我想为宇树说句公道话，二次开发才是宇树机器人的灵魂
在瑜伽球上稳如老狗，安若磐石。在山地丛林越障跑酷，如履平地。
在绿茵场上带球运球，精准射门。在不同步态丝滑切换，行云流水
只需买一只二手的老款宇树机器狗Go1 EDU，上面这些人工智能和机器人领域顶尖学术会议的开源论文和算法，都可以在你自己的机器狗上运行复现。
宇树Go1机器狗论文清单：
【瑜伽球】DrEureka: Language Model Guided Sim-to-Real Transfer
https://github.com/eureka-research/DrEureka
【跑酷】Robot Parkour Learning
https://robot-parkour.github.io
【运球】DribbleBot: Dynamic Legged Manipulation in the Wild
https://gmargo11.github.io/dribblebot
【射门】Hierarchical Reinforcement Learning for Precise Soccer Shooting Skills using a Quadrupedal Robot
https://xbpeng.github.io/projects/A1SoccerShooting/index.html
【步态】Walk These Ways: Tuning Robot Control for Generalization with Multiplicity of Behavior
https://gmargo11.github.io/walk-these-ways
【AI设计奖励函数】Eurekaverse: Environment Curriculum Generation via Large Language Models
https://eureka-research.github.io/eurekaverse
