Technology‎ > ‎

Simulating Radar for Driving Autonomy

postat 7 iul. 2018, 17:46 de Emil Francu
 Though many of us probably first encountered radar in a movie involving warships and submarines — the radar beam represented by a line sweeping clockwise around a circular monitor, with an occasional blip indicating the presence of an enemy vessel — a radar prototype was first developed by the German inventor Christian Hülsmeyer in the early 1900s as a way for ships to avoid collision in a fog.

Today, radar is one of the fundamental technologies being harnessed by companies racing to develop autonomous vehicles of all sorts. Advanced driver assistance systems (ADAS) in automobiles already use this technology to detect vehicles in the driver’s blind spot. Radar systems installed in the vehicle can detect objects that the driver may have missed and send a signal that automatically applies the brakes to avoid a crash. Applications of radar, as well as lidar, ultrasound sensors and visual cameras, will continue to expand as autonomous technology increases in cars, drones, robots and other devices.

autonomous vehicle radar

Engineering simulation plays a major role in millimeter-wave automotive radar development. However, developing and applying the technology present major engineering challenges. Here at ANSYS, we are always working on new ways to enable pervasive simulation solutions to aid engineers in solving these challenges.

One of the critical challenges is the computational size of the problem. Because the physical size of a target can be large and the electromagnetic frequency is very high, significant time and compute resources are needed to solve the resultant, extremely large computation. To solve this issue, ANSYS has introduced significant new capability to model radar signatures of electrically very large targets and scenes with the integration of ANSYS HFSS SBR+ within the ANSYS Electronics Desktop. This integration of HFSS SBR+ to the available high-frequency electromagnetic (EM) solver technologies allows designers to apply the best analysis technologies for predicting radar signatures of structures ranging from sub-wavelength to kilo-wavelengths. Shooting and bouncing rays (SBR) is a ray tracing technique within the physical optics (PO) framework, and is suitable for efficiently solving electromagnetic problems that are hundreds and thousands of wavelengths in size.

In the latest issue of ANSYS Advantage, two articles focus on engineering simulation of automotive radar systems — one looking at the fundamental issues surrounding radar antenna design, and the other on the challenges of producing operating radar systems for the wide variety of vehicle designs in the market today.

In the first, Autonomous Vehicle Radar: Improving Radar Performance with Simulation, Shawn Carpenter, product manager for high frequency electronics at ANSYS, describes the step-by-step process used to develop a 77 GHz automotive radar sensor on a two-sided printed circuit board (PCB) using a slotted waveguide. The focus is on the antenna, which is the interface between the sensor and the real world. ANSYS HFSS and HFSS SBR+ simulation is used to design the antenna array for the desired frequency and determine interaction with the much larger fascia and bumper using its high-frequency ray tracing methods. Simulation is also applied to interaction with external environment of vehicles and pedestrians. Read the article to learn more about how ANSYS simulation can save you radar antenna development time.

Autonomous vehicle radar

The second article, On the Radar, by Clyde Callewaert, principle radio frequency (RF) engineer at Autoliv Electronics describes the challenges faced by a world-leading automotive safety systems company in developing radar systems that will function in the wide variety of automobile designs on the market today. Specifically, radar systems are often located behind the plastic fascia that covers the bumpers to make them look aesthetically pleasing. The positioning of the radar and its mounting bracket is crucial to success. Failure can mean a redesign that might take 12 weeks and cost a million dollars or more.

Autoliv engineers leverage HFSS to generate radiation patterns for the radar mounted in a certain location. From the simulation they can determine the angular and distance range of effective coverage. Read the article to learn more about the details of Autoliv’s process to supply electronics safety systems to automobile manufacturers around the world.

radar in fascia

For even more insight, view the webinar or read the white paper on improving radar performance with simulation.

Today we most often observe radar in action on the daily weather report as it tracks the movement of storm systems, or when a police officer armed with a radar gun pulls us over for speeding (a much rarer occasion, we hope). Radar has now come full circle in the emerging era of autonomous vehicles as a means, once again, of preventing collisions. 

About Mark Ravenstahl

H. Mark Ravenstahl, Technical Director, Strategic Partnerships and Business Development, Electronic Business Unit Mr. Ravenstahl is responsible for strategic partnerships, product strategy and marketing for the ANSYS Electronics Business Unit. Previously, he held various sales and marketing positions at Ansoft Corporation, now ANSYS, Inc. He is a graduate of the University of Pittsburgh.