Audi’s A8 will use Intel’s chips to power level 3 autonomous driving systems
Through the use of processors from Intel’s Programmable Solutions Group and an operating system from Intel’s Wind River subsidiary, Audi will be able to power a system that allows for level 3 autonomous driving, whereby drivers can functions critical to car safety to the autonomous system.
Level 3 autonomy requires drivers to make the switch to and from driverless systems, which still requires them to be behind the wheel, unlike the more advanced level 4 autonomy, but level 3 still requires a solid dose of compute performance.
Autonomous driving requires more than just a driverless system; it needs the hardware and software to enable the training of autonomous systems through the processing of huge amounts of data ,often requiring the power of the cloud and data centre grade servers.
Once a system is trained it needs to be put into action using a technique called inference, whereby a smart system not only puts into action what it has learnt but also absorbs new information and effectively learns through experience,
Intel’s general manager of its Automated Driving Division. Kathy Winter, noted that Intel is in the position to provide processors that cover the compute power needed for both large scale autonomous system training in data centres through to powering driverless systems in cars.
Intel is alone in being able to reach this computational value and performance scale across the car, the network and the data centre. Intel recognised this multifaceted challenge needs a multifaceted toolkit – with an optimised mix of scalable CPUs that can extend to data centre server-class products, FPGAs and, yes, artificial intelligence,” she said.
“And Intel is the only tech company that offers a complete, end-to-end toolkit that spans the full technology challenge car-to-cloud to make fully autonomous driving possible.”
However, Intel is not alone in providing technology to facilitate autonomous driving, with graphics giant Nvidia providing hardware and software to allow for the training and running of driverless cars, with an eye for pursuing levle 4 autonomy.
What do you know about tech in transport? Take our quiz!