Lidar measures distance with laser light. It fires pulses and times how long they take to return. Multiply by the speed of light and divide by two, and you have the distance. Spin the sensor or steer the beam and you get millions of points per second, forming a 3D map of the surroundings.

The term is a portmanteau of light and radar. Early lidar was bulky and expensive, used for atmospheric research and military mapping. The 2007 DARPA Urban Challenge put lidar on the map for self-driving cars. Velodyne's spinning units became the standard. Solid-state lidar is now cheaper and more compact, though spinning units still dominate high-end applications.

Lidar characteristics

• Wavelength: typically 905 nm or 1550 nm.
• Range: from a few meters to over 200 meters.
• Accuracy: centimeter-level in good conditions.
• Field of view: 360 degrees for spinning units.
• Limitations: rain, fog, and dust scatter the beam.

Lidar gives robots precise depth information. A camera tells you what something is. Lidar tells you exactly where it is. That makes it valuable for autonomous driving, warehouse robots, and drone mapping. The main drawback is cost and complexity. A mechanical spinning lidar has moving parts that wear out. Solid-state lidar has no moving parts but a narrower field of view. Camera-based depth estimation is improving, but lidar remains the gold standard for reliable 3D perception in robotics.