aiqbench analysis
This research introduces Robotic Multiphase Interaction (RMI), a novel framework enabling robots to manipulate coupled liquid-solid systems like wet sponges by predicting complex fluid dynamics. By integrating implicit porous flow models with diffusion policies and rectified flow matching, the system achieves over 60% reduction in prediction error for retained water volume. This advancement represents a significant step forward in machine learning applications for robotics, specifically addressing the challenges of controlling deformable objects with internal fluids.
Robotic Multiphase Interaction: Manipulating Coupled Liquid and Solid Dynamics with a World Model
arXiv:2609.21448v1 Announce Type: new Abstract: This work presents \textit{Robotic Multiphase Interaction (RMI)}, a setting in which liquid enters a porous material and interacts mechanically with its deforming solid skeleton. Manipulation can therefore change pore volume, expel or redistribute retained liquid, and alter grasp stability at the same time. Spilled liquid can also create safety risks in domestic and manufacturing settings. This differs from most manipulation of solid objects and from tasks that involve both liquid and solid while keeping the phases spatially separate. We study a sponge filled with water as the first RMI example. We use implicit incompressible porous flow with smoothed particle hydrodynamics as the dynamics engine and enable robotic manipulation by adding Coulomb contact memory, hybrid velocity and force regulation, and a stability gate for lifting. The resulting environment connects robot commands to changes in the coupled liquid and solid state. A world model conditioned on actions predicts how this state evolves under candidate commands, while a temporal UNet generates actions using either Diffusion Policy or rectified flow matching. Our world model reduces retained water prediction error by more than $60\%$ compared with the baseline. The best action sequence selected by the world model from policy proposals further reduces the predicted terminal water error by about half. These improvements show that modelling the coupled liquid and solid state helps the robot predict how its actions affect both the porous object and the liquid held inside.
Abstract:This work presents \textit{Robotic Multiphase Interaction (RMI)}, a setting in which liquid enters a porous material and interacts mechanically with its deforming solid skeleton. Manipulation can therefore change pore volume, expel or redistribute retained liquid, and alter grasp stability at the same time. Spilled liquid can also create safety risks in domestic and manufacturing settings. This differs from most manipulation of solid objects and from tasks that involve both liquid and solid while keeping the phases spatially separate. We study a sponge filled with water as the first RMI example. We use implicit incompressible porous flow with smoothed particle hydrodynamics as the dynamics engine and enable robotic manipulation by adding Coulomb contact memory, hybrid velocity and force regulation, and a stability gate for lifting. The resulting environment connects robot commands to changes in the coupled liquid and solid state. A world model conditioned on actions predicts how this state evolves under candidate commands, while a temporal UNet generates actions using either Diffusion Policy or rectified flow matching. Our world model reduces retained water prediction error by more than $60\%$ compared with the baseline. The best action sequence selected by the world model from policy proposals further reduces the predicted terminal water error by about half. These improvements show that modelling the coupled liquid and solid state helps the robot predict how its actions affect both the porous object and the liquid held inside.
Robotics (cs.RO)
arXiv:2609.21448 [cs.RO]
(or arXiv:2609.21448v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2609.21448 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 · 326 words
Related coverage
Interactive 4D Volumetric Liquid Forecasting under Moving-Solid Interaction
arXiv · Oct 15, 2026

News
Quantum Computing Report · Jul 25, 2026

Developing Healthcare Robotics with GPU-Native Medical Physics Simulation
NVIDIA · Jul 28, 2026

Game Design Creativity evals
LessWrong · Oct 12, 2026