We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

Showing results for 
Search instead for 
Did you mean: 

Now, even students can build self-driving cars with off-the-shelf, FPGA-based hardware from National Instruments

Xilinx Employee
Xilinx Employee
0 0 49.3K

It’s still not a walk (or drive) in the park, but it’s easier than ever to build a self-driving car. Engineering students at KAIST (Korea Advanced Institute of Science and Technology) needed only two years to develop their first self-driving car, the EureCar (as in Eureka!). EureCar uses a high-precision positioning system, seven laser scanners, and four video cameras to drive itself along a pre-planned path while avoiding obstacles and obeying various traffic laws. EureCar Turbo, the follow-on project based on a bright yellow Hyundai Veloster, took six months to develop and only two months were needed to develop new software for the vehicle.



EureCar Turbo.jpg



The control systems in the EureCar and EureCar Turbo are based on National Instruments’ (NI) CompactRIO-9024 Real-Time Controller and the CompactRIO -9114 8-slot reconfigurable chassis, which contains a Xilinx Virtex-5 FPGA. The design also uses several NI CompactRIO modules plugged into the 8-slot chassis to communicate with the car’s sensors and actuators, which include the LIDAR and video cameras, a GPS unit, and an inertial navigation package. The sophisticated closed-loop control system is based on fuzzy logic and PID loops. It’s programmed with NI’s LabVIEW and the LabVIEW FPGA module. Here’s a block diagram of the control system:



EureCar Control System Block Diagram.jpg




Here’s a video of the driverless EureCar Turbo motoring around the KAIST campus earlier this year:







Notice at the end of the video how a pedestrian tries to play chicken with the car. At first, the car tries to avoid the pesky pedestrian. When the pedestrian gets even more aggressive, the car stops to avoid an accident. It almost makes you believe in Asimov’s entirely fictional three laws of robotics.



Note: This student project was one of three highlighted at the NI Engineering Impact Awards dinner during the recent NI Week held in Austin, Texas.