PapersMulti-source

RoboTeam 基准:多机械臂视觉实时协同控制测试

RoboTeam: Benchmarking Vision-Based Multi-Agent Control of Many Robot Arms in Real Time

Why we picked it

多机械臂协作的新基准,只靠摄像头画面实时干活,不暂停世界也不给坐标。gpt-6.1-sol 能过 28 个任务但单任务要 3.5 小时和 2 亿 token,做机器人智能体和 agent 架构选型的人应该看看这套对比数据。

arXiv 发布 RoboTeam,一个模拟环境基准,要求智能体仅凭摄像头图像实时控制 4 到 16 台机械臂,包含 36 个需要协作的长时任务,如传递木料、用数百个方块拼图。与既有基准不同,它不提供物体真实坐标,且世界运行不暂停。

Body

论文以自适应系统为起点:由单个智能体决定何时分叉子智能体。该架构下 gpt-6.1-sol 和 opus-5.5 分别完成 28 和 27 个任务,其余四个模型最多完成 3 个。但代价高昂:gpt-6.1-sol 完成一个典型任务需 3.5 小时、2.18 亿输入 token,且约三分之一步骤至少失败重试一次。

论文随后对比三种固定架构:单智能体、层级式与去中心化。四者完成任务数接近,但层级式与去中心化更快、token 消耗最高 1.9 倍,单智能体省 token 但耗时 1.6 倍,自适应系统居中。结论是没有一种模型或架构能同时做到快、省、可靠,RoboTeam 正是用长时多臂任务、纯视觉输入和实时世界来暴露这些限制。

Original · arXiv cs.AI

RoboTeam: Benchmarking Vision-Based Multi-Agent Control of Many Robot Arms in Real Time

Multimodal agents can now control robot arms by calling tools and writing code. Existing benchmarks fall short in three ways: most test one or two arms on short tasks, many pause the world while the agent thinks, and many give it ground-truth object positions. We present RoboTeam, a simulated benchmark where agents control 4 to 16 robot arms in real time from camera images alone. Its 36 tasks are long jobs that the arms must do together, such as passing timbers or building a picture from hundreds of cubes. We start from the most flexible agent system, the adaptive system, where one agent decides when to fork delegates. With it, gpt-6.1-sol and opus-5.5 complete 28 and 27 tasks, and four other models at most 3. But at a high price: a typical task takes gpt-6.1-sol 3.5 hours and 218 million input tokens, and a third of its steps fail at least once and must be retried. We then compare it with three fixed systems: single-agent, hierarchical, and decentralized. All four complete a similar number of tasks but differ in time and tokens. The hierarchical and decentralized systems are faster but use up to 1.9 times the tokens, and the single agent uses fewer tokens but takes 1.6 times as long, so the adaptive system lies between them. Even so, no model or system is yet fast, cheap, and reliable at once. RoboTeam measures these limits where earlier benchmarks cannot: long multi-arm jobs, camera images only, and a world that never pauses.

Industry Coordinates & Source Verification

This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include RoboTeam, gpt-6.1-sol, opus-5.5, agents.

Source attribution: curated from arXiv cs.AI (@Changxuan Fan, Zhongwei Xie, Boyi Zhang, Shaoli Hu, Chang Xu, Xiyu Ren, Yufei Li, Qiao Xiao, Huihao Jing, Kawai Chung, Wenbin Hu, Hongyu Luo, Lingyun Xie, Yisen Gao, Chunyang Li, Haoyu Huang, Wuganjing Song, Yuxuan Liu, Chiu Yeung Tang, Fangneng Zhan, Xi Yang). Cross-source corroboration: 3 independent sources have covered this development (including Aravind Srinivas, Browser Use), confirming industry-wide signal strength.