Wayve Robotaxi: How a Cambridge Startup Is Rivaling Waymo Without a Single LiDAR

Most self-driving programs lean heavily on LiDAR to build a precise 3D map of the world around the vehicle. Wayve has taken the opposite bet: an end-to-end model trained largely on camera video, reasoning about the road the way a human driver does rather than fusing several dedicated sensor modalities.

That architectural choice has real trade-offs. Camera-only systems are cheaper to deploy at scale and generalize more easily to new cities without pre-mapping, but they ask more of the underlying model in adverse weather and low-light conditions. Wayve's Cambridge, UK origins — outside the Bay Area's established self-driving cluster — has also shaped a research-first culture that leans on published papers as much as pilot fleets.

Whether an end-to-end, camera-first approach can match the reliability of LiDAR-anchored stacks from incumbents like Waymo is still being tested on public roads, but Wayve has become one of the most closely watched challengers in the space.