LatentMimic: Terrain-Adaptive Locomotion via Latent Space Imitation

• URL: http://arxiv.org/abs/2604.16440v1
• Date: Tue, 07 Apr 2026 03:49:15 GMT
• Title: LatentMimic: Terrain-Adaptive Locomotion via Latent Space Imitation
• Abstract summary: We introduce LatentMimic, a novel locomotion learning framework that decouples stylistic fidelity from geometric constraints.<n>By minimizing the marginal latent divergence between the policy's state-action distribution and a learned mocap prior, our approach provides a conditional relaxation of rigid pose-tracking objectives.<n>We validate our method across four locomotion styles and four terrains, demonstrating that LatentMimic enables effective terrain-adaptive locomotion.
• Score: 22.247608050046125
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Developing natural and diverse locomotion controllers for quadruped robots that can adapt to complex terrains while preserving motion style remains a significant challenge. Existing imitation-based methods face a fundamental optimization trade-off: strict adherence to motion capture (mocap) references penalizes the geometric deviations required for terrain adaptability, whereas terrain-centric policies often compromise stylistic fidelity. We introduce LatentMimic, a novel locomotion learning framework that decouples stylistic fidelity from geometric constraints. By minimizing the marginal latent divergence between the policy's state-action distribution and a learned mocap prior, our approach provides a conditional relaxation of rigid pose-tracking objectives. This formulation preserves gait topology while permitting independent end-effector adaptations for irregular terrains. We further introduce a terrain adaptation module with a dynamic replay buffer to resolve the policy's distribution shifts across different terrains. We validate our method across four locomotion styles and four terrains, demonstrating that LatentMimic enables effective terrain-adaptive locomotion, achieving higher terrain traversal success rates than state-of-the-art motion-tracking methods while maintaining high stylistic fidelity.

Related papers

• MILD: Tractable Terrain Modeling for Learning Improved Bipedal Locomotion on Deformable Surfaces [27.312811759047023]
  We present MILD, featuring a physics-grounded discrete-element contact solver.<n>We train a terrain-aware controller via deep reinforcement learning.<n>We demonstrate the system's capability online identification and adaptation across a wide range of surface stiffness.
  arXiv  Detail & Related papers  (2026-08-20T12:24:34Z)
• AutoPath: Learning Transferable Goal-Conditioned Stochastic Path Prior for Safe Navigation Without Human Demonstrations [103.61864552260316]
  Real-time navigation in dynamic environments requires collision-free and dynamically feasible motion under limited perception.<n>We formulate navigation as learning a transferable goal-aligned path prior that models a reusable distribution over goal-aligned geometry-consistent local paths conditioned on local observations.<n>We develop a structured prior learning framework that parameterizes local paths using a geometry-aware polar action manifold and incorporates risk-sensitive utility shaping with multi-goal distributional rollouts for stable and safety-aware planning.
  arXiv  Detail & Related papers  (2026-07-13T16:00:41Z)
• Mind Your Steps: A General Learning Framework for Accurate Humanoid Foothold Tracking [78.30338456206984]
  We introduce a novel framework for training general-purpose 3D foothold-tracking policies.<n>By dynamically providing footstep support through a goal sampler, this method enables the learned policy to be agnostic to specific terrains.<n>Our new target representation effectively mitigates challenges arising in the real world, such as noisy and inaccurate pose estimation and foot contact estimation.
  arXiv  Detail & Related papers  (2026-06-06T16:44:56Z)
• T-GMP: Terrain-conditioned Generative Motion Priors for Versatile and Natural Humanoid Locomotion [10.258225375602231]
  We propose a module that captures a terrain-conditioned latent motion manifold from a few expert state-terrain demonstrations.<n>The learned priors enable smooth style transitions, facilitating a unified policy that adapts to terrain variations.<n>We integrate T-GMP into an adversarial learning pipeline with our proposed Foothold Penalty, where a discriminator dynamically modulates naturalness constraints conditioned on local terrain features.
  arXiv  Detail & Related papers  (2026-06-05T06:15:42Z)
• CoRe-MoE: Contrastive Reweighted Mixture of Experts for Multi-Terrain Humanoid Locomotion with Gait Adaptation [18.803751724563448]
  We propose CoRe-MoE, a two-stage reinforcement learning framework that decouples gait generation from terrain adaptation.<n>In the first stage, a stable locomotion policy is learned to produce natural walking and running behaviors with smooth transitions.<n>In the second stage, a terrain-aware MoE branch is introduced, and the gating network is trained with a contrastive objective to learn structured terrain representations.
  arXiv  Detail & Related papers  (2026-06-03T10:51:46Z)
• START: Traversing Sparse Footholds with Terrain Reconstruction [11.632418196552324]
  We propose START, a single-stage learning framework that enables agile, stable locomotion on highly sparse footholds.<n> START leverages only low-cost onboard vision and proprioception to accurately reconstruct local terrain heightmap.<n> Experimental results demonstrate that START achieves zero-shot transfer across diverse real-world scenarios.
  arXiv  Detail & Related papers  (2025-12-15T10:02:41Z)
• Learning Terrain-Specialized Policies for Adaptive Locomotion in Challenging Environments [0.4588028371034407]
  This work introduces a hierarchical reinforcement learning framework that leverages terrain-specialized policies and curriculum learning to enhance agility and tracking performance in complex environments.<n>We validated our method on simulation, where our approach outperforms a generalist policy by up to 16% in success rate and achieves lower tracking errors as the velocity target increases.
  arXiv  Detail & Related papers  (2025-09-25T00:17:39Z)
• Multi-Agent Path Finding in Continuous Spaces with Projected Diffusion Models [57.45019514036948]
  Multi-Agent Path Finding (MAPF) is a fundamental problem in robotics.<n>This work proposes a novel approach that integrates constrained optimization with diffusion models for MAPF in continuous spaces.
  arXiv  Detail & Related papers  (2024-12-23T21:27:19Z)
• SAFE-SIM: Safety-Critical Closed-Loop Traffic Simulation with Diffusion-Controllable Adversaries [94.84458417662407]
  We introduce SAFE-SIM, a controllable closed-loop safety-critical simulation framework. Our approach yields two distinct advantages: 1) generating realistic long-tail safety-critical scenarios that closely reflect real-world conditions, and 2) providing controllable adversarial behavior for more comprehensive and interactive evaluations. We validate our framework empirically using the nuScenes and nuPlan datasets across multiple planners, demonstrating improvements in both realism and controllability.
  arXiv  Detail & Related papers  (2023-12-31T04:14:43Z)
• 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)
• Next Steps: Learning a Disentangled Gait Representation for Versatile Quadruped Locomotion [69.87112582900363]
  Current planners are unable to vary key gait parameters continuously while the robot is in motion. In this work we address this limitation 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, foot step height and full stance duration.
  arXiv  Detail & Related papers  (2021-12-09T10:02:02Z)
• First Steps: Latent-Space Control with Semantic Constraints for Quadruped Locomotion [73.37945453998134]
  Traditional approaches to quadruped control employ simplified, hand-derived models. This significantly reduces the capability of the robot since its effective kinematic range is curtailed. In this work, these challenges are addressed by framing quadruped control as optimisation in a structured latent space. A deep generative model captures a statistical representation of feasible joint configurations, whilst complex dynamic and terminal constraints are expressed via high-level, semantic indicators. We validate the feasibility of locomotion trajectories optimised using our approach both in simulation and on a real-worldmal quadruped.
  arXiv  Detail & Related papers  (2020-07-03T07:04:18Z)

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.