Research

Paper

TESTING March 16, 2026

ReLU Barrier Functions for Nonlinear Systems with Constrained Control: A Union of Invariant Sets Approach

Authors

Pouya Samanipour, Hasan A. Poonawala

Abstract

Certifying safety for nonlinear systems with polytopic input constraints is challenging because CBF synthesis must ensure control admissibility under saturation. We propose an approximation--verification pipeline that performs convex barrier synthesis on piecewise-affine (PWA) surrogates and certifies safety for the original nonlinear system via facet-wise verification. To reduce conservatism while preserving tractability, we use a two-slope Leaky ReLU surrogate for the extended class-$\mathcal{K}$ function $α(\cdot)$ and combine multiple certificates using a Union of Invariant Sets (UIS). Counterexamples are handled through local uncertainty updates. Simulations on pendulum and cart-pole systems with input saturation show larger certified invariant sets than linear-$α$ designs with tractable computation time.

Metadata

arXiv ID: 2603.15286
Provider: ARXIV
Primary Category: eess.SY
Published: 2026-03-16
Fetched: 2026-03-17 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.15286v1</id>\n    <title>ReLU Barrier Functions for Nonlinear Systems with Constrained Control: A Union of Invariant Sets Approach</title>\n    <updated>2026-03-16T13:48:30Z</updated>\n    <link href='https://arxiv.org/abs/2603.15286v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.15286v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Certifying safety for nonlinear systems with polytopic input constraints is challenging because CBF synthesis must ensure control admissibility under saturation. We propose an approximation--verification pipeline that performs convex barrier synthesis on piecewise-affine (PWA) surrogates and certifies safety for the original nonlinear system via facet-wise verification. To reduce conservatism while preserving tractability, we use a two-slope Leaky ReLU surrogate for the extended class-$\\mathcal{K}$ function $α(\\cdot)$ and combine multiple certificates using a Union of Invariant Sets (UIS). Counterexamples are handled through local uncertainty updates. Simulations on pendulum and cart-pole systems with input saturation show larger certified invariant sets than linear-$α$ designs with tractable computation time.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='eess.SY'/>\n    <published>2026-03-16T13:48:30Z</published>\n    <arxiv:comment>Accepted to ACC 2026</arxiv:comment>\n    <arxiv:primary_category term='eess.SY'/>\n    <author>\n      <name>Pouya Samanipour</name>\n    </author>\n    <author>\n      <name>Hasan A. Poonawala</name>\n    </author>\n  </entry>"
}