The Role of Domain Randomization in Training Diffusion Policies for Whole-Body Humanoid Control

• URL: http://arxiv.org/abs/2411.01349v1
• Date: Sat, 02 Nov 2024 19:33:28 GMT
• Title: The Role of Domain Randomization in Training Diffusion Policies for Whole-Body Humanoid Control
• Abstract summary: Policies Diffusion (DPs) have shown impressive results in robotic manipulation. This paper investigates how dataset diversity and size affect the performance of DPs for humanoid whole-body control.
• Score: 14.36344580057985
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Humanoids have the potential to be the ideal embodiment in environments designed for humans. Thanks to the structural similarity to the human body, they benefit from rich sources of demonstration data, e.g., collected via teleoperation, motion capture, or even using videos of humans performing tasks. However, distilling a policy from demonstrations is still a challenging problem. While Diffusion Policies (DPs) have shown impressive results in robotic manipulation, their applicability to locomotion and humanoid control remains underexplored. In this paper, we investigate how dataset diversity and size affect the performance of DPs for humanoid whole-body control. In a simulated IsaacGym environment, we generate synthetic demonstrations by training Adversarial Motion Prior (AMP) agents under various Domain Randomization (DR) conditions, and we compare DPs fitted to datasets of different size and diversity. Our findings show that, although DPs can achieve stable walking behavior, successful training of locomotion policies requires significantly larger and more diverse datasets compared to manipulation tasks, even in simple scenarios.

Related papers

• What Matters in Humanoid General Motion Tracking? An Empirical Study [2.728639503114606]
  We develop YAHMP, an open-source framework for training, evaluating, and deploying whole-body motion tracking policies on the Unitree G1.<n>We evaluate the resulting policies on a test set of retargeted human motions and compare the nominal policy with TWIST2 as an external baseline trained on the same motion set.
  arXiv  Detail & Related papers  (2026-07-22T08:37:19Z)
• HumanoidMimicGen: Data Generation for Loco-Manipulation via Whole-Body Planning [52.022681285103126]
  We present HumanoidMimicGen, a method for generating humanoid legged loco-manipulation data.<n>Our method adapts contact-rich whole-body skills from a handful of source demonstrations to new states.<n>We show that whole-body visuomotor policies co-trained with data generated by HumanoidMimicGen outperform those trained only on real-world data by 20%.
  arXiv  Detail & Related papers  (2026-05-26T21:57:11Z)
• BifrostUMI: Bridging Robot-Free Demonstrations and Humanoid Whole-Body Manipulation [6.733928872257592]
  BifrostUMI is a portable, efficient, and robot-free data collection framework for humanoid robots.<n>We demonstrate the efficacy and versatility of the proposed framework across two distinct experimental scenarios.
  arXiv  Detail & Related papers  (2026-05-05T07:35:09Z)
• Grasp as You Dream: Imitating Functional Grasping from Generated Human Demonstrations [23.294838162593184]
  We present GraspDreamer, a method to enable zero-shot functional grasping without labor-intensive data collection.<n>The key idea is that VGMs pre-trained on internet-scale human data implicitly encode generalized priors about how humans interact with the physical world.<n>Extensive experiments on the public benchmarks with different robot hands demonstrate the superior data efficiency and generalization performance of GraspDreamer.
  arXiv  Detail & Related papers  (2026-04-08T18:52:16Z)
• Data Analogies Enable Efficient Cross-Embodiment Transfer [72.88252238231269]
  Generalist robot policies are trained on demonstrations collected across a wide variety of robots, scenes, and viewpoints.<n>We ask: what form of demonstration data is most useful for enabling transfer across robot set-ups?<n>Our simulated experiments show that while perceptual shifts such as viewpoint benefit most from broad diversity, morphology shifts benefit far less from unstructured diversity.
  arXiv  Detail & Related papers  (2026-03-06T16:42:46Z)
• EgoHumanoid: Unlocking In-the-Wild Loco-Manipulation with Robot-Free Egocentric Demonstration [67.13034606664333]
  EgoHumanoid is the first framework to co-train a vision-language-action policy using egocentric human demonstrations.<n>A portable system for scalable human data collection is developed.
  arXiv  Detail & Related papers  (2026-02-10T18:59:03Z)
• Opening the Sim-to-Real Door for Humanoid Pixel-to-Action Policy Transfer [59.02729900344616]
  GPU-accelerated, photorealistic simulation has opened a scalable data-generation path for robot learning.<n>We develop a teacher-student-bootstrap learning framework for vision-based humanoid loco-manipulation.<n>This represents the first humanoid sim-to-real policy capable of diverse articulated loco-manipulation using pure RGB perception.
  arXiv  Detail & Related papers  (2025-11-30T20:07:13Z)
• DemoHLM: From One Demonstration to Generalizable Humanoid Loco-Manipulation [29.519071338337685]
  We present DemoHLM, a framework for humanoid loco-manipulation on a real humanoid robot from a single demonstration in simulation.<n>whole-body controller maps whole-body motion commands to joint torques and provides omnidirectional mobility for the humanoid robot.<n> Experiments show a positive correlation between the amount of synthetic data and policy performance.
  arXiv  Detail & Related papers  (2025-10-13T10:49:40Z)
• ResMimic: From General Motion Tracking to Humanoid Whole-body Loco-Manipulation via Residual Learning [59.64325421657381]
  Humanoid whole-body loco-manipulation promises transformative capabilities for daily service and warehouse tasks.<n>We introduce ResMimic, a two-stage residual learning framework for precise and expressive humanoid control from human motion data.<n>Results show substantial gains in task success, training efficiency, and robustness over strong baselines.
  arXiv  Detail & Related papers  (2025-10-06T17:47:02Z)
• Dexplore: Scalable Neural Control for Dexterous Manipulation from Reference-Scoped Exploration [58.4036440289082]
  Hand-object motion-capture (MoCap) offer large-scale, contact-rich demonstrations and hold promise for dexterous robotic scopes.<n>We introduce Dexplore, a unified single-loop optimization that performs repositories and tracking to learn robot control policies directly from MoCap at scale.
  arXiv  Detail & Related papers  (2025-09-11T17:59:07Z)
• Latent Policy Steering with Embodiment-Agnostic Pretrained World Models [8.265847442964045]
  We aim to reduce data collection efforts when learning visuomotor robot policies by leveraging existing or cost-effective data.<n>We use optic flow as an embodiment-agnostic action representation to train a World Model (WM) across multi-embodiment datasets.<n>We develop a method, Latent Policy Steering (LPS), to improve the output of a behavior-cloned policy by searching in the latent space of the WM for better action sequences.
  arXiv  Detail & Related papers  (2025-07-17T17:57:57Z)
• Is Diversity All You Need for Scalable Robotic Manipulation? [50.747150672933316]
  We investigate the nuanced role of data diversity in robot learning by examining three critical dimensions-task (what to do), embodiment (which robot to use), and expert (who demonstrates)-challenging the conventional intuition of "more diverse is better"<n>We show that task diversity proves more critical than per-task demonstration quantity, benefiting transfer from diverse pre-training tasks to novel downstream scenarios.<n>We propose a distribution debiasing method to mitigate velocity ambiguity, the yielding GO-1-Pro achieves substantial performance gains of 15%, equivalent to using 2.5 times pre-training data.
  arXiv  Detail & Related papers  (2025-07-08T17:52:44Z)
• Inference-Time Policy Steering through Human Interactions [54.02655062969934]
  During inference, humans are often removed from the policy execution loop. We propose an Inference-Time Policy Steering framework that leverages human interactions to bias the generative sampling process. Our proposed sampling strategy achieves the best trade-off between alignment and distribution shift.
  arXiv  Detail & Related papers  (2024-11-25T18:03:50Z)
• DivDiff: A Conditional Diffusion Model for Diverse Human Motion Prediction [9.447439259813112]
  We propose a conditional diffusion-based generative model, called DivDiff, to predict more diverse and realistic human motions. Specifically, the DivDiff employs DDPM as our backbone and incorporates Discrete Cosine Transform (DCT) and transformer mechanisms. We design a diversified reinforcement sampling function (DRSF) to enforce human skeletal constraints on the predicted human motions.
  arXiv  Detail & Related papers  (2024-08-16T04:51:32Z)
• Movement Primitive Diffusion: Learning Gentle Robotic Manipulation of Deformable Objects [14.446751610174868]
  Movement Primitive Diffusion (MPD) is a novel method for imitation learning (IL) in robot-assisted surgery. MPD combines the versatility of diffusion-based imitation learning (DIL) with the high-quality motion generation capabilities of Probabilistic Dynamic Movement Primitives (ProDMPs) We evaluate MPD across various simulated and real world robotic tasks on both state and image observations.
  arXiv  Detail & Related papers  (2023-12-15T18:24:28Z)
• H-GAP: Humanoid Control with a Generalist Planner [45.50995825122686]
  Humanoid Generalist Autoencoding Planner (H-GAP) is a generative model trained on humanoid trajectories derived from human motioncaptured data. For 56 degrees of freedom humanoid, we empirically demonstrate that H-GAP learns to represent and generate a wide range of motor behaviours. We also do a series of empirical studies on the scaling properties of H-GAP, showing the potential for performance gains via additional data but not computing.
  arXiv  Detail & Related papers  (2023-12-05T11:40:24Z)
• Robust Visual Sim-to-Real Transfer for Robotic Manipulation [79.66851068682779]
  Learning visuomotor policies in simulation is much safer and cheaper than in the real world. However, due to discrepancies between the simulated and real data, simulator-trained policies often fail when transferred to real robots. One common approach to bridge the visual sim-to-real domain gap is domain randomization (DR)
  arXiv  Detail & Related papers  (2023-07-28T05:47:24Z)
• Executing your Commands via Motion Diffusion in Latent Space [51.64652463205012]
  We propose a Motion Latent-based Diffusion model (MLD) to produce vivid motion sequences conforming to the given conditional inputs. Our MLD achieves significant improvements over the state-of-the-art methods among extensive human motion generation tasks.
  arXiv  Detail & Related papers  (2022-12-08T03:07:00Z)
• DeXtreme: Transfer of Agile In-hand Manipulation from Simulation to Reality [64.51295032956118]
  We train a policy that can perform robust dexterous manipulation on an anthropomorphic robot hand. Our work reaffirms the possibilities of sim-to-real transfer for dexterous manipulation in diverse kinds of hardware and simulator setups.
  arXiv  Detail & Related papers  (2022-10-25T01:51:36Z)
• DexTransfer: Real World Multi-fingered Dexterous Grasping with Minimal Human Demonstrations [51.87067543670535]
  We propose a robot-learning system that can take a small number of human demonstrations and learn to grasp unseen object poses. We train a dexterous grasping policy that takes the point clouds of the object as input and predicts continuous actions to grasp objects from different initial robot states. The policy learned from our dataset can generalize well on unseen object poses in both simulation and the real world.
  arXiv  Detail & Related papers  (2022-09-28T17:51:49Z)
• Learning Latent Traits for Simulated Cooperative Driving Tasks [10.009803620912777]
  We build a framework capable of capturing a compact latent representation of the human in terms of their behavior and preferences. We then build a lightweight simulation environment, HMIway-env, for modelling one form of distracted driving behavior. We finally use this environment to quantify both the ability to discriminate drivers and the effectiveness of intervention policies.
  arXiv  Detail & Related papers  (2022-07-20T02:27:18Z)
• TrafficSim: Learning to Simulate Realistic Multi-Agent Behaviors [74.67698916175614]
  We propose TrafficSim, a multi-agent behavior model for realistic traffic simulation. In particular, we leverage an implicit latent variable model to parameterize a joint actor policy. We show TrafficSim generates significantly more realistic and diverse traffic scenarios as compared to a diverse set of baselines.
  arXiv  Detail & Related papers  (2021-01-17T00:29:30Z)

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.