创造人机竞速新历史——人形机器人马拉松见证中国未来产业奔跑之路
Humanoid robot surpasses human half-marathon world record in Beijing

4月19日，破风闪电队遥控人形机器人选手闪电在比赛中冲线。4月19日，2026北京亦庄半程马拉松暨人形机器人半程马拉松在北京亦庄举行。[新华社记者 张晨霖 摄]
Humanoid robot "Flash" of Pofengshandian Team in the remote-controlled category crosses the finish line during the Beijing E-Town half-marathon and humanoid robots half-marathon in the Beijing Economic-Technological Development Area in southeast Beijing, China, April 19, 2026. (Xinhua/Zhang Chenlin)

在2026北京亦庄半程马拉松暨人形机器人半程马拉松赛事中，深圳荣耀智慧科技开发有限公司齐天大圣队自主人形机器人选手“闪电”以50分26秒的净时成绩夺冠，创造人机竞速新历史。

Humanoid robot named "Flash" from Shenzhen Honor Smart Technology Development Co., Ltd. claimed victory at the 2026 Beijing E-Town half-marathon with a time of 50 minutes and 26 seconds in autonomous navigation on Sunday, beating the human world record of 57:20.

这一成绩超越乌干达名将基普利莫在今年3月里斯本半程马拉松赛中创造的57分20秒的人类男子半程马拉松世界纪录。

The men's half-marathon world record was created by Uganda's star distance runner Jacob Kiplimo at the Lisbon Half Marathon in Portugal last month.

去年首届人形机器人半程马拉松中，天工队的人形机器人“选手”以2小时40分42秒夺冠，完赛队伍仅有6支。

At the inaugural E-Town half-marathon in 2025, Tiangong Ultra seized victory in two hours, 40 minutes, and 42 seconds, and only six out of the 20 competing teams successfully finished the 21.0975-kilometer race to make history.

今年机器人跑者的“进化”能力让人们感到惊喜，奔跑速度和稳定性均有提升。参赛队伍由20支增至百余支，国际赛队涵盖德国、法国、巴西等多国团队。

In the second edition of the event on Sunday, humanoid robots made remarkable progress and demonstrated technological breakthroughs to become faster and more stable in running. The number of participating teams increased to more than 100 this year, including participants from Germany, France and Brazil.

根据规则，人形机器人与人类跑者同时从起点出发，赛道中间设置隔离带进行安全分隔，双方共同完成21.0975公里的半程马拉松赛程。今年机器人马拉松的关门时间为3小时40分钟，参赛方式包括自主导航与遥控操作两类。

In terms of competition rules, robots and human runners followed the same route but separate lanes to ensure safety, and the robot's cutoff time was set at three hours and 40 minutes. The newly added format this year includes two ways of participation - remote control and autonomous navigation.

采用遥控方式的参赛队伍，净计时成绩需乘以1.2系数。而近四成队伍采用自主导航技术“自己找路”。

For teams in remote control, their finishing times need to be multiplied by a coefficient of 1.2. About 40 percent of the teams adopted the autonomous navigation to achieve better results.

“1.2系数的设定，旨在引导、鼓励自主导航技术研发，为人形机器人在日常生活中更多实际应用场景的落地打下基础。”中国电子学会副秘书长梁靓说。

"The setting of coefficient aims to guide and encourage the research and development of autonomous navigation, which represents a technological foundation for humanoid robots to be applied in more scenarios in our daily life towards future," explained Liang Liang, deputy secretary-general of the Chinese Institute of Electronics.