提出双系统机器人操控框架，兼顾通用性与高效执行
Towards Synergistic, Generalized, and Efficient Dual-System for Robotic Manipulation
- 融合通用模型与扩散变压器专家模型，实现任务理解与动作生成协同
- 真实场景表现提升26.7%，仅用5%演示数据仍保持强性能
- 适合需高控制频率的现实部署场景，参数量仅20M
随着对多样化动态环境中多功能机器人系统的需求增加，具备泛化能力的通用策略因其能利用大规模跨平台数据而备受重视，但其推理效率低且训练成本高。相比之下，专精策略针对特定领域数据设计，虽在任务层面精度高且高效，却缺乏广泛适用性。受此启发，我们提出RoboDual——一种协同双系统架构，融合通用与专精策略优势。基于视觉-语言-动作（VLA）的通用模型提供高层任务理解与离散动作输出，再由基于扩散变换器的专家模型生成多步动作序列。相比OpenVLA，RoboDual在真实环境提升26.7%，在CALVIN数据集上提升12%，仅使用20M可训练参数；仅需5%演示数据即保持良好性能，并实现3.8倍更高的实时控制频率。代码将公开。
原文摘要 · Abstract (English)
The increasing demand for versatile robotic systems to operate in diverse and dynamic environments has emphasized the importance of a generalist policy, which leverages a large cross-embodiment data corpus to facilitate broad adaptability and high-level reasoning. However, the generalist would struggle with inefficient inference and cost-expensive training. The specialist policy, instead, is curated for specific domain data and excels at task-level precision with efficiency. Yet, it lacks the generalization capacity for a wide range of applications. Inspired by these observations, we introduce RoboDual, a synergistic dual-system that supplements the merits of both generalist and specialist policy. A diffusion transformer-based specialist is devised for multi-step action rollouts, exquisitely conditioned on the high-level task understanding and discretized action output of a vision-language-action (VLA) based generalist. Compared to OpenVLA, RoboDual achieves 26.7% improvement in real-world setting and 12% gain on CALVIN by introducing a specialist policy with merely 20M trainable parameters. It maintains strong performance with 5% of demonstration data only, and enables a 3.8 times higher control frequency in real-world deployment. Code would be made publicly available. Our project page is hosted at: https://opendrivelab.com/RoboDual/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。