A new kind of traffic officer went on duty at the 2026 Hangzhou Marathon on November 1 in China. Traffic and security humanoid robots developed by Hangzhou-based DEEP Robotics, together with local traffic authorities, joined a four-point patrol team around the Huanglong Sports Center in Hangzhou, Zhejiang Province.
The model in operation is the Hangxing No.2, based on DEEP Robotics' DR02 humanoid robot. According to the company, the robot is 1.78 metres tall, carries an IP66 rating against dust and water jets, and operates in temperatures from -20 C to 55 C. It is equipped with a 275 TOPS computing unit and a multi-sensor fusion system with LiDAR, a depth camera and a wide-angle camera.
Its tasks include intersection monitoring, autonomous patrols on preset routes, traffic violation warnings and public inquiry services. The robot detects situations such as pedestrians crossing the road and vehicles cutting in, in real time, performs standardized traffic control gestures and broadcasts traffic safety advisories, covering blind spots left by fixed surveillance.
The event drew more than 300,000 applicants, with 36,000 runners taking part. For DEEP Robotics, dense crowds over a short period and highly repetitive tasks make a marathon a good first scenario for robots, and the company describes the deployment as an exploration of a standardized model for humanoid robots in traffic management at large events.
The robot is not presented as a replacement for human officers: according to the company, it takes on repetitive roadside advisory tasks and reduces the workload of frontline officers, turning fixed surveillance points into mobile patrol units. A precedent cited in the report is the Beijing Yizhuang Half Marathon in April, which also used embodied-intelligence traffic robots.
Image credit: DEEP Robotics, via People's Daily. The image shows the traffic robots on duty at the 2026 Hangzhou Marathon.
Sources: People's Daily, available at People's Daily, with statements from DEEP Robotics and the Global Times reproduced in the report.
Transparency note: this article was produced with the assistance of artificial intelligence, cross-checked against public sources on the internet, and reviewed editorially. Please consult the original sources to confirm the details.
Muse
