REBot: Reflexive Evasion Robot for Instantaneous Dynamic Obstacle Avoidance
Authors:
- Zihao XuZXZihao Xu
- Ce HaoCHCe Hao
- Chunzheng WangCWChunzheng Wang
- Kuankuan SimaKSKuankuan Sima
- Fan ShiFSFan Shi
- Jin Song DongJDJin Song Dong
Preprint
Abstract
Dynamic obstacle avoidance (DOA) is critical for quadrupedal robots operating in environments with moving obstacles or humans. Existing approaches typically rely on navigation-based trajectory replanning, which assumes sufficient reaction time and leading to fails when obstacles approach rapidly. In such scenarios, quadrupedal robots require reflexive evasion capabilities to perform instantaneous, low-latency maneuvers. This paper introduces Reflexive Evasion Robot (REBot), a control framework that enables quadrupedal robots to achieve real-time reflexive obstacle avoidance. REBot integrates an avoidance policy and a recovery policy within a finite-state machine. With carefully designed learning curricula and by incorporating regularization and adaptive rewards, REBot achieves robust evasion and rapid stabilization in instantaneous DOA tasks. We validate REBot through extensive simulations and real-world experiments, demonstrating notable improvements in avoidance success rates, energy efficiency, and robustness to fast-moving obstacles. Videos and appendix are available on https://rebot-2025.github.io/.Keywords
All Related Versions
This publication has 20 references indexed in Scilit:
- ARMOR: Egocentric Perception for Bimanual Robot Collision Avoidance and Motion Planning2025
- Catch It! Learning to Catch in Flight with Mobile Dexterous Hands2025
- Flying in Highly Dynamic Environments With End-to-End Learning ApproachIEEE Robotics and Automation Letters, 2025
- FAPP: Fast and Adaptive Perception and Planning for UAVs in Dynamic Cluttered EnvironmentsIEEE Transactions on Robotics, 2024
- Mobile Robot Collision Avoidance Based on Deep Reinforcement Learning With Motion ConstraintsIEEE Transactions on Intelligent Vehicles, 2024
- MorAL: Learning Morphologically Adaptive Locomotion Controller for Quadrupedal Robots on Challenging TerrainsIEEE Robotics and Automation Letters, 2024
- Learning-Based Navigation and Collision Avoidance Through Reinforcement for UAVsIEEE Transactions on Aerospace and Electronic Systems, 2023
- Momentum-Aware Trajectory Optimization and Control for Agile Quadrupedal LocomotionIEEE Robotics and Automation Letters, 2022
- RLOC: Terrain-Aware Legged Locomotion Using Reinforcement Learning and Optimal ControlIEEE Transactions on Robotics, 2022
- Adaptive gait planning for quadruped robot based on center of inertia over rough terrainBiomimetic Intelligence and Robotics, 2022