Shield-Loco: Shielding Locomotion Policies with Predictive Safety Filtering

• URL: http://arxiv.org/abs/2606.07193v1
• Date: Fri, 05 Jun 2026 11:59:43 GMT
• Title: Shield-Loco: Shielding Locomotion Policies with Predictive Safety Filtering
• Abstract summary: Large-scale offline safe learning is impractical for covering all edge cases.<n>We propose a predictive safety filter that post-hoc filters the nominal contact locations fed to the RL policy.<n>We validate the filter on a quadruped robot in dense, cluttered environments.
• Score: 10.242244578585897
• License: http://creativecommons.org/licenses/by/4.0/
• 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.

Related papers

• FastBridge: Closing the Model-Based Realization Gap in Safety Filters on 3D Gaussian Splatting for Fast Quadrotor Flight [0.0]
  Current 3DGS safety filters ignore actuator limits and assume commanded accelerations are realized instantaneously.<n>We introduce a nonlinear, actuator-aware safety filter enforced through the full quadrotor dynamics.<n>Compared with a state-of-the-art 3DGS safety filter, our approach reduces trajectory jerk by 47% and runs 2.25 times faster.
  arXiv  Detail & Related papers  (2026-07-01T17:33:01Z)
• Policy Library CBF: Finite-Horizon Safety at Runtime via Parallel Rollouts [11.99821055563793]
  Policy Library Control Barrier Function(PL-CBF) is a runtime safety filter.<n>Policy Library Control Barrier Function(PL-CBF) is a runtime safety filter.
  arXiv  Detail & Related papers  (2026-05-15T19:45:52Z)
• Selective Safety Steering via Value-Filtered Decoding [54.87935112120107]
  Large language models (LLMs) are trained to align with human values, but their generations may still violate safety constraints.<n>Existing decoding-time steering methods often intervene unnecessarily, modifying generations that would have been safe under the base model.<n>We propose a new test-time steering method designed to reduce such unnecessary interventions while improving the safety of unsafe responses.
  arXiv  Detail & Related papers  (2026-05-14T12:13:08Z)
• Layered Safety: Enhancing Autonomous Collision Avoidance via Multistage CBF Safety Filters [18.386652442201665]
  This paper presents a general end-to-end framework for constructing robust and reliable layered safety filters.<n>Given a robot-centric point cloud, we begin by constructing an occupancy map which is used to synthesize a Poisson safety function (PSF)<n>The PSF is employed as a control barrier function (CBF) within two distinct safety filtering stages.
  arXiv  Detail & Related papers  (2026-02-27T22:19:45Z)
• Q-realign: Piggybacking Realignment on Quantization for Safe and Efficient LLM Deployment [55.14890249389052]
  Existing defenses either embed safety recovery into fine-tuning or rely on fine-tuning-derived priors for post-hoc correction.<n>We propose textttQ-realign, a post-hoc defense method based on post-training quantization.<n>Our work provides a practical, turnkey solution for safety-aware deployment.
  arXiv  Detail & Related papers  (2026-01-13T00:07:24Z)
• Rethinking Safety in LLM Fine-tuning: An Optimization Perspective [56.31306558218838]
  We show that poor optimization choices, rather than inherent trade-offs, often cause safety problems, measured as harmful responses to adversarial prompts.<n>We propose a simple exponential moving average (EMA) momentum technique in parameter space that preserves safety performance.<n>Our experiments on the Llama families across multiple datasets demonstrate that safety problems can largely be avoided without specialized interventions.
  arXiv  Detail & Related papers  (2025-08-17T23:46:36Z)
• Shape it Up! Restoring LLM Safety during Finetuning [65.75757313781104]
  Finetuning large language models (LLMs) enables user-specific customization but introduces critical safety risks.<n>We propose dynamic safety shaping (DSS), a framework that uses fine-grained safety signals to reinforce learning from safe segments of a response while suppressing unsafe content.<n>We present STAR-DSS, guided by STAR scores, that robustly mitigates finetuning risks and delivers substantial safety improvements across diverse threats, datasets, and model families.
  arXiv  Detail & Related papers  (2025-05-22T18:05:16Z)
• Modular Control Architecture for Safe Marine Navigation: Reinforcement Learning and Predictive Safety Filters [0.0]
  Reinforcement learning is increasingly used to adapt to complex scenarios, but standard frameworks ensuring safety and stability are lacking. Predictive Safety Filters (PSF) offer a promising solution, ensuring constraint satisfaction in learning-based control without explicit constraint handling. We apply this approach to marine navigation, combining RL with PSF on a simulated Cybership II model. Results demonstrate the PSF's effectiveness in maintaining safety without hindering the RL agent's learning rate and performance, evaluated against a standard RL agent without PSF.
  arXiv  Detail & Related papers  (2023-12-04T12:37:54Z)
• Learning Predictive Safety Filter via Decomposition of Robust Invariant Set [6.94348936509225]
  This paper presents advantages of both RMPC and RL RL to synthesize safety filters for nonlinear systems. We propose a policy approach for robust reach problems and establish its complexity.
  arXiv  Detail & Related papers  (2023-11-12T08:11:28Z)
• A Multiplicative Value Function for Safe and Efficient Reinforcement Learning [131.96501469927733]
  We propose a safe model-free RL algorithm with a novel multiplicative value function consisting of a safety critic and a reward critic. The safety critic predicts the probability of constraint violation and discounts the reward critic that only estimates constraint-free returns. We evaluate our method in four safety-focused environments, including classical RL benchmarks augmented with safety constraints and robot navigation tasks with images and raw Lidar scans as observations.
  arXiv  Detail & Related papers  (2023-03-07T18:29:15Z)
• Online Safety Property Collection and Refinement for Safe Deep Reinforcement Learning in Mapless Navigation [79.89605349842569]
  We introduce the Collection and Refinement of Online Properties (CROP) framework to design properties at training time. CROP employs a cost signal to identify unsafe interactions and use them to shape safety properties. We evaluate our approach in several robotic mapless navigation tasks and demonstrate that the violation metric computed with CROP allows higher returns and lower violations over previous Safe DRL approaches.
  arXiv  Detail & Related papers  (2023-02-13T21:19:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.