There has been a lot of interest recently in the development of next generation radar sensors for automotive and industrial applications. Unlike first-generation radar systems, which collected only speed, range, and angle-of-arrival, state-of-the-art 4D radar systems can determine an object’s position in range, azimuth, elevation, and relative speed, providing more detailed information.
4D radar sensors provide better resolution than previous radar versions and equal the performance of a LiDAR system, which is a much more expensive solution and suffers in poor visibility conditions such as rain and fog. By offering all-weather sub-degree horizontal and elevation spatial resolution on long-range around a wide field of view, 4D enhanced resolution and sensitivity unlock the potential of market-proven commercial radars.
The technology on which 4D radar systems are based can tell when and how fast a vehicle is moving in all types of weather and environmental conditions. Cameras can be harmed by strong sunlight or darkness, and depth and contrast issues can arise. In bad weather, standard radar works fine, but it can’t detect objects in great detail. LiDARs are excellent for identifying objects in space. However, they are hampered by adverse weather.
Steradian 4D Imaging Radar
Steradian Semiconductor, a fabless semiconductor company headquartered in Bangalore (India), is focused on CMOS millimeter wave products for 4D radar imaging. Founded by industry experts with several years of experience in designing cellular/RF and microwave transceiver ICs, Steradian provides a multi-patented hardware and perception software 4D solution. The target markets for Steradian’s products include applications ranging from automotive to industrial.
“We are primarily a solution provider, focusing on perception software and the hardware side,” said the company’s spokesperson. “That means we provide both the hardware, which includes semiconductors, modules, antennas, and the software solution.”
Today, most of cars offer advanced driver assistance systems (ADAS), corresponding to level 2 of the driving automation scale. Autonomous or semi-autonomous vehicles require sensors able to provide a high level of safety that, for automotive applications, means the sensor shall be all-weather reliable and not be affected by weather conditions such as sun, rain, or fog. None of the above-mentioned parameters should impact the sensor behavior, and that is one of the main reasons why radar systems have been used in cars, from almost two decades.
Compared to a LiDAR solution, 4D radar imaging provides three major benefits:
- Cost: Radar cost is a fraction of LiDAR.
- Safety: Radar provides 300m+ visibility under all weather conditions.
- Reliability: Since 2019, more than 30 million conventional radars have been shipped.
Radar has a long history of use in the automotive industry. It is not a new technology like LiDARs or cameras, which which have appeared in recent years. As an example, in order to get the highest safety rating (5 stars) of the European New Car Assessment Programme (Euro NCAP), radar systems become mandatory.
“Next-generation sensors are expected to provide that safety feature and, at the same time, more real-time information about the surrounding,” said the company’s spokesperson.
According to Steradian, the current generation of radars, which is also referred to as 2.5D radar, does not provide any other information than the presence or absence of an object. It is a kind of binary information, and it lacks details such as the width, the height, or the shape of the object in front of the vehicle. Today, LiDARs and cameras are doing that, but this is not possible with conventional radars.
“This is where multiple dimensions need to come into the radar system so that it can identify the size, as well the posture of the objects, and being able to do much more than what it is doing today,” said the company’s spokesperson.
Steradian 4D imaging radar includes a front end operating in millimeter wave (mmWave) in the 76 to 81 GHz band, which pairs up their custom radar perception software. The radar IC (SVR) is an integrated 16 channel E-band MIMO radar transceiver that interfaces with the Imaging Radar Unit (IRU), shown in Figure 1, an E-band 256 channel MIMO Radar Module. The IRU is capable of 4D imaging thanks to its antenna array generating a 1.2° azimuth beam width and supports multimode through its wide field of view (FOV), 120° in azimuth and 30° in elevation.