---
title: "面向人形机器人的自适应运动规划与跟踪框架AdaPT：从广播视频学习职业网球风格"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.20087v1"
permalink: "https://feed.astralor.com/items/itm_d209575210cd712739709c99"
published: "2026-08-20T14:19:36.000Z"
score: 70
channel: "tech"
tags: ["AdaPT","人形机器人","仿真到现实迁移","宇树科技","自适应运动规划","越疆科技"]
---

# 面向人形机器人的自适应运动规划与跟踪框架AdaPT：从广播视频学习职业网球风格

## AI 点评

> 这篇论文最值得关注的地方在于，它没有只追求仿真里的花哨动作，而是认真处理了真实机器人上“风格越复杂、误差越容易滚雪球”的痛点。对做机器人控制和sim2real的团队来说，AdaPT的分层思路和速度适配器设计提供了很直接的参考；对关心具身智能产品化的读者，Unitree和Dobot上的真机演示也证明了这项技术距离实用并没想象中那么远。

## 速览

人形机器人在球类运动上已展现潜力，但要在保持任务性能的同时实现专业动作风格仍困难。研究者提出AdaPT——自适应运动规划与跟踪框架，从广播视频直接学习职业网球发球与对打风格。其分层设计让规划器生成风格化运动，跟踪器以最小干预执行。针对仿真到现实的性能衰减，通过学习随机化执行速度的跟踪和运动速度适配器缓解误差累积。在Unitree G1和全尺寸Dobot Atom上的实验验证了有效性。

## 关键点

- AdaPT采用分层架构：规划器负责生成具有风格的运动轨迹，跟踪器负责执行并最小化对规划目标的干扰。
- 提出适应机制，通过随机化执行速度训练跟踪器，并结合运动速度适配器缓解规划中的误差累积。
- 在Unitree G1和1.7米高的Dobot Atom全尺寸人形机器人上完成真实世界验证，无需动作捕捉即可完成户外发球。

## 全文

Humanoid robots have recently demonstrated promising capabilities in real-world ball sports. However, achieving professional motion styles while maintaining strong task performance remains challenging. In this work, we propose AdaPT, an Adaptive Motion Planning and Tracking framework that learns professional tennis serving and rally styles directly from broadcast videos. This hierarchical design is motivated by the key insight that the planner generates stylistic kinematic motions, while the tracker executes them with minimal interference with planning. Despite its effectiveness in simulation, a substantial sim-to-real gap emerges: tracking performance inevitably degrades on real robots, and this degradation is partially overlooked by autoregressive planning and further compounded by noisy perception. To address these issues, our adaptation mechanism improves tracking robustness by learning to track randomized execution speeds, while conditioning the planner on a learned motion-speed adapter to mitigate compounding errors. Real-world experiments on the Unitree G1 demonstrate the effectiveness of our adaptation mechanism in bridging the sim-to-real gap. We further deploy AdaPT policies on the full-size Dobot Atom humanoid robot (1.7m) and demonstrate in-the-wild serving without motion capture. Beyond these results, our real-world experiments reveal both algorithmic and engineering insights for future humanoid ball-sports systems. Videos and code are available on our \href{https://humanoidtennis.github.io/AdaPT/}{project website}.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_d209575210cd712739709c99