Physical SLAM couples spring-mass dynamics with 4D Gaussians to predict deformable object motion

zhenjun_zhao · x · 2026-10-17

• Goal: Reconstruct a predictive 4D representation of the physical world from RGB-D video — jointly building dynamic Gaussian radiance fields and inferring underlying physical dynamics to forecast future states under external actions.
• Method:
• Semantically decouples a static background Gaussian field (for camera localization) from a dynamic object field driven by a spring-mass model with elastic deformation, inertia, and damping;
• Introduces virtual observation forces that gradually pull simulated surface nodes toward observed depth points;
• Physical parameters are updated from simulation errors aggregated over the full observation window;
• A physics-aware Gaussian transformation keeps dynamic Gaussians aligned with the deforming surface.
• Results: 1.43cm average ATE on TUM RGB-D dynamic sequences, beating 4DGS-SLAM (1.8cm); rendering PSNR 22.31 / LPIPS 0.209 on par with or better than strongest baselines.

Original post →

More from Embodied

• Contact-UAN: MIT's residual world model fixes humanoid sim-to-real with 6.5 minutes of data — kevin_zakka · 2026-10-17
• ROS 2 LiDAR localization runs 390 m outdoors on handheld Livox MID-360 — rsasaki0109 · 2026-10-17
• Embodied Astra tops RoboMME at 92.1% — beating humans by just keeping full visual history in context — hankyang94 · 2026-10-17
• ROS weekly: ROSCon China kicks off in Shanghai, ROS By-The-Bay set for Oct 22 at Intrinsic — kscottz · 2026-10-17
• Simulated robotics eval: Claude Opus 5.5 preferred over Opus 5 in 14 of 16 comparisons — qinzytech · 2026-10-17
• FoundationVIO fuses IMU with foundation-model predictions to fix metric scale in feed-forward SLAM — zhenjun_zhao · 2026-10-17

Embodied channel →