Huawei 4D Imaging Radar
On the eve of Auto Shanghai 2021, Huawei unveiled a new generation of intelligent components and solutions, including 4D imaging radar, AR-HUD, and MDC810. Huawei’s high-resolution 4D imaging radar uses a 12T24R large antenna array (12 transmit channels, 24 receive channels), which is officially claimed to be 24 times higher than the antenna configuration of conventional millimeter-wave radar 3T4R and 50% more receive channels than typical imaging radars in the industry. High-resolution Huawei 4D imaging radar can see the car in front of the front.
It is understood that compared to LIDAR and cameras, millimeter-wave radar has excellent speed measurement capabilities, excellent weather and light robustness (unaffected by light and rain and snow), and unique detection capabilities for non-visual distance obscured targets.
The high-resolution 4D imaging radar inherits these advantages while dramatically improving resolution, target detection confidence, and detection range (e.g. distance and FOV), while evolving a high-density point cloud-like LIDAR, which can bring rich perception enhancement applications, such as environment portrayal, radar composition, positioning, etc. It can also better achieve 360° detection around the vehicle through point cloud-level fusion of multiple radars.
Full target, full coverage, full working conditions, and all-weather are the goals of the ideal autonomous driving sensor.
Huawei believes that the next evolution of millimeter wave is high-resolution 4D imaging radar, 4D refers to the detection of the four dimensions of the target, including its speed, distance, horizontal angle, and vertical angle, the traditional radar horizontal resolution capability is not enough, does not support the vertical resolution, resulting in an inability to see.
In recent years, the lack of perception is one of the important reasons for autonomous driving accidents, such as not effectively identifying stationary vehicles or accident vehicles, and inaccurate judgment of isolation piles or guardrails leading to serious accidents.
Huawei 4D imaging radar significantly improves the horizontal and vertical angle measurement capability to meet the perceived requirements of a full target, full coverage, and multiple working conditions, gradually approaching the ideal sensor target, which will form effective fusion and redundancy with camera and LIDAR. According to the introduction, Huawei’s high-resolution 4D imaging radar realizes the intergenerational improvement of capability in 3 aspects:
- The first is large array high resolution: Huawei’s high-resolution 4D imaging radar uses 12 transmit channels and 24 receive channels, which is 24 times higher than the conventional millimeter wave 3 transmit 4 receive antenna configuration and 50% more receive channels than the industry’s typical imaging radar. This is also the largest antenna configuration imaging radar available for mass production in the short term.
- The second is a large field of view without blur: The horizontal field of view is increased from 90° to 120°, the vertical field of view is increased from 18° to 30°, and the longitudinal detection distance is increased from 200 meters to more than 300 meters.
- The third is the 4D high-density point cloud: 4D point cloud of imaging radar, including velocity, distance, horizontal angle, and vertical angle, has an additional velocity dimension compared to LIDAR point cloud, which can have more dimensions to interpret the target object.
It is reported that Huawei’s high-resolution 4D imaging radar can achieve comfortable braking at 130km/h when the front is blocked; the cone bucket detection distance reaches 110 meters, and can also achieve guardrail stationary car discrimination.
Besides, obscured or partially obscured target perception is a unique capability of millimeter-wave. 4D imaging radar can see the car in front, the car in front, and the car in front of the car in front. It can even see the approximate chassis outline of the car in front of it. By perceiving the car in front of it, for example, if there is a sharp deceleration, it can anticipate the front car’s action and reduce the risk of serial tailgating.
In Huawei’s view, the 4D imaging radar will certainly bring revolutionary changes and become the next necessary asset for higher-order autonomous driving.