# Skild AI 用 140 年自博弈训练人形机器人踢足球

- 来源：Rohan Paul (@rohanpaul_ai)
- 发布时间：2026-09-24 01:44
- AIHOT 分数：48
- AIHOT 链接：https://aihot.news/items/cmuef56t607cwrox36vu2d3w1
- 原文链接：https://x.com/rohanpaul_ai/status/2102816500235337860

## AI 摘要

Skild AI 训练出一台会踢足球的人形机器人 Messinator，靠自博弈在虚拟世界杯中累计训练 140 年，与自身历史策略版本反复对抗并从得分动作中学习。机器人每秒输出 50 次关节角度指令，而非从离散动作空间中挑选，需同时处理平衡、控球、碰撞恢复、对手建模、运动与策略。

## 正文

This is just so tough for robots.
The balance, ball control, collision recovery, opponent modelling, locomotion, and strategy all have to run at once.

here the humanoid is producing joint-angle commands 50 times a second, not choosing from a neat discrete action space.

The training setup is even more interesting, it basically spent 140 simulated years getting cooked by itself.
it played millions of accelerated virtual soccer games, repeatedly competing against saved earlier versions of its own policy and learning from which actions helped it score.

### 引用推文

> Skild AI：We trained a robot to play football. How? Self-play for 140 years in a virtual World Cup. Meet the #Messinator: