A key technology for robotic mobility and autonomous driving is pulsed laser scanning lidar. It computes depth and performs three-dimensional imaging using pulses of directed light. However, innovations are required to improve FoV, imaging frame rate, ambiguity range, and lower manufacturing costs. A research team suggests a creative alternative to satisfy the demands of automotive lidar.
The researchers demonstrate an experimental pulsed metasurface-scanning lidar with a very high field of view. It cascades a laser diode with a metasurface to increase the field of view (FoV) by up to 150° in both the horizontal and vertical axes. Additionally, the narrow FoV of the deflector is continually expanded by varying the optical characteristics of the metasurface. Finally, the system’s detecting component uses a very sensitive photodetector digitalized using an analog-to-digital converter.
Despite operating on a straightforward physical concept, pulse scanning lidar often has low SNR and poor accuracy for objects positioned at great distances inside the ambiguity range or the greatest measurable distance. Inherent to d-ToF imaging is a trade-off between speed and ambiguity range. In response to these problems, the researchers suggest a unique imaging method that mitigates the pulse scanning lidar’s trade-off.
The researchers point out that the new pulse scanning lidar almost satisfies the specifications for automotive lidar and shows potential for unique applications by harnessing the skills of metasurfaces. It can potentially be shrunk to chip-scale dimensions and is compact. This downsizing will create new opportunities and intriguing potential for the autonomous car and robotic sectors.
Related Content: LiDAR Based On Diode-Pumped Alexandrite To Monitor Upper Atmosphere