Neuro-Symbolic Predicate Learning for Semantic Safe Robot Control

2026-09-30

AI Summary

This work addresses the challenges of model uninterpretability, high task coupling, and insufficient semantic insight in safe robotic control by proposing a differentiable reasoning framework that integrates neural and symbolic paradigms. By combining human-defined symbolic rules with learned feature extraction, the method enables end-to-end gradient propagation to acquire reusable safety representations, thereby overcoming the limitations of conventional black-box optimization. The proposed approach facilitates transparent constraint evaluation and cross-task transfer while providing explicit semantic explanations for policy violations. Furthermore, this project introduces REASON, the first interpretable safety benchmark dataset collected from real-world robots. Collectively, these contributions significantly enhance both the generalizability and interpretability of safety-critical robotic control systems.

Abstract

As robots are increasingly deployed in everyday environments, ensuring their safety has become a central challenge. Existing methods often encode safety requirements as opaque mathematical/logical formulations or dense cost functions. While effective in specific tasks, they remain difficult to interpret, tightly coupled to individual tasks, and offer limited insight into why a robot action is considered safe or unsafe. To address this limitation, we propose ``Neuro-Symbolic Predicate Learning for Semantic Safe Robot Control'' (NEUPRO), which leverages a differentiable reasoner that can learn reusable safety representations from human-specified safety knowledge. NEUPRO allows practitioners to express task-related safety requirements as transparent symbolic rules, while enabling gradients to propagate through these rules to a feature extractor that maps raw observations to safety-relevant concepts. As a result, the learned feature extractor is (softly) grounded in human-understandable semantics, supports transparent constraint evaluation, and is transferable across tasks. By coupling interpretability with differentiability, NEUPRO moves beyond opaque cost design toward reusable safety reasoning. To evaluate NEUPRO's capability, we collect and release REASON, the first real robot benchmark dataset for interpretable robot safety specification. Experiments on REASON show that NEUPRO learns safety-critical features that generalize across tasks, mitigate the interpretability limitations of conventional black-box cost formulations, and provide explicit explanations of safety violation.

Problem

• robot safety
• interpretability
• neuro-symbolic learning
• safety specification
• transferability

Innovation

• Neuro-Symbolic Learning
• Differentiable Reasoning
• Semantic Safety
• Interpretable Robot Control
• Transferable Representations