In recent years, solid-state radar has developed towards miniaturization, software, and ASIC integration. According to SinoLink Securities, while the lidar is solidified, the volume of the product is getting smaller and smaller, from a huge mechanical rotating lidar on the roof of the car in the initial test stage, gradually developing to the size of a mobile phone and even hiding around the body, eliminating mechanical parts. On the other hand, lidar manufacturers have changed from simply selling hardware before, to gradually matching software algorithms, and then packaging complete solutions.
Furthermore, at present, lidar manufacturers are trying to integrate ASIC, packaging hundreds of electronic components such as laser emitters, detectors, and amplifiers into ASIC special chips, and using a single chip to achieve overall control, which can effectively reduce components, reduce volume, reduce power consumption, and greatly reduce costs. An ASIC is an integrated circuit that is designed and customized for specific application needs. The design and manufacturing process of an ASIC involves multiple stages, including requirements analysis, architecture design, RTL design, verification, logic synthesis, place-and-route, mass production preparation, and follow-up maintenance. The main advantages of ASICs include high performance, low power consumption, high integration, high security and confidentiality, but they also have disadvantages such as long design cycles, high manufacturing costs, low flexibility, and high risks.
Figure: Quanergy software solutions
Millimeter-wave radar: all-weather service is indispensable
Compared with lidar, millimeter-wave radar technology is very mature, and it can be used in adaptive cruise control as early as the 90s of the last century. With its absolute advantage of being able to penetrate dust, fog, rain and snow, and not be affected by bad weather, millimeter-wave radar has become an indispensable main sensor for autonomous driving. At present, the top manufacturers in the millimeter-wave radar market are Bosch, Continental, Delphi, etc., but with the continuous improvement of ADAS penetration, independent brand car manufacturers have gradually applied various ADAS modules for loading, and domestic millimeter-wave manufacturers have also risen. At present, millimeter-wave radar is basically monopolized by foreign Tier 1 suppliers, and Tier1 in this part of the market is closely bound to mainstream car manufacturers.
Figure: The world's major vehicle-mounted millimeter-wave radars
In-vehicle cameras: ADAS mainstream sensors, glasses for autonomous driving
According to SinoLink Securities, the camera technology is the most mature, and the on-board application started relatively early, as an absolute visual sensor in the ADAS stage, 4 to 8 cameras are needed according to different functions, and the range of applications is very wide. Currently, in-vehicle cameras are moving towards higher pixels to provide sharper images to meet the needs of higher levels of autonomous driving. For example, the 8-million-pixel camera has begun to be adopted by some new energy vehicle companies, such as NIO, Ideal, etc. In addition, the optical design of the automotive camera has been continuously optimized to improve imaging stability, including the elimination of stray light and ghosting, the improvement of night vision, and LED flicker suppression.
Judging from the analysis of SinoLink Securities, the multi-sensor solutions of major car companies are aimed at active layout and mass production of the most reasonable solutions. At present, most models are equipped with ADAS functions, and the solution includes several cameras, millimeter-wave radar and ultrasonic sensors, etc., and a few solutions are installed or reserved for lidar positions. Compared with the production models, it is not difficult to find that the vast majority of car companies have gradually "evolved" from ADAS functions to more advanced autonomous driving functions in a step-by-step manner. At the same time, it is also in line with the development of autonomous driving technology, the improvement of government policies and regulations, and the acceptance of consumers.
Relateds:
Development Trend of Multi-Sensor Fusion for Autonomous Driving (1)
Development Trend of Multi-Sensor Fusion for Autonomous Driving (2)