Sensors: When the Car Sees Around the Corner

A new type of radar system makes it possible for vehicles to “see” around corners and thus recognize other road users in time. Sooner or later, the optimization of the traffic flow might also be conceivable.

Mit entsprechenden Systemen kann das Auto um die Ecke schauen. Foto: Popp/DEKRA

Future music: With the new radar system, cars will be able to look around the corner. Photo: Popp/DEKRA

To increase comfort and safety, information and assistance systems have been standard in modern vehicles for years. Whether it’s a navigation system with congestion avoidance recommendation, proximity cruise control, lane departure warning, emergency brake assistant, blind spot assistant, fatigue warning, camera-based active light systems, night vision assistant, driving dynamics control, and much more: “All of these systems contribute to informing and supporting the driver and, if necessary, compensating for his or her misconduct in order to further reduce the number of people killed and injured in road traffic,” says Peter Rücker, Head of DEKRA Accident Research.

The vehicle picks up objects outside the direct field of view

Much of this equipment requires additional sensor technology – from ultrasonic sensors to radar and cameras. In addition, the vehicle must evaluate all information in real time and then make decisions. High-tech is also the basis of a new type of radar system that scientists from the universities of Princeton (New Jersey, USA), Ulm, and Kassel have developed together with engineers of the Mercedes-Benz AG: The system enables so-called Non-Line-of-Sight (NLOS) imaging and allows vehicles to “see around corners” and to pick up objects that are outside of their direct field of view. “This is made possible by the evaluation of reflected radar waves using artificial intelligence,” explains Felix Heide, Professor of Computer Science at Princeton University.

Mit Doppler-Radarmessungen unter Verwendung statischer Gebäudefassaden oder geparkter Fahrzeuge als Relaiswände ist es möglich, sich bewegende Objekte außerhalb der direkten Sichtlinie in großen Fahrzeugumgebungen zu detektieren. Foto: Daimler AG

With Doppler radar measurements using static building facades or parked vehicles as relay walls, it’s possible to detect moving objects outside the direct line of sight in large vehicle environments. Photo: University of Princeton

With this NLOS method, the Doppler radar system installed in the vehicle points radio waves against a static building facade or parked cars, for example, so that signals reflected by them hit an object located around the corner. “The radio waves reflected by the Doppler radar system bounce back onto the wall and a computer calculates an image of the vehicle’s surroundings. A sophisticated algorithm even makes objects such as cyclists or pedestrians visible in the indirect field of vision,” continues Felix Heide.

A warning message is shown on the head-up display

The system, which can be used in cars, vans, or trucks, also detects whether the concealed object is moving. A warning message is then shown on the head-up display in the windshield or on the vehicle display and the driver can reduce his speed in time to avoid a collision. It’s also possible to connect to the emergency brake assistant, which then automatically brakes the vehicle or brings it to a complete stop.

Das neben einer GNSS-Antenne (GNSS = Global Navigation Satellite System) unter anderem mit einer inertialen Messeinheit (IMU), Lidar- und Radarsensoren sowie einer Kamera ausgestattete Testfahrzeug von Mercedes-Benz. Foto: Daimler AG

In addition to a GNSS antenna (GNSS = Global Navigation Satellite System), the test vehicle from Mercedes-Benz is equipped with an inertial measurement unit (IMU), lidar and radar sensors, and a camera. Photo: Daimler AG

“The technology is basically ready for series production, and test vehicles configured accordingly by Mercedes-Benz have shown that the system works,” reports Felix Heide. Numerous other automobile manufacturers and suppliers have already expressed great interest. However, Nicolas Scheiner, project member and responsible for radar-based detection of road users at Mercedes-Benz, stresses that there are very high standards for series production of systems in the automotive industry: “Yes, our concept study has shown very promising results on a test vehicle. However, these are initial research results. Many more iterations and studies will be needed to make the system ready for series production.”

Sooner or later, the technology could be used to optimize traffic flow. Because by detecting the environment that is not directly visible, the system could also determine whether another vehicle is approaching at high speed on a highway ramp, for example, and then adjust its own driving behavior accordingly. In the case of fully automated driving, this would be done by the vehicle itself. But it will probably be a while until this stage is reached.

 

Related articles
 
Magazine Topics
 
Newsletter