aiqbench analysis
This paper proposes an interface that lets a single learned humanoid controller handle object-level goals such as pushing, pulling, and two-arm transport, without task-specific retraining. The simulation results and sim-to-real transfer on the Digit robot are promising, but they come from one platform and a limited set of tasks, so broader generalization is still unproven. For AI readers, the main value is the idea of adding object-centric commands to learned whole-body control.
OIL: An Object Interaction Layer for Humanoid Whole-Body Control
arXiv:2610.14033v1 Announce Type: new Abstract: Recent progress in learned whole-body controllers for humanoid robots has enabled robust free-space motion by tracking kinematic targets such as joint or end-effector trajectories. This interface provides a useful abstraction for upstream systems, including motion capture, teleoperation, or generative models, which can specify desired motion without reasoning about forces or balance. However, this abstraction has largely been limited to robot motion, requiring task-specific controllers or retraining for object interaction. We introduce OIL, an Object Interaction Layer that extends the kinematic command interface of whole-body control to include desired object motion and contact. OIL augments a standard controller with object-centric commands and translates them into coordinated whole-body behavior while preserving locomotion and balance. This design enables a single learned policy to execute non-prehensile interactions, including pushing, pulling, two-arm pick-and-place, and locomotion-coupled transport, directly from object-level kinematic targets without task-specific retraining. In simulation, we demonstrate that object-level commands improve performance over robot-centric controllers while matching task-specific policies across variations in object properties such as mass and size. We further demonstrate zero-shot generalization to novel behavior combinations and sim-to-real transfer on the Digit humanoid robot.
Abstract:Recent progress in learned whole-body controllers for humanoid robots has enabled robust free-space motion by tracking kinematic targets such as joint or end-effector trajectories. This interface provides a useful abstraction for upstream systems, including motion capture, teleoperation, or generative models, which can specify desired motion without reasoning about forces or balance. However, this abstraction has largely been limited to robot motion, requiring task-specific controllers or retraining for object interaction. We introduce OIL, an Object Interaction Layer that extends the kinematic command interface of whole-body control to include desired object motion and contact. OIL augments a standard controller with object-centric commands and translates them into coordinated whole-body behavior while preserving locomotion and balance. This design enables a single learned policy to execute non-prehensile interactions, including pushing, pulling, two-arm pick-and-place, and locomotion-coupled transport, directly from object-level kinematic targets without task-specific retraining. In simulation, we demonstrate that object-level commands improve performance over robot-centric controllers while matching task-specific policies across variations in object properties such as mass and size. We further demonstrate zero-shot generalization to novel behavior combinations and sim-to-real transfer on the Digit humanoid robot.
Robotics (cs.RO)
arXiv:2610.14033 [cs.RO]
(or arXiv:2610.14033v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.14033 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 275 words
Related coverage
Exploiting Hierarchical Controller Structure in Contextual Parameter Learning for Humanoid Loco-Manipulation
arXiv · Oct 6, 2026
EgoExplorer: A Vision-Motion-Touch Observatory for Human Loco-Manipulation In Situ
arXiv · Oct 14, 2026
Dyna-2: General Purpose Physical Agents with Hour-Long Autonomy — Jason Ma, Dyna Robotics
Latent Space · Oct 13, 2026
Google DeepMind Ships Three Physical AI Models For Whole Body Control, Dexterity And Multi Robot Collaboration
MarkTechPost · Jul 30, 2026