PoSafeNet: Safe Learning with Poset-Structured Neural Nets
This addresses safety-critical deployment in robotics by handling partial priority relations among constraints, though it is incremental as it builds on existing safety layer methods.
The paper tackled the problem of enforcing heterogeneous safety constraints in learning-based robotic controllers by modeling them as a partially ordered set, and proposed PoSafeNet, a neural safety layer that improved feasibility, robustness, and scalability in experiments on navigation, manipulation, and driving tasks.
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.