统一机器人策略部署与数据收集，让每次测试都直接用于下轮训练。
EVA-Client: A Unified Data Collection, Inference, and Deployment Framework for Embodied Policies on Real Robots
- 模块解耦架构，新增机器人或推理策略只需改对应层。
- 评估即采集，每轮测试自动生成训练可用的完整轨迹数据。
- 集成多种实时推理模式，配置统一，适合真实机器人研发者使用。
我们提出EVA-Client，一个开源框架，用于在真实机器人上部署、收集数据并评估训练好的操作策略。该框架位于策略服务器与物理硬件之间，将策略迭代循环中的真实机器人阶段统一在一个代码库中。其三大贡献为：第一，组件解耦架构，机器人后端、推理策略和传输中间件构成正交网格，新增机器人或策略仅影响自身层；第二，可追溯执行，支持Debug、Collect和Eval工作流，覆盖从开环仿真到连续实时控制的多种模式；第三，每次评估自动记录完整轨迹（以训练就绪格式）、详尽日志及侧边对比查看器，使评估结果直接反馈至下一轮训练，而非成为无记录的观测。EVA-Client还整合了主要实时推理策略，包括同步与异步执行、ACT风格的时间集成、实时分块（Real-Time Chunking）以及朴素异步基线，均通过单一配置界面实现。
原文摘要 · Abstract (English)
We present EVA-Client, an open-source framework for deployment, data collection, and evaluation of trained manipulation policies on real robots. Sitting between a policy server and the physical hardware, EVA-Client unifies the real-robot stages of the policy iteration loop within a single codebase. It makes three contributions. First, a component-decoupled architecture in which robot backends, inference strategies, and transport middlewares form an orthogonal grid: adding a robot or a strategy touches only its own layer. Second, inspectable execution through Debug, Collect, and Eval workflows, with modes ranging from open-loop simulation to continuous real-time control. Third, every evaluation run doubles as a data collection, recording full rollouts in training-ready format alongside exhaustive logs and a side-by-side comparison viewer, so each evaluation feeds the next round of training rather than ending as an unrecorded impression. EVA-Client further consolidates major real-time inference strategies, synchronous and asynchronous execution, ACT-style temporal ensembling, Real-Time Chunking, and a naive-async ablation baseline, behind a single configuration surface.
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。