SCDP: Learning Humanoid Locomotion from Partial Observations via Mixed-Observation Distillation

• URL: http://arxiv.org/abs/2603.09574v1
• Date: Tue, 10 Mar 2026 12:21:11 GMT
• Title: SCDP: Learning Humanoid Locomotion from Partial Observations via Mixed-Observation Distillation
• Abstract summary: We present Sensor-Conditioned Diffusion Policies (SCDP) that enables humanoid locomotion using only onboard sensors.<n>SCDP decouples sensing from supervision through mixed-observation training.<n>We validate SCDP on velocity-commanded locomotion and motion reference tracking tasks.
• Score: 3.149420237200927
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Distilling humanoid locomotion control from offline datasets into deployable policies remains a challenge, as existing methods rely on privileged full-body states that require complex and often unreliable state estimation. We present Sensor-Conditioned Diffusion Policies (SCDP) that enables humanoid locomotion using only onboard sensors, eliminating the need for explicit state estimation. SCDP decouples sensing from supervision through mixed-observation training: diffusion model conditions on sensor histories while being supervised to predict privileged future state-action trajectories, enforcing the model to infer the motion dynamics under partial observability. We further develop restricted denoising, context distribution alignment, and context-aware attention masking to encourage implicit state estimation within the model and to prevent train-deploy mismatch. We validate SCDP on velocity-commanded locomotion and motion reference tracking tasks. In simulation, SCDP achieves near-perfect success on velocity control (99-100%) and 93% tracking success in AMASS test set, performing comparable to privileged baselines while using only onboard sensors. Finally, we deploy the trained policy on a real G1 humanoid at 50 Hz, demonstrating robust real robot locomotion without external sensing or state estimation.

Related papers

• Learning-Based Dynamic Obstacle Avoidance for a UAV Using Only Three Range Sensors [1.2246649738388389]
  We present a learning-based approach to kinodynamic online motion planning for an Unmanned Aerial Vehicle (UAV) operating at a fixed altitude in unknown dynamic environments.<n>The proposed framework integrates a behavior grid map representation with Deep Reinforcement Learning (DRL) for stable policy learning in continuous control.<n>The key idea is the co-design of a state representation and control policy that enables reliable navigation using only three low-cost directional range sensors.
  arXiv  Detail & Related papers  (2026-09-13T10:53:43Z)
• Dynamical System-Based Imitation Learning and Neuroadaptive Control for Trajectory Recovery in Autonomous Ships [0.7734726150561086]
  Repetitive maritime operations can be effectively learned using the Imitation Learning (IL) paradigm.<n> Dynamical Systems (DS) are widely used to model non-linear human demonstrations.<n>We present a hybrid learning-control architecture that integrates a DS-based IL reference generator with a neuroadaptive controller.
  arXiv  Detail & Related papers  (2026-08-25T00:04:39Z)
• FutureRTC: Real-Time Robot Execution with Anticipatory-Conditioned Action Chunking [51.97839311685636]
  FutureRTC is a plug-and-play adaptation framework that predicts execution-time observations and states for asynchronous VLA control without modifying the underlying policy.<n>We show that FutureRTC achieves robustness to inference delays, resulting in smoother trajectories, faster execution, and consistently higher task success rates.
  arXiv  Detail & Related papers  (2026-07-27T05:10:16Z)
• End-to-end Conditional Diffusion for Realistic and Controllable Visual Traffic Scenario Generation [59.188006170485046]
  E2E-CDiff is an end-to-end conditional diffusion framework for controllable and realistic scenario generation.<n>Differentiable guidance regulates speed, enforces drivable-area compliance, and supports collision-avoidance or collision-seeking behaviors.<n>E2E-CDiff also performs competitively as a learning-based ego planner.
  arXiv  Detail & Related papers  (2026-07-21T02:13:25Z)
• Predictive Training with Latent Imagination for Visual Quadruped Navigation [2.687590979817173]
  Reinforcement-learning navigation policies for legged robots select actions reactively from current observations and short-term memory.<n>Lightweight predictive supervision applied to the policy's recurrent state during training can encode anticipatory obstacle dynamics without modifying the inference-time controller.<n>On simulated and real-world navigation benchmarks with dynamic obstacles, our method substantially improves navigation success while reducing collision rates through the predictive training signal alone.
  arXiv  Detail & Related papers  (2026-07-20T05:33:22Z)
• KinematicRL: A Sim-to-Real Reinforcement Learning Framework For Social Navigation With Kinodynamic Feasibility [27.73146983499444]
  Deep Reinforcement Learning (DRL) has shown promise for social navigation, yet its real-world deployment remains hindered by a persistent sim-to-real gap.<n>This work presents a unified framework that addresses these limitations to produce dynamically feasible navigation policies.<n> Experiments in real-world environments demonstrate that, when combined with the proposed tracking pipeline, KinematicRL can be deployed on a real differential drive robot with minimal modifications.
  arXiv  Detail & Related papers  (2026-06-10T13:07:03Z)
• When Sensors Fail: Temporal Sequence Models for Robust PPO under Sensor Drift [64.37959940809633]
  We study robustness of Proximal Policy Optimization (PPO) under temporally persistent sensor failures.<n>We show Transformer-based sequence policies substantially outperform, RNN, and SSMs in robustness, maintaining high returns even when large fractions of sensors are unavailable.
  arXiv  Detail & Related papers  (2026-03-04T22:21:54Z)
• ULTRA: Unified Multimodal Control for Autonomous Humanoid Whole-Body Loco-Manipulation [55.467742403416175]
  We introduce a physics-driven neural algorithm that translates large-scale motion capture to humanoid embodiments.<n>We learn a unified multimodal controller that supports both dense references and sparse task specifications.<n>Results show that ULTRA generalizes to autonomous, goal-conditioned whole-body loco-manipulation from egocentric perception.
  arXiv  Detail & Related papers  (2026-03-03T18:59:29Z)
• Control and State Estimation of Vehicle-Mounted Aerial Systems in GPS-Denied, Non-Inertial Environments [5.191474416940848]
  We present a robust control and estimation framework for quadrotors operating in Global Navigation Satellite System (GNSS)-denied, non-inertial environments.<n>Our method relies exclusively on external position measurements combined with a Extended Kalman Filter with Unknown Inputs (EKF-UI) to account for platform motion.<n>Compared to standard EKF, the proposed method significantly improves stability and trajectory tracking without requiring inertial feedback.
  arXiv  Detail & Related papers  (2026-02-03T22:45:34Z)
• Mitigating Covariate Shift in Imitation Learning for Autonomous Vehicles Using Latent Space Generative World Models [60.87795376541144]
  A world model is a neural network capable of predicting an agent's next state given past states and actions.<n>During end-to-end training, our policy learns how to recover from errors by aligning with states observed in human demonstrations.<n>We present qualitative and quantitative results, demonstrating significant improvements upon prior state of the art in closed-loop testing.
  arXiv  Detail & Related papers  (2024-09-25T06:48:25Z)
• Hybrid Internal Model: Learning Agile Legged Locomotion with Simulated Robot Response [25.52492911765911]
  We introduce Hybrid Internal Model to estimate external states according to the response of the robot. The response, which we refer to as the hybrid internal embedding, contains the robot's explicit velocity and implicit stability representation. A wealth of real-world experiments demonstrates its agility, even in high-difficulty tasks and cases never occurred during the training process.
  arXiv  Detail & Related papers  (2023-12-18T18:59:06Z)
• Safe Output Feedback Motion Planning from Images via Learned Perception Modules and Contraction Theory [6.950510860295866]
  We present a class of uncertain control-affine nonlinear systems which guarantees runtime safety and goal reachability. We train a perception system that seeks to invert a subset of the state from an observation, and estimate an upper bound on the perception error. Next, we use contraction theory to design a stabilizing state feedback controller and a convergent dynamic state observer. We derive a bound on the trajectory tracking error when this controller is subjected to errors in the dynamics and incorrect state estimates.
  arXiv  Detail & Related papers  (2022-06-14T02:03:27Z)
• Stochastic Trajectory Prediction via Motion Indeterminacy Diffusion [88.45326906116165]
  We present a new framework to formulate the trajectory prediction task as a reverse process of motion indeterminacy diffusion (MID) We encode the history behavior information and the social interactions as a state embedding and devise a Transformer-based diffusion model to capture the temporal dependencies of trajectories. Experiments on the human trajectory prediction benchmarks including the Stanford Drone and ETH/UCY datasets demonstrate the superiority of our method.
  arXiv  Detail & Related papers  (2022-03-25T16:59:08Z)
• Robust Behavioral Cloning for Autonomous Vehicles using End-to-End Imitation Learning [0.4588028371034407]
  We present a lightweight pipeline for robust behavioral cloning of a human driver using end-to-end imitation learning. The proposed pipeline was employed to train and deploy three distinct driving behavior models onto a simulated vehicle.
  arXiv  Detail & Related papers  (2020-10-09T19:03:15Z)

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.