Electronics Suppliers Preach Gospel of Openness for Autonomous Cars

Semiconductor companies want to build an open foundation for the autonomous vehicle’s solution stack. They just need automakers to accept the concept.
May 19, 2017

As the era of the autonomous car approaches, electronics suppliers are encouraging a culture of technological openness in the historically competitive arena of the auto industry.

The shift to open platforms, they say, would enable automakers to bring self-driving cars to the market more quickly, while saving on the extensive research and development costs that are sure to follow. “The more we align on a common platform, and not reinvent the wheel, the more automakers can focus on what really matters to them from a product differentiation standpoint,” Jack Weast, chief systems architect for autonomous driving at Intel Corp. , told Design News.

To make that vision happen, electronics suppliers are teaming with automakers and Tier Ones, as well as rolling out new hardware and software platforms. On May 16, Intel announced a four-way partnership with BMW Group, MobilEye N.V., and Delphi Automotive PLC to “deploy a cooperation model” to deliver solutions. In April, Renesas Electronics America introduced an automated driving platform known as Autonomy, focused on openness and aiming to provide a programmable platform that would allow automakers and Tier Ones the opportunity to “plug in their own IP.”


Intel Corp., BMW Group, MobilEye N.V., and Delphi Automotive PLC are teaming up to “deploy a cooperation model” to deliver solutions for autonomous vehicles. (Source: Intel Corp.)


The key is for the auto industry to let electronics companies develop the base platform of the autonomous car’s solution stack. Intel, for example, wants to provide solutions for such technologies as 5G connectivity, data center infrastructure and cloud computing, all of which would be components of the autonomous car’s architecture. Automakers would then take their unique intellectual property for, say, automatic braking or collision avoidance, and layer it atop that foundation.

“They shouldn’t have to worry about, ‘What radio technology do I use?’ or ‘What frequency bands can be supported?’ or ‘What’s the Internet packet protocol?’” Weast said. “They should start their work at a higher level in the stack.”

The question is whether automakers will buy into that vision. In a recent blog on the Intel website, Doug Davis, general manager of the Automated Driving Group at Intel, bluntly stated that “there are plenty in the automotive industry who don’t understand how open collaboration can enable differentiation and innovation.” Davis argued that “technology best solves problems when it’s organized around common platforms and predictable interfaces.”

Indeed, there is historical precedence to support Davis’ point. When IBM built its first PCs using off-the-shelf parts and outsourced operating systems in the 1980s, the technology grew by a factor of 150X in two decades. Similarly, data centers took off when they adopted PC technology and wireless technology thrived after standardized hardware and software made it easier for users to set up connections.

“We’ve done this in a number of different industries,” Weast said. “It brings costs down, accelerates time to market, and allows the customers to invest in places where they’ll get the most return.”

Experts say the auto industry is open to such ideas, but its long, competitive history may be slowing


For a lot of folks there is a big difference between most consumer junk and the car we buy. A 3 year old car is still "fairly new", while a 3 year old computer is obsolete, although it still works perfectly. And while computer failures can be very inconvenient, unlike cars, the computer is seldom a "life critical" part of our existence. So the extra effort on vehicle control systems is well warranted.

Add new comment

By submitting this form, you accept the Mollom privacy policy.