Publikation
Neuro-Symbolic Predicate Learning for Semantic Safe Robot Control
Zihan Ye; Jiayi Liu; Puze Liu; Jiayun Li; Georgia Chalvatzaki; Jan Peters; Kristian Kersting
In: Computing Research Repository eprint Journal (CoRR), Vol. abs/2609.39594, Pages 1-9, arXiv, 2026.
Zusammenfassung
As robots are increasingly deployed in everyday environments,
ensuring their safety has become a central challenge. Existing
methods often encode safety requirements as opaque math-
ematical/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 Learn-
ing for Semantic Safe Robot Control” (NEUPRO), which
leverages a differentiable reasoner that can learn reusable
safety representations from human-specified safety knowl-
edge. NEUPRO allows practitioners to express task-related
safety requirements as transparent symbolic rules, while en-
abling gradients to propagate through these rules to a fea-
ture extractor that maps raw observations to safety-relevant
concepts. As a result, the learned feature extractor is (softly)
grounded in human-understandable semantics, supports trans-
parent constraint evaluation, and is transferable across tasks.
By coupling interpretability with differentiability, NEUPRO
moves beyond opaque cost design toward reusable safety rea-
soning. To evaluate NEUPRO’s capability, we collect and re-
lease REASON, the first real robot benchmark dataset for
interpretable robot safety specification. Experiments on REA-
SON show that NEUPRO learns safety-critical features that
generalize across tasks, mitigate the interpretability limita-
tions of conventional black-box cost formulations, and provide
explicit explanations of safety violation.