TECH Signal 508
Waymo Waymo CEO explains why Tesla's camera-only self-driving falls short
↗
Waymo’s co-CEO argued that a camera-only perception stack cannot achieve the safety levels required for fully driverless operation.
Engineers building autonomous systems must recognize that relying solely on vision limits reliability, especially in adverse lighting or weather. Multi-sensor fusion with lidar and radar extends perception into conditions where cameras fail, raising the overall safety “nines”. The cost advantage of a camera-only design may be offset by the need for additional safety engineering and the risk of hitting a performance ceiling.
Written by elseif from the cluster below · every claim links back to a sourceThe three things worth knowing
Cameras can approximate human driving ability but hit a safety ceiling before reaching superhuman performance needed for full autonomy.
Waymo combines cameras, lidar, and radar, each providing distinct data that remains usable in darkness, glare, dust, rain, or when the lens is obstructed.
Achieving higher reliability levels (additional “nines”) requires exponentially more effort, and the price of lidar has fallen but still represents a non-trivial hardware investment.
THE CLUSTER