4D imaging radar, unlike conventional radar
Measurement of speed, distance, altitude, azimuth, and relative velocity
Xilinx FPGAs for Real-Time Radar Sensor Aggregation


Unlike conventional automotive radars that only capture speed and azimuth, 4D imaging radar additionally utilizes distance, altitude, and relative speed to determine object location, providing more detailed driving environment information.

Continental announced on the 24th that it has developed the 'ARS540 (Advanced Radar Sensor 540)' 4D imaging radar based on Xilinx's Zynq® UltraScale+™ MPSoC platform, which provides a detection range of up to 300m and ±60°.

The ARS540 supports SAE Level 2, where the driver supervises vehicle control, and can further expand the sensor's application range by expanding to Level 5, which is fully autonomous driving.

“Xilinx’s Zynq UltraScale+ MPSoC platform provides the performance and DSP capabilities needed to implement the ARS540,” said Norbert Hammerschmidt, head of program management radar at Continental. “It provides a network capable of processing a wide range of antenna data with high adaptability and throughput.”He explained the reason for adoption, saying, “It has a network interface.”

Xilinx's automotive-grade (XA) Zynq UltraScale+ MPSoC is an adaptable platform, enabling Continental's 4D imaging radar to be agnostic to multiple sensor platform configurations and tailored to OEM specifications.

Parallel processing within the device's programmable logic enables fully independent, simultaneous processing pipelines, crucial for the ARS540's 4D sensing. Additionally, numerous DSP (Digital Signal Processing) slices enable hardware acceleration of real-time radar sensor input.

Meanwhile, Cédric Malaquin, technology and market analyst for RF devices and technology at Yole Development, said, “4D imaging radar will initially be applied to luxury cars and robotaxis,” and predicted that “the market size will grow at a compound annual growth rate of 124% from 2020 to 2025, reaching $550 million.”