aiqbench analysis
TechPulse take: Shield-Loco addresses a critical gap in reinforcement learning for legged robots by introducing a predictive safety filter that mitigates collisions without degrading task performance. The approach combines a sampling-based optimizer with a learned value function to handle discontinuous contact landscapes in real-world, cluttered environments. This method offers a practical solution for deploying RL policies safely where traditional offline training or conservative recovery controllers fall short.
Shield-Loco: Shielding Locomotion Policies with Predictive Safety Filtering
arXiv:2606.07193v1 Announce Type: new Abstract: Reinforcement learning (RL) policies enable dynamic legged locomotion but lack mechanisms to avoid violations of safety constraints that are absent during training. Large-scale offline safe learning is impractical for covering all edge cases. Existing safety frameworks either rely on reduced-order models that cannot reason about whole-body behaviors or require conservative recovery controllers that degrade task performance. We propose a predictive safety filter that post-hoc filters the nominal contact locations fed to the RL policy. When a collision is predicted, a sampling-based optimizer asynchronously searches for safer contact sequences using a full-physics model, while a learned value function bootstraps long-horizon returns. Our three algorithmic components (geometric projection of sampled contacts, momentum-augmented updates, and replica-exchange) make the optimization tractable in a discontinuous contact landscape. We validate the filter on a quadruped robot in dense, cluttered environments, both in simulation and in the real world, showing substantial reductions in safety violations with minimal deviation from the nominal input.
Abstract:Reinforcement learning (RL) policies enable dynamic legged locomotion but lack mechanisms to avoid violations of safety constraints that are absent during training. Large-scale offline safe learning is impractical for covering all edge cases. Existing safety frameworks either rely on reduced-order models that cannot reason about whole-body behaviors or require conservative recovery controllers that degrade task performance. We propose a predictive safety filter that post-hoc filters the nominal contact locations fed to the RL policy. When a collision is predicted, a sampling-based optimizer asynchronously searches for safer contact sequences using a full-physics model, while a learned value function bootstraps long-horizon returns. Our three algorithmic components (geometric projection of sampled contacts, momentum-augmented updates, and replica-exchange) make the optimization tractable in a discontinuous contact landscape. We validate the filter on a quadruped robot in dense, cluttered environments, both in simulation and in the real world, showing substantial reductions in safety violations with minimal deviation from the nominal input.
Robotics (cs.RO)
arXiv:2606.07193 [cs.RO]
(or arXiv:2606.07193v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2606.07193 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 248 words
Related coverage
Safe Reinforcement Learning with Receding-Horizon Safety Critic and Differentiable Action Shields
arXiv · Oct 15, 2026
Predictive Safety Curricula for Robust Legged Locomotion
arXiv · Sep 30, 2026
Calibrated Predictive Safety for Heterogeneous Robots: An Action-Conditioned JEPA Framework with Model-Based Safety Shields
arXiv · Aug 19, 2026
Backup-Based Safety Filters: A Comparative Review of Backup CBF, Model Predictive Shielding, and gatekeeper
arXiv · Sep 21, 2026