ParticleFormer: A 3D Point Cloud World Model for Multi-Object, Multi-Material Robotic Manipulation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Huang, Suning, Chen, Qianzhong, Zhang, Xiaohan, Sun, Jiankai, Schwager, Mac
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908503121068032
author Huang, Suning
Chen, Qianzhong
Zhang, Xiaohan
Sun, Jiankai
Schwager, Mac
author_facet Huang, Suning
Chen, Qianzhong
Zhang, Xiaohan
Sun, Jiankai
Schwager, Mac
contents 3D world models (i.e., learning-based 3D dynamics models) offer a promising approach to generalizable robotic manipulation by capturing the underlying physics of environment evolution conditioned on robot actions. However, existing 3D world models are primarily limited to single-material dynamics using a particle-based Graph Neural Network model, and often require time-consuming 3D scene reconstruction to obtain 3D particle tracks for training. In this work, we present ParticleFormer, a Transformer-based point cloud world model trained with a hybrid point cloud reconstruction loss, supervising both global and local dynamics features in multi-material, multi-object robot interactions. ParticleFormer captures fine-grained multi-object interactions between rigid, deformable, and flexible materials, trained directly from real-world robot perception data without an elaborate scene reconstruction. We demonstrate the model's effectiveness both in 3D scene forecasting tasks, and in downstream manipulation tasks using a Model Predictive Control (MPC) policy. In addition, we extend existing dynamics learning benchmarks to include diverse multi-material, multi-object interaction scenarios. We validate our method on six simulation and three real-world experiments, where it consistently outperforms leading baselines by achieving superior dynamics prediction accuracy and less rollout error in downstream visuomotor tasks. Experimental videos are available at https://suninghuang19.github.io/particleformer_page/.
format Preprint
id arxiv_https___arxiv_org_abs_2506_23126
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ParticleFormer: A 3D Point Cloud World Model for Multi-Object, Multi-Material Robotic Manipulation
Huang, Suning
Chen, Qianzhong
Zhang, Xiaohan
Sun, Jiankai
Schwager, Mac
Robotics
3D world models (i.e., learning-based 3D dynamics models) offer a promising approach to generalizable robotic manipulation by capturing the underlying physics of environment evolution conditioned on robot actions. However, existing 3D world models are primarily limited to single-material dynamics using a particle-based Graph Neural Network model, and often require time-consuming 3D scene reconstruction to obtain 3D particle tracks for training. In this work, we present ParticleFormer, a Transformer-based point cloud world model trained with a hybrid point cloud reconstruction loss, supervising both global and local dynamics features in multi-material, multi-object robot interactions. ParticleFormer captures fine-grained multi-object interactions between rigid, deformable, and flexible materials, trained directly from real-world robot perception data without an elaborate scene reconstruction. We demonstrate the model's effectiveness both in 3D scene forecasting tasks, and in downstream manipulation tasks using a Model Predictive Control (MPC) policy. In addition, we extend existing dynamics learning benchmarks to include diverse multi-material, multi-object interaction scenarios. We validate our method on six simulation and three real-world experiments, where it consistently outperforms leading baselines by achieving superior dynamics prediction accuracy and less rollout error in downstream visuomotor tasks. Experimental videos are available at https://suninghuang19.github.io/particleformer_page/.
title ParticleFormer: A 3D Point Cloud World Model for Multi-Object, Multi-Material Robotic Manipulation
topic Robotics
url https://arxiv.org/abs/2506.23126