LiDAR Time-of-Flight Sensor

Light leaves, bounces off a wall 100 metres away, and returns — and the sensor measures that 667-nanosecond round trip to within tens of picoseconds.

Fires short pulses of near-infrared laser light and measures the time each pulse takes to return after reflecting from a surface. Since light travels at a known speed, that elapsed time directly encodes distance. Scanning the laser beam — mechanically, with a MEMS mirror, or via a phased array — turns single-distance samples into dense 3D point clouds used for autonomous navigation, robot perception, mapping, and obstacle avoidance.

A driver circuit fires a laser diode in bursts of 2–20 nanoseconds. The outgoing pulse is timed by a reference clock. Reflected photons arrive at a detector — typically an avalanche photodiode (APD) or single-photon avalanche diode (SPAD) — which generates an electrical pulse on photon arrival. A time-to-digital converter (TDC) measures the interval between emission and detection with a resolution of tens of picoseconds. For a 100-metre round trip at the speed of light, the flight time is approximately 667 ns; a 10-ps timing error corresponds to a 1.5-mm range error. SPADs can trigger on single photons, enabling detection even at very long range or low reflectivity, but require quenching circuits to reset between detections. Modern sensors fire many laser channels simultaneously or in rapid sequence to build full 3D scans at 10–20 Hz.

In plain terms

Imagine standing in a dark cave and shouting. You count the milliseconds until you hear the echo — and from that you know how far the wall is. LiDAR does exactly the same thing with flashes of light instead of sound, but millions of times per second and with enough precision to measure the distance to something the size of a tennis ball a football field away.

Also called: pulsed ToF LiDAR, direct time-of-flight LiDAR, dToF LiDAR, laser rangefinder.

Why it matters

Where it lives

Key numbers

Surprising facts

A short history

The first LiDAR prototype was built in 1961 at Hughes Aircraft Company — the same laboratory where Theodore Maiman had demonstrated the first working laser only a year earlier. Hughes introduced the Hughes Mark II Colidar commercially in 1962. Pulsed laser ranging for space applications came of age with the Apollo 15 Lunar Laser Altimeter in 1971, which measured the Moon's surface elevation profile from 100 km orbit at roughly one point per second and 2–3 m accuracy. Terrestrial LiDAR evolved steadily through airborne topographic mapping in the 1990s. The pivotal moment for robotics came in 2005, when David Hall designed the first spinning multi-beam 3D sensor for Velodyne, used in the DARPA Grand Challenge. The 64-beam HDL-64E debuted at the 2007 DARPA Urban Challenge and became the defining sensor of a decade of autonomous-driving research. Simultaneously, solid-state and MEMS-mirror designs reduced cost and size; Hesai's AT128 automotive-grade sensor entered mass production in 2022, combining 200-metre range with 1.53-million-points-per-second measurement frequency.

Sources

Related