It’s easy to imagine what people will do when their cars take care of all the driving: check email, help a child finish their homework, spread cream cheese on a bagel, all without causing an accident. But before that dream becomes a reality, technologists and automakers will need to collaborate on key innovation challenges, from compute to security. With the development of next-generation advanced driver assistance systems (ADAS), we are entering an era in which cars will become much safer and more efficient as they grow increasingly aware of and react to the surrounding driving environment and conditions. So while you shouldn’t take your hands off the wheel just yet, you can certainly see what the future holds.
Intel is driving the conversation around the growing need for compute in next-gen cars, with security and safety supported by sophisticated systems that sense, anticipate, and react to changing conditions. In the recently released paper, Technology and Computing Requirements for Self-Driving Cars, we explore the complexities and excitement surrounding ADAS, including a section devoted to the top requirements for self-driving cars to become a reality. Here’s a quick look; please check out the full paper for many more details and insights.
Enabling ADAS and Self-Driving Cars: Top Requirements
» Compute Power
It will take a lot of compute muscle to power next-generation ADAS technology. About 1GB of data per second will need to be processed in the vehicle’s real-time operating system, and that data will need to be analyzed fast enough to allow the car to react to changes in less than a second. The car must compute to compete.
In order to facilitate usage of different sensors in an ADAS or autonomous driving scenario, we need an open framework that allows the sensor vendors to provide information that is consumable by any driving controller. An open framework, such as AUTOSAR, is the ideal solution because it allows auto manufacturers to choose the optimal sensor for a given vehicle using cost vs. performance as a metric. Furthermore, having a standard operating environment allows auto manufacturers to develop autonomous driving IP that is portable across an entire suite of vehicles and customizable based on the sensors and capabilities of a given vehicle model.
» Reliable Suppliers
As the “brain” of the car changes, automakers will be looking to supply chain partners who are able to source, build, deliver, and service these new vehicles. Semiconductor companies will play a key role in bringing ADAS systems to life. Companies that collaborate with them will be well-positioned to help lead the change.
» A Centralized Approach
As the industry moves toward more advanced driving experiences, systems will become more complex. It will be increasingly difficult to support the growing number of technologies if each one is a separate embedded control unit (ECU) with its own computer and software. The best strategy will be a more centralized model that integrates functions and uses one source of computing power to carry them out.
» Efficient Solutions
Increased processing power must be matched with increased efficiency—and a small footprint. In fact, automakers have already determined that computing resources need to be located in the safest place in the car: under the driver’s seat. To do this while maintaining driver comfort is challenging. The answer is semiconductors, which deliver lots of processing capability and use very little power.
» Secure Data
Security and privacy are on everyone’s minds these days—at home, at work, and in their cars. While vehicles with next-generation ADAS (and ultimately, self-driving capabilities) will largely rely on data that is generated from sources within the vehicle, IVI systems regularly get data from the web, CDs, and other sources. That’s why it’s critical to safeguard IVI systems from malicious intent. As the IVI also begins to function as the car’s user interface, this becomes even more important, and when vehicles begin relying on data to make quick, accurate decisions, it is critical. Car computing systems need to be able to identify and isolate threats. Tech partners with security track records will be essential.
Want to learn the rest of the story? Read Technology and Computing Requirements for Self-Driving Cars.
About the Automotive Blog Series:
From in-vehicle infotainment to autonomous driving, Intel is using its proven expertise and R&D in computing technology, automotive systems, and consumer electronics to help automotive industry partners accelerate the evolution of connected, intelligent vehicles. This series is designed to offer the insights and observations of the Intel experts and engineers working to advance the next generation of driving experiences.