Application of Lidar in the field of Atmospheric Detection And Target Capture

Energy saving and environmental protection, new energy vehicles, intelligent vehicles has become the direction of China's automobile development, intelligent vehicles is mainly the development of autonomous intelligent vehicles and network-connected intelligent vehicles. Autonomous intelligent vehicles and network-connected intelligent vehicles to achieve the above functions can not be realized only by assembling ordinary microwave radar, to achieve the above functions, LiDAR is an essential component in the control system of the electrical system of autonomous intelligent vehicles and network-connected intelligent vehicles.

Lidar has the characteristics of narrow beam, small size, non-contact measurement, etc. It can detect cloud, aerosol, air wind field, air pollutants, temperature and humidity changes and other parameters. It uses optical frequency band for detection, which is several orders of magnitude higher than millimeter wave, and the detection accuracy is more advantageous than microwave radar. Therefore, LiDAR has a broader application prospect in the field of atmospheric detection and target capture.

At present, the scanning LIDAR that can be used in intelligent vehicles are mechanical rotating LIDAR, microelectromechanical system scanning radar and phased array LIDAR. Mechanical rotating LiDAR (LiDAR that transmits, receives, and rotates on a common axis) is relatively mature at present, and there are already unmanned concept cars trying mechanical rotating LiDAR devices. Scanning radar based on microelectromechanical systems (MEMS) currently belongs to the state of the art research, which is based on the principle of changing the optical path through MEMS scanning mirrors. Phased-array lidar is realized by point-by-point scanning, i.e., changing the optical path by the emission phase of the laser light emitted between multiple small antennas. A surface-array LIDAR emits a surface array of light, and the main problem is the short detection range.

In time-of-flight ToFLiDAR, the laser emits light pulses of duration τ that activate the internal clock of the timing circuit at the instant of emission. The light pulse reflected from the target arrives at the photodetector with an output electrical signal that disables the clock. This electronic measurement round trip ToFΔt calculates the distance from the target to the point of reflection R. If in reality the laser and the photodetector are located at the same position, the distance R is influenced by two factors: c is the speed of light in a vacuum, and n is the refractive index of the propagation medium (which is close to 1 in air). These two factors influence the distance resolution ΔR: if the diameter of the laser dot is larger than the size of the target to be resolved, the uncertainty in the measurement Δt and the spatial width of the pulse w (w = cτ) is ΔΔt. In a typical automotive LiDAR system, the laser produces pulses with durations of about 4 ns, and therefore a minimum beam divergence angle is necessary.

The most critical aspect for the designer of an automotive LiDAR system is the selection of the wavelength of the light. The two most popular wavelengths are 905nm and 1550nm, which is a complex issue for automotive LiDAR due to the numerous weather conditions and reflective surface type possibilities. In real world environments, there is actually less light loss at 905nm due to stronger absorption at 1550nm than at 905nm.

Lidar-related technology algorithms in the traditional key modules still need to achieve better lightweight, accuracy, robustness, and generalization, semantic maps and the integration of deep learning has become a trend, and other sensors that can achieve autonomous localization of the source, such as depth camera, millimeter-wave radar, etc., multi-source fusion is also a hot spot of the current research, and Lidar-related technology for unmanned platforms to achieve the development of autonomous intelligence is bound to have a LIDAR related technology will have a far-reaching impact on the development of unmanned platforms to realize autonomous intelligence.