paper-with-me

홈 › Papers

PhysMani: Physics-principled 3D World Model for Dynamic Object Manipulation

2026-07-02 · Peng Yun, Shouwang Huang, Hao Li, Jinxi Li, Jianan Wang, Bo Yang arxiv

Manipulating fast and dynamically moving targets in unstructured 3D environments remains challenging for embodied AI. Existing visual-language-action models and world models struggle with accurate 3D geometry and physically meaningful forecasting. We propose PhysMani, a framework that couples a physics-principled 3D Gaussian world model with a future-aware action policy model. The world model learns a divergence-free Gaussian velocity field via online optimization for fast and physically grounded future dynamics prediction. The policy model integrates the predicted 3D scene future dynamics through a learnable token based cross-attention module. We introduce PhysMani-Bench, a dynamic manipulation benchmark with 16 tasks, and demonstrate a superior success rate over strong baselines in both simulation and real-world robot experiments.

📄 PDF Abstract BibTeX arXiv:2607.01938

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

PAINET: A Principled Efficient Transformer for 3D Dynamics Modeling

2025-10-05 · Kai Yang, Yuqi Huang, Junheng Tao, Wanyu Wang 외 arxiv

Modeling 3D dynamics is a fundamental problem in multi-body systems across scientific and engineering domains and has important practical implications in object trajectory prediction and simulation. While recent GNN-base…

Trajectory Prediction

HUSKY: Humanoid Skateboarding System via Physics-Aware Whole-Body Control

2026-02-03 · Jinrui Han, Dewei Wang, Chenyun Zhang, Xinzhe Liu 외 arxiv

While current humanoid whole-body control frameworks predominantly rely on the static environment assumptions, addressing tasks characterized by high dynamism and complex interactions presents a formidable challenge. In …

A Compositional Object-Based Approach to Learning Physical Dynamics

2016-12-01 · Michael B. Chang, Tomer Ullman, Antonio Torralba, Joshua B. Tenenbaum

We present the Neural Physics Engine (NPE), a framework for learning simulators of intuitive physics that naturally generalize across variable object count and different scene configurations. We propose a factorization o…

Object

Incremental Few-Shot Adaptation for Non-Prehensile Object Manipulation using Parallelizable Physics Simulators

2024-09-20 · Fabian Baumeister, Lukas Mack, Joerg Stueckler

Few-shot adaptation is an important capability for intelligent robots that perform tasks in open-world settings such as everyday environments or flexible production. In this paper, we propose a novel approach for non-pre…

Model Predictive Control

Physics Supernova: AI Agent Matches Elite Gold Medalists at IPhO 2025

2025-09-01 · Jiahao Qiu, Jingzhe Shi, Xinzhe Juan, Zelin Zhao 외 arxiv

Physics provides fundamental laws that describe and predict the natural world. AI systems aspiring toward more general, real-world intelligence must therefore demonstrate strong physics problem-solving abilities: to form…