Research Analyzer
← Back ICRA 2026

Causality-Based Parametric Control Barrier Function for Safe Multi-Vehicle Interaction

Yiwei Lyu, Caleb Chang, John M. Dolan

PDF

AI summary

Key figure (auto-extracted from paper)
Embedding causality inference into Parametric-CBF enables accurate learning of heterogeneous driver safety behaviors, yielding safer and more efficient multi-vehicle control without restrictive assumptions.
Control Barrier Functions Causality Inference Multi-Vehicle Interaction Safe Control Autonomous Driving Parametric-CBF

Problem

Existing controller-learning methods struggle to isolate causal influences in multi-vehicle settings or rely on overly conservative worst-case analyses, often assuming homogeneous drivers or fully saturated safety constraints.

Approach

The authors apply Cross Map Smoothness causality detection to identify relevant pairwise interactions, then use this to selectively learn safety parameters for a Parametric-CBF, driving an adaptive safety-critical controller.

Key results

  • CMS-based causality detection for pairwise vehicle interactions
  • Causality-embedded Parametric-CBF for learning heterogeneous safety specifications
  • Adaptive safety-critical controller framework for cooperative driving
  • Demonstrated improvements in task efficiency and collective safety in multi-vehicle scenarios

Why it matters

Provides a scalable, data-driven foundation for safe autonomous driving in complex interactive environments where driver behaviors are heterogeneous and unpredictable.

Abstract

Safe control has been widely studied in various safety-critical applications, for instance, autonomous driving. In order to ensure the autonomous vehicle does not collide with other vehicles, it is essential to obtain an accurate expectation of surrounding vehicles’ behavior and react adaptively. Instead of assuming fully cooperative and homogeneous vehicles using the same safety-critical controllers, recent works have been explor- ing different data-driven approaches to model the neighboring vehicles’ underlying controllers with observed data. However, existing works either suffer from 1) the inter-vehicle influence during the multi-vehicle interaction, which makes it hard to determine the causality of surrounding vehicles’ behavior in controller modeling, or 2) being dominated by the worst-case analysis, which may lead to overly conservative behavior. In this paper, we extend the prior work on Parametric-Control Barrier Function (Parametric-CBF) to multi-robot interactions with embedded causality inference to explicitly reason over the inter-vehicle influence. Given the learned Causality-based Parametric-CBF, we present an adaptive safety-critical con- troller that allows the ego vehicle to safely react to surrounding vehicles with the learned expectation. We demonstrate that by leveraging the motion flexibility among multi-vehicle systems, task efficiency can be greatly improved in various interaction- intensive scenarios.

Index terms

Intelligent Transportation Systems Robot Safety

Related papers