LiDAR, or light detection and ranging, measures distance with laser light and uses those measurements to construct information about the surrounding environment. In autonomous vehicles, it contributes to a 3D map used alongside other sensors. section 4.1.2
LiDAR fits into an autonomous system as one part of a larger sensing and decision stack:
- Radar uses radio waves to detect objects and their speeds, particularly in poor visibility. section 4.1.2
- Cameras provide visual information for object recognition and lane detection, while ultrasonic sensors handle nearby objects during parking and low-speed maneuvers. section 4.1.2
- Processing units, perception algorithms, localization, planning, and control systems turn sensor data into real-time driving decisions. section 4.1.2
- Applications extend beyond autonomous driving to drones, smart cities, industrial automation, and research by geologists, geographers, and archaeologists. section 4.2
Common system designs include time-of-flight, frequency-modulated continuous wave (FMCW), and solid-state LiDAR. Solid-state systems avoid moving parts, potentially reducing wear and size, while vendors trade range, resolution, accuracy, scalability, and cost. section 4.2 section 4.2.5
Technology and deployment comparison
The investigated systems illustrate different designs: Aeva uses FMCW; Luminar uses hybrid solid-state LiDAR with emphasis on long-range 1550 nm lasers; Innoviz uses MEMS-based solid-state LiDAR; and Ouster uses digital solid-state LiDAR based on VCSELs. Deployments include Volvo, BMW, Daimler Trucks, and Zoox, while Ouster also targets smart-city adoption. section 4.2.5
LiDAR remains strategically uncertain. Its cost makes camera-only alternatives attractive, while automotive adoption depends on performance, regulation, partnerships, and the difficulty of handling edge cases across different operating conditions. section 4.2 section 4.2.5 section 4.3.1
