The Next Plateau For Autonomous Driving

April 24, 2018


The Next Plateau For Autonomous Driving

Building on a chipset announcement of a year ago, Siemens his reducing the time required to test autonomous driving systems.

About a year ago, I reported on a story out of Mentor Graphics, that the company was releasing a chipset aimed at autonomous driving. The chipset, the DRS360, was based on a Xilinx Zynq FPGA. Fast forward to today, and Mentor Graphics is now a division of Siemens, and that gives them a little more muscle.

That muscle has resulting in what the company is calling “a breakthrough solution for the development of autonomous driving systems.” As part of the Simcenter portfolio, the solution minimizes the need for extensive physical prototyping while reducing the number of logged test miles needed to demonstrate autonomous vehicle safety thanks to some innovative testing methods, including the latest simulation techniques.

Available in the third quarter of this year, Simcenter leverages advanced, physics-based simulation and advanced sensor data processing to significantly reduce the development, verification, and validation of self-driving cars.

A big piece of the Siemens solution hails from TASS International, which is now a Siemens property. The company’s PreScan simulation environment produces realistic, physics-based simulated raw sensor data for an unlimited number of driving scenarios and traffic situations. That data is then fed into Mentor’s DRS360 platform, where it’s fused in real time to create a high-resolution model of the vehicle’s environment and driving conditions. The DRS360 tests and refines the algorithms for things like object recognition.

The platform works in environments that take advantage of LiDAR, radar, and other vision-sensing technologies to develop physics-based, 3D simulated versions of specific sensor modules.