Google Cloud and RemotiveLabs put a whole vehicle's E/E architecture in the cloud

Google Cloud and RemotiveLabs published a joint account, dated October 20, of running a whole vehicle’s electronic architecture as software in the cloud, extending a practice already routine for cockpit stacks to the rest of the car.

The post, co-written by a Google Cloud automotive industry architect and RemotiveLabs’ chief executive, set out the gap it addresses. Gateways, body controllers, zone control units and sensor and actuator nodes, it said, are still frequently hardwired to physical chips when they are built, which pushes integration to the end of a program where defects are most expensive.

Android Automotive OS runs inside Cuttlefish, Google’s official virtual testing environment for Android, on Arm-based Google Cloud instances. The authors said the recent launch of Android 17 and AAOS SDV extends that capability to further domains. RemotiveTopology is described as the harness around the virtual vehicle: the layer that uses a manufacturer’s existing network design files to build a working virtual version of its networks, and connects the cockpit in Cuttlefish to the rest of the car’s computers, networks and signals. The authors called this vehicle-as-code, an E/E architecture expressed as an executable specification and provisioned like infrastructure.

Instruction-set parity is the stated architectural argument. Google’s Axion C4A bare-metal Arm instance is offered for automotive systems and Android software development, so the cloud and the car share an instruction set.

The post said early customer engagements with RemotiveLabs had measured more than tenfold faster integration cycles. No customer was named, and the figure is the two companies’ own.