Robots safely reach and remain in targets using black-box control filters
Winning a Won Game: Strict Reach-Avoid-Stay Control Barrier Functions for High-Dimensional Black-Box Systems
Robotics
Summary
Robots need to safely reach certain goal areas and stay there without causing accidents. The authors created a new method that can learn these safe behaviors even when the robot’s inner workings are unknown, treating it like a 'black box.' Their approach combines ways to make sure the robot can reach the goal safely and stay there permanently, despite possible uncertainties. They tested this on simulated and real robots jumping gaps and racing safely, showing it works in tricky, high-dimensional problems.
What this means in practice
- •For robotics software engineers: Implement safety filters that ensure robots safely reach and stay in target zones without knowing their underlying dynamics.
- •For autonomous vehicle developers: Enable safe overtaking and persistent safe behavior in high-speed racing by approximating safety values from black-box simulations.
Authors
Donggeon David Oh, Duy P. Nguyen, Gongkai Yuan, Qingchen Li, Jaime Fernández Fisac, Haimin Hu
Abstract
Robots must complete their tasks and maintain the achieved outcomes while avoiding safety failures at all times. Strict reach-avoid-stay (sRAS) formalizes this requirement: safely reaching a target and remaining there indefinitely after first entry. We propose an sRAS Q-control barrier function (CBF) safety filter for high-dimensional black-box systems under bounded uncertainty. Our construction combines a stay value encoding safe permanent residence in a target subset with a reach-avoid value encoding safe reachability of this subset while avoiding target states from which safe permanent residence cannot be guaranteed. We prove that these values jointly yield a valid robust discrete-time CBF and lift them to state-action Q-functions for runtime intervention. For exact values and under a measure-zero condition, our filter preserves sRAS feasibility from almost every winnable initial state and keeps the system safely within the target after first entry, against all admissible uncertainty realizations. We adopt reachability-based adversarial reinforcement learning for scalable value approximation using only black-box interactions. Notably, neither synthesis nor deployment of our filter requires known dynamics, affine structure, value derivatives, or hand-designed barriers. We validate our framework in quadruped gap jumping in simulation and hardware, where the robot crosses the gap, lands safely, and remains safe afterward. Simulated F1TENTH races further demonstrate safe overtaking and lead retention.