When robots join the race: 5G-A powers a new kind of marathon

Human and humanoid runners share the track as next-gen connectivity underpins real-time AI performance

Published wed 29 Apr 2026 // 07:08 UTC

SPONSORED POST The half marathon has long been a test of human endurance. In Beijing’s E-Town, it has become something else entirely. The Beijing E-Town Half Marathon and Humanoid Robot Half Marathon introduced a world-first format where humanoid robots ran alongside human participants, supported by a high-performance 5G-A network designed to handle the unique demands of embodied AI.

Behind the scenes, China Unicom Beijing and Huawei used the event as a live showcase for 5G-A GigaUplink capabilities. Large-scale sporting events are notoriously challenging for connectivity, with dense crowds generating sustained uplink demand from streaming, sharing and broadcasting. In this case, those requirements were compounded by the needs of robot runners transmitting continuous video, sensor data and navigation inputs in real time.

To meet that demand, the network delivered consistent uplink speeds of 20 Mbps surpassing 99.6% of coverage, with peak speeds reaching 677Mbps. This ensured reliable support for ultra-high-definition broadcasting and seamless user-generated content, even at peak moments.

For the robots themselves, connectivity was not just about performance, but viability. Each machine required stable uplink throughput of around 10 Mbps to support environment sensing, gait control and autonomous navigation. Dedicated network slicing and low latency, averaging under 30 milliseconds, enabled precise positioning and responsive decision-making on the move.

The result was more than a technical demonstration. It highlighted how 5G-A is shifting network priorities towards uplink performance, reflecting the growing importance of AI-driven, data-intensive applications. The event also demonstrated how intelligent network operations, including dynamic resource allocation and AI-assisted monitoring, can maintain service quality across multiple high-demand scenarios simultaneously.

As embodied AI continues to evolve, scenarios like this offer a glimpse into how connectivity, computing and intelligent systems will converge in real-world environments. From autonomous robots to immersive spectator experiences, the race in Beijing signals a broader transition where networks are no longer just supporting events, but actively enabling new forms of participation.

Sponsored by Huawei