Huajin Securities pointed out that in the current mainstream multi-sensor fusion scheme in the industry, lidar is mostly used as the main sensor because of its higher angular resolution and high imaging quality, but its price is relatively expensive. In the case of intensified cost pressure on domestic downstream manufacturers, on the one hand, 4D millimeter-wave radar with obvious cost advantages has imaging capabilities, and with the gradual improvement of performance, car companies are expected to use 4D millimeter-wave radar to replace lidar; On the other hand, with the decline in scale and raw material prices, the price of lidar has also fallen.
4D millimeter-wave radar has a strong cost advantage, and the performance has reached the imaging level
After years of development, the millimeter-wave radar industry chain has matured, chip development has become increasingly sophisticated, and algorithms have been steadily improved. This makes 4D mmWave radar a cost-effective sensor choice in the field of autonomous driving with significant potential for cost reduction. Huajin Securities pointed out that the overall price of 4D millimeter-wave radar is about 1,000 yuan, such as Arbe's imaging radar price is 690-1036 yuan, and the unit price of ZF, Continental and other Tier1 products is 1036-1381 yuan; At the same time, its price is rapidly approaching that of traditional millimeter-wave radar. At present, the domestic price of lidar is 450-500 US dollars; Compared with lidar, 4D millimeter-wave radar has more advantages in cost. Lidar is typically more expensive, while 4D mmWave radar is able to provide near-imaging-level performance while keeping costs low.
From the perspective of imaging principle, millimeter-wave radar has a longer wavelength than visible light and infrared, and is able to penetrate small obstacles and is not affected by weather conditions. This enables the 4D millimeter-wave radar to work stably in adverse weather conditions such as rain, fog, and snow. In addition, the point cloud data output by 4D millimeter-wave radar contains four dimensions (velocity, azimuth, pitch and distance), which greatly improves the point cloud density compared with traditional millimeter-wave radar. This allows the algorithm to perceive the environment more accurately.
The significant advantages of 4D millimeter-wave radar have also attracted many mainstream millimeter-wave manufacturers to deploy 4D millimeter-wave radar. According to Huajin Securities, since 2023, at least 10 4D millimeter-wave radar companies in China have received financing.
Figure: Comparison of the imaging effect of millimeter-wave radar and 4D millimeter-wave radar (Source: Huajin Securities)
Lidar prices are declining, and the trend is expected to continue
As LiDAR technology continues to advance, innovations in materials and manufacturing processes have significantly reduced costs. Economies of scale are another major factor in reducing the cost of lidar. When a technology is widely adopted across multiple industries, manufacturers can scale up production, reducing unit costs. The growing demand for lidar in the automotive industry has prompted manufacturers to invest heavily in large-scale production technologies. The increase in the number of units produced leads to a lower cost per unit, which in turn significantly lowers the price over time. Market competition is also driving down the price of lidar. For example, Hesai Group, a Chinese lidar manufacturer, plans to reduce the price of its lidar products by 50% by 2025 to boost the adoption of electric vehicles (EVs). Hesai Technology plans to achieve cost reductions through in-house chip manufacturing and optimization of factory operations. In addition, Hesai Technology plans to increase the annual production capacity of its Hangzhou plant to 1.5 million units to meet growing demand.
At this stage, there are still many problems to be solved in the application of 4D millimeter-wave radar
Technical
The quality of the point cloud needs to be improved: the current 4D millimeter-wave radar point cloud is generally one or two thousand points per frame, which is not dense enough, and needs to be increased to one or two thousand points per frame to achieve the desired effect. Moreover, the point cloud is random, flickering and fluttering, and compared with the relatively regular point cloud of lidar, the algorithm is more difficult to perform point-cloud-level fusion.
Limited angular resolution: Although the angular resolution of 4D mmWave radar is improved compared with traditional mmWave radar, the angular resolution may still be insufficient in some complex scenarios, such as accurately identifying and distinguishing multiple targets that are close to each other, resulting in the inability to accurately determine the specific position and attitude of the target.
Anti-interference capability still needs to be enhanced: Although 4D millimeter-wave radar has anti-interference capability to a certain extent, problems such as signal instability and misjudgment may occur in complex electromagnetic environments, such as near large power equipment, communication base stations, or other areas of interference from other devices in the same frequency band.
4D millimeter-wave radar has a broad application prospect in the field of autonomous driving, but it still faces many technical challenges. The solution of these problems will help promote the further development and application of 4D millimeter-wave radar technology.
Related:
Huawei Intelligent Driving Solution Market Analysis Report (1)
Huawei Intelligent Driving Solution Market Analysis Report (2)
Huawei Intelligent Driving Solution Market Analysis Report (3)
