Latent Dynamics · 2026-10-09
RiCo: Neural Simulation of Rigid-Body Interactions via Local Contact Reasoning
Ruixiang Ouyang, Guanren Qiao, Fansen Meng, Yueci Deng, Ruixing Jin, Kui Jia, Guiliang Liu
AI-assisted summary based on the paper. Consult the linked source for methods and limitations. Benchmark numbers appear after editor verification.
TL;DR
RiCo is a neural simulator for rigid-body motion that reasons over sparse local contact neighborhoods rather than dense scene/object interactions. It represents each surface point’s contacts via nearby candidate points (from other objects and static environment) and uses intra-object point-transformer reasoning to predict rigid motion updates via anchors. On MOVi, the authors report up to ~35% lower position error and ~38% lower orientation error (plus reported contact-fidelity improvements) and zero-shot generalization to scenes with 270 objects.
Why it matters
Rigid-body contact is local, so modeling all point-pair interactions is costly and may miss fine-grained contact effects. RiCo’s design confines cross-object reasoning to sparse nearby contact candidates and propagates contact information within each rigid body to preserve interaction details while keeping computation manageable. The paper also reports quantitative improvements on trajectory errors and contact fidelity, and evaluates large-scale generalization and preliminary sim-to-real behavior.
Method
- Construct sparse local contact neighborhoods for each surface point using broad-phase AABB filtering (contact radius ρ) and K nearest candidates within ρ, including learned features from relative geometry/motion and physical properties.
- Fuse each point’s state with aggregated contact-conditioned embeddings and run shared intra-object point-Transformer blocks (no dense cross-object attention) to integrate contacts across an object’s surface.
- Predict anchor motion updates and recover a single rigid transformation per object (anchor alignment via Kabsch), then simulate autoregressively by rebuilding contacts from predicted geometry.
Limitation
The paper provides only “preliminary evidence of sim-to-real transfer” via real-world multi-ball collision experiments.
Abstract (from arXiv)
Accurate simulation of rigid-body interactions is essential for predictive physical world models. Despite recent progress in modeling object dynamics, capturing how local contacts between surfaces shape object motion remains challenging. While end-to-end world models predict interactions across entire scenes or objects, in practice, rigid-body contact is inherently local, and only nearby surfaces can directly exchange contact forces. Motivated by this observation, we introduce Rigid-body Contact Reasoning (RiCo), which represents interactions between objects through sparse neighborhoods of contact surface points. RiCo combines each point's state with the relative geometry, motion, and physical properties of nearby surfaces, then reasons across the object's points to determine how these local contacts jointly affect its motion. By confining cross-object reasoning to nearby surfaces while propagating contact information within each rigid body, RiCo retains fine-grained interaction details without the cost of modeling every pair of scene points. Such properties enable RiCo a higher accuracy and contact fidelity. Experiments on MOVi-benchmark demonstrate that RiCo reduces 100-frame position and orientation errors by 31-35% and approximately 38%, respectively, compared with baselines. Moreover, RiCo achieves high contact fidelity, with ground-truth-relative penetration-time and mean-depth differences of 11.0% and 2.22 mm, respectively. RiCo further generalizes zero-shot from small-scale training scenarios to scenes containing 270 objects. Our real-world multi-ball collision experiments further provide preliminary evidence of sim-to-real transfer.
Related papers
- GPWM: Modeling the World as 3D Particle Motion
- Physics-Aware Object-Centric SE(3) Forecasting for Real-Time Proactive Dynamic Grasping
- ForceLDM: A Force-aware Latent Dynamics Model for Contact-Rich Manipulation
- Action Representation for Spatial Generalization in Robot World Models
- MotionCore: A Latent Neural Simulator for Video Generation with Rigid-Body Dynamics
Publication record
Bibliographic evidence describes where and when a record appeared. It does not validate scientific claims.
First public release: 2026-10-08 · Indexed by Pub-AI: 2026-10-09 UTC
RiCo: Neural Simulation of Rigid-Body Interactions via Local Contact Reasoning
Preprint posted: 2026-10-08 · Original record ↗
- arxiv
- 2610.12333
Explicit source versions (1)
- v1 · 2026-10-08
- Preprint available
Source date: 2026-10-08 · First detected: 2026-10-25 UTC · Historical baseline
Source evidence (1)
- Official evidence ↗ · atom:published (official metadata)
Abstract · arxiv
Accurate simulation of rigid-body interactions is essential for predictive physical world models. Despite recent progress in modeling object dynamics, capturing how local contacts between surfaces shape object motion remains challenging. While end-to-end world models predict interactions across entire scenes or objects, in practice, rigid-body contact is inherently local, and only nearby surfaces can directly exchange contact forces. Motivated by this observation, we introduce Rigid-body Contact Reasoning (RiCo), which represents interactions between objects through sparse neighborhoods of contact surface points. RiCo combines each point's state with the relative geometry, motion, and physical properties of nearby surfaces, then reasons across the object's points to determine how these local contacts jointly affect its motion. By confining cross-object reasoning to nearby surfaces while propagating contact information within each rigid body, RiCo retains fine-grained interaction details without the cost of modeling every pair of scene points. Such properties enable RiCo a higher accuracy and contact fidelity. Experiments on MOVi-benchmark demonstrate that RiCo reduces 100-frame position and orientation errors by 31-35% and approximately 38%, respectively, compared with baselines. Moreover, RiCo achieves high contact fidelity, with ground-truth-relative penetration-time and mean-depth differences of 11.0% and 2.22 mm, respectively. RiCo further generalizes zero-shot from small-scale training scenarios to scenes containing 270 objects. Our real-world multi-ball collision experiments further provide preliminary evidence of sim-to-real transfer.
