mlsanityData to 7 Oct 2026

Agile But Safe: Learning Collision-Free High-Speed Legged Locomotion

arXiv 2401.17583 ↗First posted 31 Jan 2024cs.RO cs.AI cs.CVcited by 53

Not on the map. Why?

Abstract (arXiv)

Legged robots navigating cluttered environments must be jointly agile for efficient task execution and safe to avoid collisions with obstacles or humans. Existing studies either develop conservative controllers (< 1.0 m/s) to ensure safety, or focus on agility without considering potentially fatal collisions. This paper introduces Agile But Safe (ABS), a learning-based control framework that enables agile and collision-free locomotion for quadrupedal robots. ABS involves an agile policy to execute agile motor skills amidst obstacles and a recovery policy to prevent failures, collaboratively achieving high-speed and collision-free navigation. The policy switch in ABS is governed by a learned control-theoretic reach-avoid value network, which also guides the recovery policy as an objective function, thereby safeguarding the robot in a closed loop. The training process involves the learning of the agile policy, the reach-avoid value network, the recovery policy, and an exteroception representation network, all in simulation. These trained modules can be directly deployed in the real world with onboard sensing and computation, leading to high-speed and collision-free navigation in confined indoor and outdoor spaces with both static and dynamic obstacles.

Key facts. Cited by 53 papers in the map (data through 7 Oct 2026). Builds on 64 of its 93 references that are in the map. First posted on arXiv 31 Jan 2024.

Builds on

64 of 93 references.

References · 64

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2017Proximal Policy Optimization Algorithmscited by 14,441
  3. 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,401
  4. 2021Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learningcited by 882
  5. 2020Learning Quadrupedal Locomotion over Challenging Terraincited by 494
  6. 2021RMA: Rapid Motor Adaptation for Legged Robotscited by 430
  7. 2019Learning agile and dynamic motor skills for legged robotscited by 402
  8. 2018Reward Constrained Policy Optimizationcited by 397
  9. 2021Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learningcited by 395
  10. 2022Learning robust perceptive locomotion for quadrupedal robots in the wildcited by 299
  11. 2019End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Taskscited by 256
  12. 2018Safe Exploration in Continuous Action Spacescited by 237
38 more
  1. 2017Hamilton-Jacobi Reachability: A Brief Overview and Recent Advancescited by 186
  2. 2020Recovery RL: Safe Reinforcement Learning with Learned Recovery Zonescited by 156
  3. 2022Goal-Conditioned Reinforcement Learning: Problems and Solutionscited by 145
  4. 2022Rapid Locomotion via Reinforcement Learningcited by 130
  5. 2023Extreme Parkour with Legged Robotscited by 120
  6. 2023Robot Parkour Learningcited by 115
  7. 2023ANYmal Parkour: Learning Agile Navigation for Quadrupedal Robotscited by 105
  8. 2020DeepReach: A Deep Learning Approach to High-Dimensional Reachabilitycited by 81
  9. 2023Reaching the Limit in Autonomous Racing: Optimal Control versus Reinforcement Learningcited by 72
  10. 2021Learning Vision-Guided Quadrupedal Locomotion End-to-End with Cross-Modal Transformerscited by 71
  11. 2020Multi-expert learning of adaptive legged locomotioncited by 62
  12. 2018Accelerated Primal-Dual Policy Optimization for Safe Reinforcement Learningcited by 61
  13. 2023State-wise Safe Reinforcement Learning: A Surveycited by 48
  14. 2020RLOC: Terrain-Aware Legged Locomotion using Reinforcement Learning and Optimal Controlcited by 46
  15. 2021Safety and Liveness Guarantees through Reach-Avoid Reinforcement Learningcited by 43
  16. 2014Reach-Avoid Problems with Time-Varying Dynamics, Targets and Constraintscited by 41
  17. 2023How to Train Your Neural Control Barrier Function: Learning Safety Filters for Complex Input-Constrained Systemscited by 40
  18. 2022Advanced Skills by Learning Locomotion and Local Navigation End-to-Endcited by 39
  19. 2023ASC: Adaptive Skill Coordination for Robotic Mobile Manipulationcited by 38
  20. 2023DTC: Deep Tracking Controlcited by 37
  21. 2023Emergence of Maps in the Memories of Blind Navigation Agentscited by 31
  22. 2021Learning a State Representation and Navigation in Cluttered and Dynamic Environmentscited by 30
  23. 2023Learning Vision-Based Bipedal Locomotion for Challenging Terraincited by 30
  24. 2022Safe Reinforcement Learning for Legged Locomotioncited by 29
  25. 2022Rethinking Sim2Real: Lower Fidelity Simulation Leads to Higher Sim2Real Transfer in Navigationcited by 28
  26. 2022Perceptive Locomotion through Nonlinear Model Predictive Controlcited by 27
  27. 2023Robust and Versatile Bipedal Jumping Control through Reinforcement Learningcited by 27
  28. 2022Sim-to-Lab-to-Real: Safe Reinforcement Learning with Shielding and Generalization Guaranteescited by 24
  29. 2023Learning Arm-Assisted Fall Damage Reduction and Recovery for Legged Mobile Manipulatorscited by 20
  30. 2022ViNL: Visual Navigation and Locomotion Over Obstaclescited by 19
  31. 2023Resilient Legged Local Navigation: Learning to Traverse with Compromised Perception End-to-Endcited by 16
  32. 2020Chance-Constrained Trajectory Optimization for Safe Exploration and Learning of Nonlinear Systemscited by 15
  33. 2022A Collision-Free MPC for Whole-Body Dynamic Locomotion and Manipulationcited by 15
  34. 2022Learning and Deploying Robust Locomotion Policies with Minimal Dynamics Randomizationcited by 15
  35. 2023Learning Agile Locomotion on Risky Terrainscited by 15
  36. 2022An Efficient Locally Reactive Controller for Safe Navigation in Visual Teach and Repeat Missionscited by 12
  37. 2023CAJun: Continuous Adaptive Jumping using a Learned Centroidal Controllercited by 12
  38. 2023IndoorSim-to-OutdoorReal: Learning to Navigate Outdoors without any Outdoor Experiencecited by 11

50 most cited of 64 All 64 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 17; 2025: 25; 2026: 11.

53 papers.

Sort:

By area:

  1. 2024OmniH2O: Universal and Dexterous Human-to-Humanoid Whole-Body Teleoperation and Learningcited by 167
  2. 2024Deep Reinforcement Learning for Robotics: A Survey of Real-World Successescited by 153
  3. 2024Learning Human-to-Humanoid Real-Time Whole-Body Teleoperationcited by 112
  4. 2025Generalizing Safety Beyond Collision-Avoidance via Latent-Space Reachability Analysiscited by 44
  5. 2025Multi-agent Embodied AI: Advances and Future Directionscited by 36
  6. 2024One Policy to Run Them All: an End-to-end Learning Approach to Multi-Embodiment Locomotioncited by 34
  7. 2025BeamDojo: Learning Agile Humanoid Locomotion on Sparse Footholdscited by 32
  8. 2025Uncertainty-aware Latent Safety Filters for Avoiding Out-of-Distribution Failurescited by 30
  9. 2025MoE-Loco: Mixture of Experts for Multitask Locomotioncited by 17
  10. 2025Learning coordinated badminton skills for legged manipulatorscited by 16
  11. 2024Commonsense Reasoning for Legged Robot Adaptation with Vision-Language Modelscited by 15
  12. 2024Gameplay Filters: Robust Zero-Shot Safety through Adversarial Imaginationcited by 12
38 more
  1. 2025Towards Embodiment Scaling Laws in Robot Locomotioncited by 12
  2. 2024CaT: Constraints as Terminations for Legged Locomotion Reinforcement Learningcited by 11
  3. 2025Manipulation as in Simulation: Enabling Accurate Geometry Perception in Robotscited by 11
  4. 2024OASIS: Conditional Distribution Shaping for Offline Safe Reinforcement Learningcited by 10
  5. 2024SoloParkour: Constrained Reinforcement Learning for Visual Locomotion from Privileged Experiencecited by 10
  6. 2024SafeDrive: Knowledge- and Data-Driven Risk-Sensitive Decision-Making for Autonomous Vehicles with Large Language Modelscited by 10
  7. 2024One Filter to Deploy Them All: Robust Safety for Quadrupedal Navigation in Unknown Environmentscited by 8
  8. 2025How to Train Your Latent Control Barrier Function: Smooth Safety Filtering Under Hard-to-Model Constraintscited by 7
  9. 2026Failure-Aware RL: Reliable Offline-to-Online Reinforcement Learning with Self-Recovery for Real-World Manipulationcited by 6
  10. 2024MAGICS: Adversarial RL with Minimax Actors Guided by Implicit Critic Stackelberg for Convergent Neural Synthesis of Robot Safetycited by 5
  11. 2025AnySafe: Adapting Latent Safety Filters at Runtime via Safety Constraint Parameterization in the Latent Spacecited by 5
  12. 2025A Real-to-Sim-to-Real Approach to Robotic Manipulation with VLM-Generated Iterative Keypoint Rewardscited by 4
  13. 2025Learning Conservative Neural Control Barrier Functions from Offline Datacited by 4
  14. 2025FACET: Force-Adaptive Control via Impedance Reference Tracking for Legged Robotscited by 4
  15. 2025AcL: Action Learner for Fault-Tolerant Quadruped Locomotion Controlcited by 3
  16. 2025Constrained Decoding for Safe Robot Navigation Foundation Modelscited by 3
  17. 2025MM-Nav: Multi-View VLA Model for Robust Visual Navigation via Multi-Expert Learningcited by 3
  18. 2024TOP-Nav: Legged Navigation Integrating Terrain, Obstacle and Proprioception Estimationcited by 2
  19. 2024GRAM: Generalization in Deep RL with a Robust Adaptation Modulecited by 2
  20. 2026PAC-MAN: Perception-Aware CBF-RL for Whole-Body Safety in Humanoid Dodgeballcited by 2
  21. 2024QuadWBG: Generalizable Quadrupedal Whole-Body Graspingcited by 1
  22. 2024Q-learning-based Model-free Safety Filtercited by 1
  23. 2025QuietPaw: Learning Quadrupedal Locomotion with Versatile Noise Preference Alignmentcited by 1
  24. 2025Toward Real-World Cooperative and Competitive Soccer with Quadrupedal Robot Teamscited by 1
  25. 2025Sight Over Site: Perception-Aware Reinforcement Learning for Efficient Robotic Inspectioncited by 1
  26. 2025Flexible Locomotion Learning with Diffusion Model Predictive Controlcited by 1
  27. 2025From Refusal to Recovery: A Control-Theoretic Approach to Generative AI Guardrailscited by 1
  28. 2026D-REX: Differentiable Real-to-Sim-to-Real Engine for Learning Dexterous Graspingcited by 1
  29. 2026When Is a Learned Command Adapter Worth It? Closed-Loop Identification and Counterfactual Auditing of Frozen Locomotion Policiescited by 1
  30. 2026Trajectory-Level Automatic Curriculum Learning for Legged Locomotion on Unstructured Terraincited by 1
  31. 2024Reinforcement Learning from Wild Animal Videoscited by 0
  32. 2025Watch Less, Feel More: Sim-to-Real RL for Generalizable Articulated Object Manipulation via Motion Adaptation and Impedance Controlcited by 0
  33. 2025Safe Beyond the Horizon: Efficient Sampling-based MPC with Neural Control Barrier Functionscited by 0
  34. 2025Quadrupedal Robot Skateboard Mounting via Reverse Curriculum Learningcited by 0
  35. 2025LEARN: Learning End-to-End Aerial Resource-Constrained Multi-Robot Navigationcited by 0
  36. 2026Co-jump: Cooperative Jumping with Quadrupedal Robots via Multi-Agent Reinforcement Learningcited by 0
  37. 2026SAGE-LLM: Towards Safe and Generalizable LLM Controller with Fuzzy-CBF Verification and Graph-Structured Knowledge Retrieval for UAV Decisioncited by 0
  38. 2026Mind Your Steps: A General Learning Framework for Accurate Humanoid Foothold Trackingcited by 0

50 most cited of 53 All 53 (JSON)