Its basic principle is to utilize the propagation characteristics of light, calculate the distance between the target and the sensor by measuring the time difference or phase difference of the laser pulse from emission to reflection back to the receiver, and then construct a three-dimensional point cloud map by combining this with a scanning mechanism. This article will systematically analyze the working principle, technical classification, core modules, and application characteristics of lidar.
Basic working principle: Time-of-flight and spatial scanning of light
The core function of lidar is to achieve "range measurement" and "3D modeling". Its workflow can be divided into three stages: laser emission, target reflection, and signal reception and processing.
The ranging principle is the cornerstone of lidar technology, mainly divided into two categories: Time-of-Flight (TOF) method and triangulation method. The TOF method calculates distance by measuring the propagation time of a laser pulse in the air. The formula is: Distance D = (Speed of light c × Time of flight t) / 2 (dividing by 2 is because the optical path is a round trip). For example, when the time from laser emission to reception is 100 nanoseconds, the corresponding distance is approximately 15 meters (c = 3 × 10⁸ m/s, t = 100 × 10⁻⁹ s, D = 3 × 10⁸ × 100 × 10⁻⁹ / 2 = 15 m). The TOF method can be further divided into pulse-based (directly measuring the pulse's round-trip time) and phase-based (calculating time by measuring the phase difference between the emitted and received lasers). The former is suitable for long-distance measurements (hundreds of meters), while the latter has higher accuracy (millimeters) but a shorter range (tens of meters).
The triangulation method calculates distance using geometric relationships: a laser transmitter and receiver are mounted at a fixed angle. After the emitted laser beam illuminates the target, the reflected light forms a spot on the imaging plane of the receiver. The distance is calculated based on the position of the spot and the baseline length (distance between the transmitter and receiver) using the principle of trigonometric similarity. This method is suitable for close-range (0.5-50 meters) and high-frame-rate scenarios (such as industrial inspection) because it does not require high-precision time measurement. However, it is limited by the size of the optical system, making it difficult to achieve long-range detection.
Spatial scanning mechanisms endow LiDAR with three-dimensional perception capabilities. By altering the propagation direction of the laser beam through mechanical rotation, electronic steering, or other methods, the laser beam covers the target area along a specific trajectory (such as a spiral or grating). Each laser pulse corresponds to a spatial point (x, y, z), and the collection of numerous points forms a point cloud. The scanning frequency (points per second) determines the point cloud density; for example, autonomous driving LiDAR typically requires ≥2 million points per second to ensure accurate identification of fast-moving targets.
