Fifteen humanoid robots have moved from pilot runs to steady shift work on a car-parts production line, a step its backers call the crossing from demo to deployment.
Gasgoo Embodied Intelligence reported on 16 October that Leju Robot and Zhaofeng Electromechanical launched a “embodied intelligence plus auto-parts feeding” solution in Hangzhou on 15 October, with 15 Kuafu humanoids entering normal operation on Zhaofeng’s parts line. The project marks, the companies say, the key leap from trial validation to scaled deployment of humanoids in car manufacturing.
What the robots actually do
The 15 Kuafu units work in a narrow, constantly changing line environment, performing more than ten actions in sequence to pick small parts with inner diameters of 80 to 120 mm and link up fetching, moving and feeding. Operating data shows a feeding success rate held steady at 98 per cent, with each cycle under 40 seconds, covering 15 stations and more than ten part types. The fleet has logged about 1,000 hours and helped produce over 60,000 parts.
A reusable system underneath
A copyable stack spans development, deployment, use, data and safety. Modular skill tools cut development from months to weeks. A multi-robot dispatch platform syncs skills with one click and shortens deployment to hours. One-click call for parts, full-loop job data return, and safety modules for dropped parts, collisions and overspeed keep the units working together reliably. Data is the other headline: 15 robots running in parallel collect over 40,000 real job records a month, which flow back through quality checks to retrain the models so the machines get sharper with use.
Alongside the launch, an “embodied intelligence plus auto-industry live training base” was unveiled, and Zhaofeng donated its real-world parts-feeding dataset to the OpenAtom OpenLET community to lower the barrier for the whole sector. Leju says the solution can now copy across stations and scenarios, and it will keep open-sourcing real-robot data and toolchains with partners to push embodied intelligence into more high-value industrial settings.
Editor’s note: This is an adapted translation of the original Gasgoo report. It has been trimmed and restructured for readability for an international business audience. The full original (in Chinese) is at Gasgoo.