cs.ROApr 14, 2026

Scalable Trajectory Generation for Whole-Body Mobile Manipulation

Authors: Yida Niu, Xinhai Chang, Xin Liu, Ziyuan Jiao, Yixin Zhu

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Robots deployed in unstructured environments must coordinate whole-body motion -- simultaneously moving a mobile base and arm -- to interact with the physical world. This coupled mobility and dexterity yields a state space that grows combinatorially with scene and object diversity, demanding datasets far larger than those sufficient for fixed-base manipulation. Yet existing acquisition methods, including teleoperation and planning, are either labor-intensive or computationally prohibitive at scale. The core bottleneck is the lack of a scalable pipeline for generating large-scale, physically valid, coordinated trajectory data across diverse embodiments and environments. Here we introduce AutoMoMa, a GPU-accelerated framework that unifies AKR modeling, which consolidates base, arm, and object kinematics into a single chain, with parallelized trajectory optimization. AutoMoMa achieves 5,000 episodes per GPU-hour (over 80×80\times faster than CPU-based baselines), producing a dataset of over 500k physically valid trajectories spanning 330 scenes, diverse articulated objects, and multiple robot embodiments. Prior datasets were forced to compromise on scale, diversity, or kinematic fidelity; AutoMoMa addresses all three simultaneously. Training downstream IL policies further reveals that even a single articulated-object task requires tens of thousands of demonstrations for SOTA methods to reach ≈80%\approx 80\% success, confirming that data scarcity -- not algorithmic limitations -- has been the binding constraint. AutoMoMa thus bridges high-performance planning and reliable IL-based control, providing the infrastructure previously missing for coordinated mobile manipulation research. By making large-scale, kinematically valid training data practical, AutoMoMa showcases generalizable whole-body robot policies capable of operating in the diverse, unstructured settings of the real world.

Figures & tables

DatasetRobot# EpisodesCoord.# ScenesActionMethod
RT-1 Robot Action [ 4 ]Google Robot73,499Yes10End-effector poseVR teleoperation
NYU VINN [ 33 ]Hello Stretch435Yes3End-effector poseKinesthetic teaching
BC-Z [ 18 ]Google Robot39,350Yes2–3End-effector poseVR teleoperation
ETH Agent Affordances [ 35 ]Franka120No50End-effector poseScripted policy
QUT Dexterous Manip. [ 5 ]Franka200No1End-effector poseVR teleoperation
CMU Stretch [ 1 , 29 ]Hello Stretch135No10End-effector poseScripted Policy
Table 1: Comparison of AutoMoMa with existing mobile manipulation datasets. Existing datasets are constrained by their acquisition methods: teleoperation yields high-fidelity but small-scale data, while scripted policies lack whole-body coordination. AutoMoMa overcomes these limitations through gpu -accelerated automated planning, simultaneously achieving large scale, broad diversity, and high-fidelity joint-space trajectories—a combination no prior dataset provides. “ Coord. ”: presence of whole-body base-arm coordination.
Figure 2: An example of the akr construction. The akr unifies independent kinematic trees into a single serial chain, enabling joint whole-body optimization of the base, arm, and object. The mobile base’s planar motion is modeled via a virtual base (blue), while a virtual joint (black) couples the manipulator (orange) to the target object (green). For articulated objects, the kinematic tree is inverted to reconfigure the kinematic root to the grasp point, forming a continuous chain (highlighted in yellow) rooted at the virtual world frame.
Figure 3: The AutoMoMa data generation pipeline. Starting from a task specification triplet (S,O,R)(\mathcal{S},\mathcal{O},\mathcal{R}) (left), AutoMoMa proceeds through four stages: (i) Task Specification defines the environmental, robotic, and object context; (ii) Problem Instantiation transforms raw scene assets into planning-ready primitives via esdf construction and akr assembly with spherical collision approximations; (iii) Trajectory Generation solves for optimal akr states under task-specific constraints to produce physically valid whole-body motions; and (iv) Rendering in NVIDIA Isaac Sim produces synchronized rgbd sequences and point clouds. The resulting trajectories span diverse scenes, objects, and robot embodiments (right).
Figure 4: Trajectory generation performance across six representative household scenes. (a) Test scenes with increasing spatial confinement. (b) Generation throughput (valid trajectories per second) decreases as scene clutter increases collision-checking overhead. (c) Average translational effort of the mobile base per trajectory (error bars: standard deviation). (d) Average rotational effort of the manipulator, reflecting compensatory whole-body motion in constrained environments.
Figure 5: Distribution of trajectory base positions. Blue and orange spheres denote start and goal base placements, respectively, illustrating the broad spatial coverage achieved by the ik clustering strategy.
Figure 6: Data scaling experiments. (a) In a single scene, the mobile base policy requires substantially more data than the fixed-base counterpart, with a persistent seen/unseen gap indicating manifold memorization. (b) Increasing scene diversity from 1 to 30 steadily improves generalization to unseen environments. (c) With 30 scenes, higher per-scene trajectory density further refines execution precision, enabling consistent generalization across seen and unseen scenes.
Figure 7: Architectural generalization of AutoMoMa . When evaluated across the same 30-scene setup as DP3 [ 49 ] , both DP [ 6 ] and ACT [ 13 ] exhibit consistent performance gains with increasing trajectory density, demonstrating AutoMoMa ’s compatibility with diverse whole-body il architectures.
Figure 8: Per-object success rates at 100k trajectories. Success rates of the DP3 policy evaluated on five representative SAPIEN [ 47 ] objects. The bar plot compares performance under unseen (orange) and seen (blue) environments.
Appendix figures & tables16 assets

Supplementary material from the paper’s appendix.

Appendix

Figure A1: Downstream scene split. 30 training and 10 test scenes are sampled from the 300 procedurally generated scenes for policy training and evaluation.
Figure A2: Two scene sources in AutoMoMa . (a) AI2-THOR scenes with articulated objects substituted from SAPIEN. (b) Procedurally generated layouts from Infinigen.
HyperparameterValue
Environment and Collision
Object offset base (quaternion)[0, 0, 0, 1, 0, 0, 0]
Expanded dimension0.3
Disable collisionFalse
Collision typeVoxel / Mesh
ik mesh pitch0.05
Appendix
Table A1: Hyperparameters for AutoMoMa data generation.
HyperparameterValue
Observation and Action
Number of points in point cloud4096
Point feature dimension3
Proprioceptive input dim11
History length (observation steps)3
Prediction horizon8
Appendix
Table A2: Hyperparameters for DP3 [ 49 ] .
HyperparameterValue
Observation and Action
Number of camera views3
Proprioceptive input dim11
History length (observation steps)3
Prediction horizon8
Number of action steps6
Appendix
Table A3: Hyperparameters for DP [ 6 ] .
HyperparameterValue
Observation and Action
Number of camera views3
Action dim11
Action chunk size6
Temporal aggregationFalse
Mobile base moderelative
Appendix
Table A4: Hyperparameters for ACT [ 13 ] .
Figure A3: Trajectory failure caused by collision. Sphere-based geometry approximations occasionally fail to capture the original shape precisely, resulting in unintended collisions during planning.
Figure A4: Trajectory failure caused by fixed-base constraint violation. The planned trajectory erroneously involves movements inconsistent with the object’s fixed-base constraint.
Figure A5: Representative successful trajectory. The planned trajectory correctly respects both collision and motion constraints, producing a physically plausible whole-body motion.
Figure A6: Representative successful inference trajectories across diverse scenes. Trained on large-scale AutoMoMa data, DP3 [ 49 ] produces feasible whole-body trajectories across varied spatial layouts and articulated-object configurations.
Figure A7: Representative inference failure. Small inconsistencies in base or arm pose predictions compound over time, eventually pushing the robot into an infeasible configuration.
Figure A8: Real-world validation on a UR5-Ridgeback platform. Planned trajectories for drawer opening and cabinet door opening are executed smoothly without collision or constraint violation.
Figure A9: Scaling analysis at 100k trajectories. Training DP3 on a 100k-trajectory dataset spanning multiple articulated objects and scenes improves cross-object capability and generalization to unseen environments, compared with smaller-scale training subsets.
Figure A10: Ablation on start-state diversity. Top: Increasing unique start states (50–1,000) monotonically improves success on the 1,000-trajectory benchmark. Bottom: For fixed dataset sizes (6,400 trajectories), more diverse start states improve performance; the full 12,800-trajectory dataset achieves the highest success rate.
Figure A11: Inference examples for the pick task using a policy trained on 1,000 AutoMoMa trajectories.
Figure A12: Fixed-grasp vs . grasp-switching. Grasp switching enables larger object opening angles by avoiding link collisions.

Explore similar work

CardsList

  1. Whole-Body Mobile Manipulation using Offline Reinforcement Learning on Sub-optimal Controllers

    Apr 14, 2026Snehal Jauhri, Vignesh Prasad, Georgia ChalvatzakiMobile ManipulationWhole-Body Robotic Manipulation

    PEARL Lab, Computer Science Department, Technische Universität Darmstadt, Germany · Hessian.AI, Darmstadt, Germany · Robotics Institute, Germany

  2. WARP: Whole-Body Retargeting for Learning from Offline Human Demonstrations

    Jun 29, 2026Zhenyang Chen, Chuizheng Kong, Chuye Zhang +4Mobile ManipulationCross-Embodiment Robot Learning

    Georgia Institute of Technology, Atlanta, Georgia 30332

  3. DreamTrajectory: Trajectory-Guided Action Generation with World Model Alignment for Mobile Manipulation

    Aug 2, 2026Zheng Yang, Wenjie Zhang, Xiangyu Chen +7Trajectory GenerationWorld Models for Robotics

    Hong Kong University of Science and Technology (Guangzhou) · Ola Dimensions