一个可跨平台通用的机器人学习框架，让不同机器人能快速学会复杂操作。
CoinRobot: Generalized End-to-end Robotic Learning for Physical Intelligence
- 统一架构支持工业机器人、协作臂等多平台无缝部署。
- 基于扩散模型在7个任务中表现优于LeRobot，泛化性更强。
- 适合需要快速适配新硬件和场景的机器人研发团队。
物理智能对推动具身智能发展具有巨大潜力，使机器人能从示范中习得复杂行为。然而，在不同机器人平台与环境中实现泛化与迁移，需精心设计模型架构、训练策略与数据多样性。现有系统常面临可扩展性差、异构硬件适应性弱及真实场景评估困难等问题。本文提出一种通用端到端机器人学习框架，以弥合这一差距。该框架采用统一架构，支持跨平台适应性，可在工业级机器人、协作机械臂及新型机体上无任务修改地部署。通过融合多任务学习与简化的网络设计，其性能优于传统方法，且兼容不同传感器配置与动作空间。我们在七个操控任务上进行了大量实验验证。值得注意的是，本框架训练的基于扩散模型在多种机器人平台与环境条件下均展现出更优性能与更强泛化能力，显著优于LeRobot框架。
原文摘要 · Abstract (English)
Physical intelligence holds immense promise for advancing embodied intelligence, enabling robots to acquire complex behaviors from demonstrations. However, achieving generalization and transfer across diverse robotic platforms and environments requires careful design of model architectures, training strategies, and data diversity. Meanwhile existing systems often struggle with scalability, adaptability to heterogeneous hardware, and objective evaluation in real-world settings. We present a generalized end-to-end robotic learning framework designed to bridge this gap. Our framework introduces a unified architecture that supports cross-platform adaptability, enabling seamless deployment across industrial-grade robots, collaborative arms, and novel embodiments without task-specific modifications. By integrating multi-task learning with streamlined network designs, it achieves more robust performance than conventional approaches, while maintaining compatibility with varying sensor configurations and action spaces. We validate our framework through extensive experiments on seven manipulation tasks. Notably, Diffusion-based models trained in our framework demonstrated superior performance and generalizability compared to the LeRobot framework, achieving performance improvements across diverse robotic platforms and environmental conditions.
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。