ForwardX scales AMRs inside a live factory

In Dalian, 484 autonomous mobile robots operate inside a Chery plant that still produces about 1,000 vehicles per day.

ForwardX Robotics announced on June 18, 2026 that its autonomous mobile robot system is still operating and scaling inside Chery Automobile’s Dalian plant, more than a year after deployment began. The verified figure is concrete: 484 AMRs, meaning mobile robots that navigate without rails or fixed guides, are used in a facility that produces about 1,000 vehicles per day. ForwardX says the deployment covers 127 material categories, with 204 robots in the body shop and 280 in final assembly.

The useful part is not only the fleet size. It is the brownfield setting, an industrial term for an existing factory constrained by its buildings, lines, IT systems and working habits. In a new site, automation can be designed into the layout from day one. In a live factory, the task is harder: robots have to fit into a dense environment already shared by workers, forklifts, conveyors, fixed robotic equipment and production priorities that cannot simply pause for an automation project.

In the deployment described by ForwardX, the AMRs support internal logistics flows, including line-side delivery, SPS cart transport, powertrain delivery and empty-container return. The body shop reportedly covers more than 80 percent of its material demand with 32 automated categories, while final assembly approaches 90 percent coverage with 95 categories. These are supplier-provided numbers, so they should be read carefully, but they give an unusually specific public scale for an AMR fleet running inside active automotive production.

The wider point reaches beyond Chery and ForwardX. Many manufacturers will not build a new plant just to adopt mobile robotics. They will need to insert robots into facilities that already exist, with short deployment windows and little tolerance for line stoppages. That is where much of industrial robotics is now being tested: not in an isolated robot demonstration, but in the ability to coordinate hundreds of machines around a real flow of parts, people and changing production priorities.

The practical lesson is that the value of an AMR is not measured by navigation alone. It also depends on fleet orchestration, integration with manufacturing software and phased rollout without breaking production rhythm. A robot that can drive cleanly in a test aisle still has to receive tasks, avoid congestion, recover from exceptions and hand work back to humans when the line changes. That operational layer is what separates a pilot from a factory system. It is less theatrical than a humanoid demonstration, but it is much closer to the real adoption path for industrial automation.