The paper proves that using a permissive safety filter in reinforcement learning does not compromise asymptotic performance. By formalizing safety through a safety‑critical Markov decision process and a filtered MDP, the authors show that optimal policies in the filtered MDP achieve the same return as the best safe policy in the original setting. Experiments on Safety Gymnasium confirm zero violations during training and performance that matches or exceeds unfiltered baselines.
By Donggeon David Oh, Duy P. Nguyen, Haimin Hu, Jaime Fern\'andez Fisac
arXiv:2605. 26452v2 Announce Type: replace-cross Abstract: Safe reinforcement learning (RL) for robotic systems requires policies that improve task performance while satisfying state and input constraints during both training and deployment.
By Dhruv S. Kushwaha, Zoleikha A. Biron
The paper introduces Safe-Stop, a task‑agnostic framework for humanoid robots that determines whether an emergency stop command can be safely executed from the current state. It couples a learned stop policy with two complementary stoppability estimators: a stop‑probability estimator trained on outcomes of a fixed stop policy, and a reach‑avoidance estimator trained via Hamilton‑Jacobi backup. By requiring agreement between both estimators before committing to a stop, Safe‑Stop can robustly decide to stop or hand off to a fall‑damping fallback without needing retraining for different upstream tasks.
By Junfeng Long, Pieter Abbeel, Koushil Sreenath, Roberto Horowitz, Guanya Shi, C. Karen Liu
arXiv:2610.29149v1 Announce Type: cross
Abstract: Learning safe policies that generalize to unseen environments is difficult under bounded actuation, where an unsafe decision may become unrecoverable...
By Benedictus C. G. Cinun, Bin Hu
arXiv:2608. 10204v1 Announce Type: new Abstract: Safe reinforcement learning maximizes reward subject to safety constraints.
By Chenhua Fan, Jiahui Zhu, Yuhang Zhang, Honghao Wei
arXiv:2609. 11697v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) and World-Action Models (WAMs) have demonstrated strong capabilities in general-purpose robotic manipulation, yet their generated actions may violate hard physical constraints and therefore be unsafe or infeasible for deployment.
By Jianming Ma, Rongjun Jin, Xiaxi Si, Yang Zhang, Yiheng Li, Yue Gao
arXiv:2609. 12014v1 Announce Type: new Abstract: Safe offline reinforcement learning assumes a cost function on every transition.
By Adam Haroon, Cody Fleming
The paper introduces LEAP-CBF, a safety filter that uses Least‑Effort Adversarial Potentials to quantify how much disturbance effort is needed to cause failure in nonlinear dynamical systems. LEAP serves as a control barrier function for the undisturbed system and can be combined with a robust safety filter that tolerates disturbances with bounded cumulative effort. The authors develop a deep reinforcement learning method to construct LEAPs and validate their effectiveness through simulations of multi‑agent systems and hardware experiments on a quadruped and quadrotors.
By Oswin So, Eric Yu, Chuchu Fan
arXiv:2607.08925v2 Announce Type: replace
Abstract: Training agents directly on the task they will be deployed on often leads to the best policy performance, but training on physical robots can cause...
By Elham Daneshmand, Majid Khadiv, Glen Berseth, Hsiu-Chin Lin
The paper introduces Safety to Competence (S2C), a two‑stage reinforcement learning framework that first learns a safety filter and then trains a competitive task policy while embedding the filter. By separating safety synthesis from task learning, S2C reduces training complexity and prevents the policy from being exploited by adversarial attacks. Experiments on simulated touchdown games show that S2C achieves higher win rates, better Elo ratings, and lower exploitability than eight safe‑RL baselines, and hardware tests confirm its competence against a human opponent.
By Ruihan Wu, Rui Yang, Donggeon David Oh, Duy Nguyen, Haimin Hu
arXiv:2604.07457v2 Announce Type: replace-cross
Abstract: While decoupled control schemes for legged mobile manipulators have shown robustness, learning holistic whole-body control policies for track...
By Ziyang Cheng, Haoyu Wei, Hang Yin, Xiuwei Xu, Bingyao Yu, Jie Zhou, Jiwen Lu
The paper investigates safe reinforcement learning by focusing on the episodic-cost tail rather than just the mean cost. It introduces the use of γαρα_{0.1} (the average cost of the worst 10% of episodes) to classify policies as tail-safe. The authors evaluate standard algorithms on Safety-Gymnasium tasks, identify tail-unsafe policies, and explore constraint families to control tail violations while maintaining return.
By Samuel Tetteh, Cody Fleming