DynaWM: Dynamics-Aware Distillation with World Model and Momentum Targets for Smooth Locomotion over Continuous Stairs

• URL: http://arxiv.org/abs/2606.24089v1
• Date: Tue, 23 Jun 2026 03:07:42 GMT
• Title: DynaWM: Dynamics-Aware Distillation with World Model and Momentum Targets for Smooth Locomotion over Continuous Stairs
• Abstract summary: We propose DynaWM, a dynamics-aware representation learning framework.<n>We introduce a world model as a regularizer to enforce forward-dynamics awareness, preserving comprehensive terrain geometry.<n>We employ a momentum target encoder to provide consistent distillation targets, preventing dimensional collapse from non-stationary teacher updates.
• Score: 8.383969829401517
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent advances in control have enabled bipedal-wheeled robots to traverse slopes and single-step obstacles, yet long staircase traversal remains challenging as current teacher-student frameworks suffer from weakened dynamics-aware representations and incomplete terrain geometry encoding. To bridge this gap, we propose DynaWM, a dynamics-aware representation learning framework. To enhance terrain encoding capability and enable transparent assessment, we introduce a world model as a regularizer to enforce forward-dynamics awareness, preserving comprehensive terrain geometry while facilitating hierarchical encoding visualization. To stabilize knowledge transfer, we employ a momentum target encoder to provide consistent distillation targets, preventing dimensional collapse from non-stationary teacher updates. Evaluation of the learned representations through Principal Component Analysis (PCA) visualization and quantitative metrics reveals that our encoder hierarchically captures terrain geometry with higher terrain encoding capability, leading to enhanced terrain adaptability and motion smoothness. Experimental results in simulation and real hardware demonstrate that our method achieves superior terrain adaptability and motion smoothness, enabling bipedal-wheeled robots to overcome diverse continuous stairs, as shown in Fig. 1.

Related papers

• WorldDirector: Building Controllable World Simulators with Persistent Dynamic Memory [106.28234880163723]
  WorldDirector is a highly controllable video world model framework.<n>Our framework explicitly decouples semantic motion orchestration from visual generation.<n>We successfully preserve the exact visual identities of dynamic entities even when they re-enter the scene after prolonged periods out of view.
  arXiv  Detail & Related papers  (2026-07-02T17:59:59Z)
• ManiSplat: Manipulation Trajectory Synthesis from Monocular Video via Decoupled 3D Gaussian Splatting [28.256286705954846]
  Reconstructing dynamic and interactive 3D scenes from real-world observations remains a fundamental challenge in computer vision and robotics.<n>We introduce ManiSplat, a unified framework that reconstructs controllable and decoupled Gaussian digital twins directly from robotic videos.<n>Our approach reconstructs interaction-driven dynamic scenes with high fidelity and controllability, effectively supporting downstream robotic tasks and policy learning.
  arXiv  Detail & Related papers  (2026-06-09T09:55:58Z)
• Kinematics-Aware Latent World Models for Data-Efficient Autonomous Driving [8.391716372682316]
  Vehicle kinematic information is incorporated into the observation encoder to ground latent transitions in physically meaningful motion dynamics.<n> geometry-aware supervision regularizes the RSSM latent state to capture task-relevant spatial structure beyond pixel reconstruction.
  arXiv  Detail & Related papers  (2026-03-07T15:47:54Z)
• MeshMimic: Geometry-Aware Humanoid Motion Learning through 3D Scene Reconstruction [54.36564144414704]
  MeshMimic is an innovative framework that bridges 3D scene reconstruction and embodied intelligence to enable humanoid robots to learn coupled "motion-terrain" interactions directly from video.<n>By leveraging state-of-the-art 3D vision models, our framework precisely segments and reconstructs both human trajectories and the underlying 3D geometry of terrains and objects.
  arXiv  Detail & Related papers  (2026-02-17T17:09:45Z)
• Self-supervised Physics-Informed Manipulation of Deformable Linear Objects with Non-negligible Dynamics [14.801939035043352]
  SPiD is a physics-informed self-supervised learning framework for dynamic manipulation of deformable linear objects.<n>We train a neural controller using a task-oriented cost, enabling end-to-end optimization through interaction with the differentiable object model.<n>We develop an affordable markerless rope perception method and demonstrate that our controller maintains performance with noisy and low-frequency state updates.
  arXiv  Detail & Related papers  (2026-02-03T15:14:09Z)
• Learning Terrain Aware Bipedal Locomotion via Reduced Dimensional Perceptual Representations [6.407946291544721]
  This work introduces a hierarchical strategy for terrain-aware bipedal locomotion.<n>It integrates reduced-dimensional representations to enhance reinforcement learning (RL)-based high-level (HL) policies for real-time gait generation.
  arXiv  Detail & Related papers  (2025-12-15T05:38:08Z)
• MoSiC: Optimal-Transport Motion Trajectory for Dense Self-Supervised Learning [66.53533434848369]
  We propose a motion-guided self-learning framework that learns densely consistent representations.<n>We improve state-of-the-art by 1% to 6% on six image and video datasets and four evaluation benchmarks.
  arXiv  Detail & Related papers  (2025-06-10T11:20:32Z)
• GeoDrive: 3D Geometry-Informed Driving World Model with Precise Action Control [50.67481583744243]
  We introduce GeoDrive, which explicitly integrates robust 3D geometry conditions into driving world models.<n>We propose a dynamic editing module during training to enhance the renderings by editing the positions of the vehicles.<n>Our method significantly outperforms existing models in both action accuracy and 3D spatial awareness.
  arXiv  Detail & Related papers  (2025-05-28T14:46:51Z)
• Gaussian Splatting to Real World Flight Navigation Transfer with Liquid Networks [93.38375271826202]
  We present a method to improve generalization and robustness to distribution shifts in sim-to-real visual quadrotor navigation tasks. We first build a simulator by integrating Gaussian splatting with quadrotor flight dynamics, and then, train robust navigation policies using Liquid neural networks. In this way, we obtain a full-stack imitation learning protocol that combines advances in 3D Gaussian splatting radiance field rendering, programming of expert demonstration training data, and the task understanding capabilities of Liquid networks.
  arXiv  Detail & Related papers  (2024-06-21T13:48:37Z)
• TrackDiffusion: Tracklet-Conditioned Video Generation via Diffusion Models [75.20168902300166]
  We propose TrackDiffusion, a novel video generation framework affording fine-grained trajectory-conditioned motion control. A pivotal component of TrackDiffusion is the instance enhancer, which explicitly ensures inter-frame consistency of multiple objects. generated video sequences by our TrackDiffusion can be used as training data for visual perception models.
  arXiv  Detail & Related papers  (2023-12-01T15:24:38Z)
• Learning to Jump from Pixels [23.17535989519855]
  We present Depth-based Impulse Control (DIC), a method for synthesizing highly agile visually-guided behaviors. DIC affords the flexibility of model-free learning but regularizes behavior through explicit model-based optimization of ground reaction forces. We evaluate the proposed method both in simulation and in the real world.
  arXiv  Detail & Related papers  (2021-10-28T17:53:06Z)

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.