Learning to Recover: Dynamic Reward Shaping with Wheel-Leg Coordination for Fallen Robots

• URL: http://arxiv.org/abs/2506.05516v2
• Date: Wed, 08 Oct 2025 13:33:28 GMT
• Title: Learning to Recover: Dynamic Reward Shaping with Wheel-Leg Coordination for Fallen Robots
• Abstract summary: This paper presents a learning-based framework integrating Episode-based Dynamic Reward Shaping and curriculum learning.<n>We demonstrate that synergistic wheel-leg coordination reduces joint torque consumption by 15.8% and 26.2%.
• Score: 10.712715061996194
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Adaptive recovery from fall incidents are essential skills for the practical deployment of wheeled-legged robots, which uniquely combine the agility of legs with the speed of wheels for rapid recovery. However, traditional methods relying on preplanned recovery motions, simplified dynamics or sparse rewards often fail to produce robust recovery policies. This paper presents a learning-based framework integrating Episode-based Dynamic Reward Shaping and curriculum learning, which dynamically balances exploration of diverse recovery maneuvers with precise posture refinement. An asymmetric actor-critic architecture accelerates training by leveraging privileged information in simulation, while noise-injected observations enhance robustness against uncertainties. We further demonstrate that synergistic wheel-leg coordination reduces joint torque consumption by 15.8% and 26.2% and improves stabilization through energy transfer mechanisms. Extensive evaluations on two distinct quadruped platforms achieve recovery success rates up to 99.1% and 97.8% without platform-specific tuning. The supplementary material is available at https://boyuandeng.github.io/L2R-WheelLegCoordination/

Related papers

• Learning Fault-Tolerant Locomotion with Adaptive Gait Timing [1.1156827035309405]
  Hardware failures require legged robots to reorganize coordination and gait timing to maintain stability and mobility.<n>We propose a deep reinforcement learning approach for fault-tolerant locomotion under actuator power loss.<n>The approach is validated in high-fidelity simulation on uneven terrain and real-world experiments on flat ground using a 68 kg quadruped robot.
  arXiv  Detail & Related papers  (2026-08-07T15:27:03Z)
• EgoRecovery: Acquiring Failure Recovery Ability Through Human Recovery Demonstration [75.13534797264485]
  We show that egocentric human data capturing failure recovery processes provides a scalable alternative to robot teleoperation.<n>By efficiently arranging task-level failure configurations and recording short recovery segments, human operators can generate more than 10x as much valid recovery data per hour.
  arXiv  Detail & Related papers  (2026-07-22T04:44:26Z)
• Recover, Discover, Plan: Learning Skills and Concepts from Robot Failures [50.668639270656115]
  ReSYNC is a robot that learns skills and concepts and refines its abstraction library to support abstract planning.<n>We show that ReSYNC's ability to continually expand and refine its abstraction library allows it to solve long-horizon, previously unseen problems.<n>Overall, ReSYNC represents a significant step toward robots that autonomously acquire abstractions for scalable, failure-aware planning.
  arXiv  Detail & Related papers  (2026-06-16T17:59:57Z)
• Robust Fall Recovery for Armless Bipedal-Wheeled Robots Via Force-Guided Learning [13.56238998710442]
  Some humanoids and quadrupeds are capable of fall recovery from diverse postures by utilizing arms or coordinating multi-legs to generate support forces.<n>Without arms or other legs to provide supportive assistance, a bi-wheeled robot must rely solely on the actuation of its legs.<n>We introduce FTSR (Force-guided Teacher-student framework with Stage-wise Rewards) to address this problem.
  arXiv  Detail & Related papers  (2026-06-12T08:51:51Z)
• Stubborn: A Streamlined and Unified Reinforcement Learning Framework for Robust Motion Tracking and Fall Recovery for Humanoids [9.395015640853973]
  We propose Stubborn, a streamlined and unified reinforcement learning framework to achieve robust humanoid motion tracking and fall recovery.<n>Specifically, Stubborn uses an asymmetric Actor-Critic architecture and consists of three major components.<n>For real-world demonstrations, please refer to https://aislab-sustech.io/Stubborn/.
  arXiv  Detail & Related papers  (2026-06-11T02:13:53Z)
• Embedding Classical Balance Control Principles in Reinforcement Learning for Humanoid Recovery [1.0883174135300417]
  Humans are vulnerable to falls and unrecoverable failure states, limiting their practical utility in unstructured environments.<n>We present a unified RL policy that addresses this limitation by embedding classical balance metrics.<n>The policy achieves a 93.4% recovery rate across randomized initial poses and unscripted fall configurations.
  arXiv  Detail & Related papers  (2026-03-09T17:02:30Z)
• A data-physics hybrid generative model for patient-specific post-stroke motor rehabilitation using wearable sensor data [17.28369129744061]
  We develop a data-physics hybrid generative framework that reconstructs an individual stroke survivor's neuromuscular control.<n>In 11 stroke survivors, the personalized controllers preserved idiosyncratic gait patterns while improving joint-angle and endpoint fidelity.
  arXiv  Detail & Related papers  (2025-12-16T11:55:11Z)
• RobotDancing: Residual-Action Reinforcement Learning Enables Robust Long-Horizon Humanoid Motion Tracking [50.200035833530876]
  RobotDancing is a simple, scalable framework that predicts residual joint targets to explicitly correct dynamics discrepancies.<n>It can track multi-minute, high-energy behaviors (jumps, spins, cartwheels) and deploys zero-shot to hardware with high motion tracking quality.
  arXiv  Detail & Related papers  (2025-09-25T03:30:34Z)
• Complementary Learning System Empowers Online Continual Learning of Vehicle Motion Forecasting in Smart Cities [50.14230518748104]
  We introduce Dual-LS, a task-free, online continual learning paradigm for deep neural network (DNN)-based motion forecasting.<n>Tests on naturalistic data spanning three countries, over 772,000 vehicles and cumulative testing mileage of 11,187 km show that Dual-LS mitigates catastrophic forgetting by up to 74.31%.
  arXiv  Detail & Related papers  (2025-08-27T06:19:21Z)
• RCR-AF: Enhancing Model Generalization via Rademacher Complexity Reduction Activation Function [28.98307539597017]
  This paper investigates activation functions as a crucial yet underexplored component for enhancing model robustness.<n>We propose a Rademacher Complexity Reduction Activation Function (RCR-AF), a novel activation function designed to improve both generalization and adversarial resilience.
  arXiv  Detail & Related papers  (2025-07-30T07:45:03Z)
• Bipedalism for Quadrupedal Robots: Versatile Loco-Manipulation through Risk-Adaptive Reinforcement Learning [21.938067330028066]
  We introduce bipedalism for quadrupedal robots, freeing the front legs for versatile interactions with the environment.<n>We propose a risk-adaptive distributional Reinforcement Learning framework designed for quadrupedal robots walking on their hind legs.
  arXiv  Detail & Related papers  (2025-07-27T18:51:34Z)
• ReCoM: Realistic Co-Speech Motion Generation with Recurrent Embedded Transformer [58.49950218437718]
  We present ReCoM, an efficient framework for generating high-fidelity and generalizable human body motions synchronized with speech.<n>The core innovation lies in the Recurrent Embedded Transformer (RET), which integrates Dynamic Embedding Regularization (DER) into a Vision Transformer (ViT) core architecture.<n>To enhance model robustness, we incorporate the proposed DER strategy, which equips the model with dual capabilities of noise resistance and cross-domain generalization.
  arXiv  Detail & Related papers  (2025-03-27T16:39:40Z)
• Spatial-Temporal Graph Diffusion Policy with Kinematic Modeling for Bimanual Robotic Manipulation [88.83749146867665]
  Existing approaches learn a policy to predict a distant next-best end-effector pose.<n>They then compute the corresponding joint rotation angles for motion using inverse kinematics.<n>We propose Kinematics enhanced Spatial-TemporAl gRaph diffuser.
  arXiv  Detail & Related papers  (2025-03-13T17:48:35Z)
• Humanoid Whole-Body Locomotion on Narrow Terrain via Dynamic Balance and Reinforcement Learning [54.26816599309778]
  We propose a novel whole-body locomotion algorithm based on dynamic balance and Reinforcement Learning (RL)<n> Specifically, we introduce a dynamic balance mechanism by leveraging an extended measure of Zero-Moment Point (ZMP)-driven rewards and task-driven rewards in a whole-body actor-critic framework.<n> Experiments conducted on a full-sized Unitree H1-2 robot verify the ability of our method to maintain balance on extremely narrow terrains.
  arXiv  Detail & Related papers  (2025-02-24T14:53:45Z)
• LoRA-IR: Taming Low-Rank Experts for Efficient All-in-One Image Restoration [62.3751291442432]
  We propose LoRA-IR, a flexible framework that dynamically leverages compact low-rank experts to facilitate efficient all-in-one image restoration. LoRA-IR consists of two training stages: degradation-guided pre-training and parameter-efficient fine-tuning. Experiments demonstrate that LoRA-IR achieves SOTA performance across 14 IR tasks and 29 benchmarks, while maintaining computational efficiency.
  arXiv  Detail & Related papers  (2024-10-20T13:00:24Z)
• Back-stepping Experience Replay with Application to Model-free Reinforcement Learning for a Soft Snake Robot [15.005962159112002]
  Back-stepping Experience Replay (BER) is compatible with arbitrary off-policy reinforcement learning algorithms. We present an application of BER in a model-free RL approach for the locomotion and navigation of a soft snake robot.
  arXiv  Detail & Related papers  (2024-01-21T02:17:16Z)
• Sparse MoE as the New Dropout: Scaling Dense and Self-Slimmable Transformers [107.3726071306935]
  We propose a new plug-and-play training framework, SMoE-Dropout, to enable scaling transformers to better accuracy in their full capacity without collapse. SMoE-Dropout consists of a randomly and fixed router network to activate experts and gradually increases the activated expert number as training progresses over time. Our experiments demonstrate the superior performance and substantial computation savings of SMoE-Dropout, compared to dense training baselines with equivalent parameter counts.
  arXiv  Detail & Related papers  (2023-03-02T22:12:51Z)
• VAE-Loco: Versatile Quadruped Locomotion by Learning a Disentangled Gait Representation [78.92147339883137]
  We show that it is pivotal in increasing controller robustness by learning a latent space capturing the key stance phases constituting a particular gait. We demonstrate that specific properties of the drive signal map directly to gait parameters such as cadence, footstep height and full stance duration. The use of a generative model facilitates the detection and mitigation of disturbances to provide a versatile and robust planning framework.
  arXiv  Detail & Related papers  (2022-05-02T19:49:53Z)

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.