Learning Constraints from Stochastic Partially-Observed Closed-Loop Demonstrations
Published in IEEE Control Systems Letters, Under Review, 2025
This paper studies how we can infer robotic safety constraints from demonstrations, where the expert demonstrations are corrupted by dynamics noise.
Recommended citation: Zhouyu Zhang, Chih-Yuan Chiu, and Glen Chou (2025). https://arxiv.org/pdf/2509.15109
Download Paper