PoSafeNet: Safe Learning with Poset-Structured Neural Nets

TL;DR

PoSafeNet introduces a novel neural safety layer that enforces heterogeneous safety constraints modeled as a poset, improving safety feasibility and robustness in robotic systems.

Contribution

It formalizes safety constraints as a poset and develops a differentiable safety layer that respects partial orderings, enabling adaptive safety enforcement.

Findings

Enhanced safety feasibility in robotic navigation and manipulation

Improved robustness and scalability over existing safety layers

Effective handling of heterogeneous safety constraints

Abstract

Safe learning is essential for deploying learningbased controllers in safety-critical robotic systems, yet existing approaches often enforce multiple safety constraints uniformly or via fixed priority orders, leading to infeasibility and brittle behavior. In practice, safety requirements are heterogeneous and admit only partial priority relations, where some constraints are comparable while others are inherently incomparable. We formalize this setting as poset-structured safety, modeling safety constraints as a partially ordered set and treating safety composition as a structural property of the policy class. Building on this formulation, we propose PoSafeNet, a differentiable neural safety layer that enforces safety via sequential closed-form projection under poset-consistent constraint orderings, enabling adaptive selection or mixing of valid safety executions while preserving priority semantics by construction. Experiments on multi-obstacle navigation, constrained robot manipulation, and vision-based autonomous driving demonstrate improved feasibility, robustness, and scalability over unstructured and differentiable quadratic program-based safety layers.