Data-Attributed Adaptive Control Barrier Functions: Safety-Certified Training Data Curation via Influence Analysis

TL;DR

This paper introduces DA-CBF, a framework that uses influence analysis for data attribution to improve safety and performance in adaptive control barrier functions for autonomous navigation.

Contribution

It provides a formal analysis of how training data quality affects safety guarantees and proposes influence-based data curation to enhance control barrier function learning.

Findings

DA-CBF reduces prediction RMSE by 35.6%.

Expands the safe operating set by 39%.

Achieves collision-free navigation in complex environments.

Abstract

Learning-based adaptation of Control Barrier Function (CBF) parameters offers a promising path toward safe autonomous navigation that balances conservatism with performance. Yet the accuracy of the underlying safety predictor is ultimately constrained by training data quality, and no prior work has formally characterized how prediction errors propagate through the adaptive pipeline to degrade closed-loop safety guarantees. We introduce Data-Attributed Adaptive CBF (DA-CBF), a framework that integrates TracIn-based data attribution into adaptive CBF learning. Our theoretical contributions are fourfold: (i) corrected two-sided bounds relating the safety-loss surrogate to the CBF constraint margin; (ii) a safety margin preservation theorem showing that prediction error induces quantifiable margin degradation and, via a smooth parameter selector, yields a genuine closed-loop forward invariance guarantee not conditioned on a fixed trajectory; (iii) a CBF-QP constraint perturbation bound that links prediction accuracy directly to recursive feasibility; and (iv) a principled leave-one-out justification for influence-based data curation under explicit smoothness assumptions. On a DynamicUnicycle2D benchmark, DA-CBF reduces prediction RMSE by 35.6\%, expands the certified safe operating set by 39\%, and achieves collision-free navigation in a 16-obstacle environment where the uncurated baseline incurs 3 collisions.