I3L: Iterative Interactive Imitation Learning for Whole-Body Mobile Manipulation
作者: Wensi Ai, Christopher Luey, Karthik Dharmarajan, Evans Han, Yunfan Jiang, Stef Ren, Jianwen Xie, Jiajun Wu, Li Fei-Fei, Ruohan Zhang
分类: cs.RO, cs.AI
发布日期: 2026-10-13
备注: Project website: https://i-3l.github.io/
🔗 代码/项目: PROJECT_PAGE
💡 一句话要点
提出I3L以解决长时间全身移动操控的可靠性问题
🎯 匹配领域: 支柱一:机器人控制 (Robot Control) 支柱二:RL算法与架构 (RL & Architecture)
关键词: 全身移动操控 离线训练 策略优化 机器人技术 迭代学习
📋 核心要点
- 现有的离线训练策略在长时间全身移动操控中表现不佳,尤其在训练分布外容易出现性能下降。
- I3L通过结合全身修正与迭代策略优化,利用JoyLo+接口实现协调修正,提升了策略的可靠性。
- 实验结果显示,I3L在多个任务中平均提升自主成功率68.3个百分点,且每个任务的成功率均超过90%。
📝 摘要(中文)
本论文提出I3L,一个集成硬件与学习系统,旨在解决离线训练策略在长时间全身移动操控中的可靠性缺口。现有方法在训练分布外表现不佳,尤其是在长时间任务中,误差累积和协调运动使得全面示范覆盖变得不切实际。I3L结合了全身修正与迭代策略优化,通过低成本的JoyLo+接口实现了协调修正,并通过迭代部署与再训练来针对策略的失败进行调整。实验结果表明,I3L在多个机器人和任务中显著提高了自主成功率,平均提升68.3个百分点,且在每个任务上均达到了90%的成功率。
🔬 方法详解
问题定义:本论文旨在解决离线训练策略在长时间全身移动操控中的可靠性问题。现有方法在训练分布外表现不佳,导致误差累积和协调运动的挑战,使得全面示范覆盖变得不切实际。
核心思路:I3L通过集成硬件与学习系统,结合全身修正与迭代策略优化,旨在缩小离线训练与可靠部署之间的差距。设计上,I3L利用低成本的JoyLo+接口实现策略跟踪和触发修正,从而增强了策略的适应性。
技术框架:I3L的整体架构包括多个模块:首先是策略训练模块,利用示范数据进行初步训练;接着是全身修正模块,通过JoyLo+接口实现协调修正;最后是迭代部署与再训练模块,针对策略失败进行调整与优化。
关键创新:I3L的主要创新在于其迭代优化机制和全身协调修正能力,这与传统的示范训练方法有本质区别,后者往往无法有效应对长时间任务中的动态变化。
关键设计:在设计上,I3L采用了低成本的硬件接口,设置了策略跟踪的领导臂,并通过触发机制实现了高效的修正。此外,损失函数和网络结构经过精心设计,以确保在迭代过程中能够有效捕捉策略的失败点。
🖼️ 关键图片
📊 实验亮点
实验结果表明,I3L在四轮HG-DAgger协议下,平均提升自主成功率68.3个百分点,且在每个任务上均达到了90%的成功率。与仅依赖示范训练的方法相比,I3L的表现提升了35.6个百分点,且每集的接管时间减少了72%。
🎯 应用场景
I3L的研究成果在机器人操作、智能制造和服务机器人等领域具有广泛的应用潜力。通过提升长时间全身移动操控的可靠性,I3L能够在复杂环境中实现更高效的自动化操作,进而推动智能机器人技术的实际应用与发展。
📄 摘要(原文)
Policies trained offline from teleoperated demonstrations can degrade outside their training distribution. This gap is particularly acute in long-horizon whole-body mobile manipulation, where long horizons amplify compounding errors and coordinated arm, torso, and base motion makes exhaustive demonstration coverage impractical. We introduce I3L, an integrated hardware and learning system that combines whole-body correction with iterative policy refinement to address this last-mile reliability gap. Its low-cost, open-source JoyLo+ interface uses policy-tracking leader arms and touch-triggered takeover to enable coordinated corrections across the body parts, while iterative deployment and retraining allow operators to target failures of the updated policy. We evaluate I3L across three robot embodiments and eight simulated and real-world tasks, including four long-horizon whole-body mobile manipulation tasks. Controlled simulation experiments show that four-round HG-DAgger achieves the highest final success on both tasks. Under this protocol, I3L improves autonomous success over the base policies by an average of 68.3 percentage points and achieves at least 90% success on every task, outperforming demonstration-only training by 35.6 percentage points on average with the same number of training trajectories. Averaged across tasks, takeover duration per episode decreases by 72% over four rounds. Together, these results show that I3L helps close the last-mile gap between offline training and reliable deployment and provides practical insights into collecting and using corrections for long-horizon whole-body mobile manipulation. Videos, visualizations, and appendix are at https://i-3l.github.io/.