard-to-model factors such as social norms and contextual cues. Understanding these influences can aid in the design and evaluation of socially-aware autonomous agents whose behaviors are aligned with human values. In this work, we seek to codify factors governing safe multi-agent interactions via the lens of responsibility, i.e., an agent's willingness to deviate from their desired control to accommodate safe interaction with others. Specifically, we propose a data-driven modeling approach based on control barrier functions and differentiable optimization that efficiently learns agents' responsibility allocation from data. We demonstrate on synthetic and real-world datasets that we can obtain an interpretable and quantitative understanding of how much agents adjust their behavior to ensure the safety of others given their current environment.

Overview

MY ALT TEXT

We define responsibility as being how much each agent contributes to a shared control barrier function (CBF) safety constraint, and given the agents' desired control, we can solve for the safe control using a quadratic program (QP). With advances in differentiable optimization, we can then learn the responsibility allocations that solve this QP and infer, for a given state, the responsibility each agent has in satisfying the CBF constraint.

Experiments

MY ALT TEXT

We show our approach can work on N agents using synthetically generated data, and train a neural network to infer responsibilities for a traffic weaving scenario.

BibTeX

@InProceedings{RemyFridovichKeilEtAl2025,
        title={Learning Responsibility Allocations from Multiagent Interactions},
        author={Remy, Isaac and Fridovich-Keil, David, and Leung, Karen},
        booktitle={American Control Conference},
        year={2025}
      }