MobileManiBench: Simplifying Model Verification for Mobile Manipulation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Wenbo, Wei, Fangyun, Li, QiXiu, Chen, Xi, Liang, Yaobo, Xu, Chang, Yang, Jiaolong, Guo, Baining
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917249309212672
author Wang, Wenbo
Wei, Fangyun
Li, QiXiu
Chen, Xi
Liang, Yaobo
Xu, Chang
Yang, Jiaolong
Guo, Baining
author_facet Wang, Wenbo
Wei, Fangyun
Li, QiXiu
Chen, Xi
Liang, Yaobo
Xu, Chang
Yang, Jiaolong
Guo, Baining
contents Vision-language-action models have advanced robotic manipulation but remain constrained by reliance on the large, teleoperation-collected datasets dominated by the static, tabletop scenes. We propose a simulation-first framework to verify VLA architectures before real-world deployment and introduce MobileManiBench, a large-scale benchmark for mobile-based robotic manipulation. Built on NVIDIA Isaac Sim and powered by reinforcement learning, our pipeline autonomously generates diverse manipulation trajectories with rich annotations (language instructions, multi-view RGB-depth-segmentation images, synchronized object/robot states and actions). MobileManiBench features 2 mobile platforms (parallel-gripper and dexterous-hand robots), 2 synchronized cameras (head and right wrist), 630 objects in 20 categories, 5 skills (open, close, pull, push, pick) with over 100 tasks performed in 100 realistic scenes, yielding 300K trajectories. This design enables controlled, scalable studies of robot embodiments, sensing modalities, and policy architectures, accelerating research on data efficiency and generalization. We benchmark representative VLA models and report insights into perception, reasoning, and control in complex simulated environments.
format Preprint
id arxiv_https___arxiv_org_abs_2602_05233
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle MobileManiBench: Simplifying Model Verification for Mobile Manipulation
Wang, Wenbo
Wei, Fangyun
Li, QiXiu
Chen, Xi
Liang, Yaobo
Xu, Chang
Yang, Jiaolong
Guo, Baining
Robotics
Vision-language-action models have advanced robotic manipulation but remain constrained by reliance on the large, teleoperation-collected datasets dominated by the static, tabletop scenes. We propose a simulation-first framework to verify VLA architectures before real-world deployment and introduce MobileManiBench, a large-scale benchmark for mobile-based robotic manipulation. Built on NVIDIA Isaac Sim and powered by reinforcement learning, our pipeline autonomously generates diverse manipulation trajectories with rich annotations (language instructions, multi-view RGB-depth-segmentation images, synchronized object/robot states and actions). MobileManiBench features 2 mobile platforms (parallel-gripper and dexterous-hand robots), 2 synchronized cameras (head and right wrist), 630 objects in 20 categories, 5 skills (open, close, pull, push, pick) with over 100 tasks performed in 100 realistic scenes, yielding 300K trajectories. This design enables controlled, scalable studies of robot embodiments, sensing modalities, and policy architectures, accelerating research on data efficiency and generalization. We benchmark representative VLA models and report insights into perception, reasoning, and control in complex simulated environments.
title MobileManiBench: Simplifying Model Verification for Mobile Manipulation
topic Robotics
url https://arxiv.org/abs/2602.05233