Event
PVF Method Uses Temporal-Difference Learning to Compute Probabilistic Reach-Avoid Sets
1 reports · 1 sources
Researchers propose a self-supervised method, the probabilistic reach-avoid value function (PVF), based on temporal-difference learning, to compute reach-avoid sets for robotic tasks under stochastic, possibly unbounded disturbances, applicable to black-box systems. Compared with a Hamilton-Jacobi reachability (HJR)-based baseline on systems of increasing complexity (up to the Hopper locomotion task), results show that when disturbances are non-adversarial, the probabilistic reach-avoid sets obtained by PVF are less conservative and more accurate than adversarial HJR sets. The method has been submitted to ICRA 2027.
Report timeline
- Probabilistic Reach-Avoid Value Function (PVF): A Temporal-Difference Learning Method for Computing Probabilistic Reach-Avoid Sets
Proposes a self-supervised method, the Probabilistic Reach-Avoid Value Function (PVF), for computing reach-avoid sets for robot tasks under stochastic, potentially unbounded disturbances, applicable to black-box systems. Compared with Hamilton-Jacobi reachability (HJR)-based baselines on systems of increasing complexity, up to the Hopper locomotion task, results show that when disturbances are non-adversarial, PVF's probabilistic reach-avoid sets are less conservative and more accurate than adversarial HJR sets. The method has been submitted to ICRA 2027.
arXiv cs.RO ·