Standard safety and control electronic control units (ECUs) – braking systems, power windows – have been standardized in AUTOSAR for almost 20 years. The software in these ECUs is comparatively simple; the largest have a few megabytes of code. On the other hand, we have multifaceted systems for autonomous driving (AD) systems. This software can be incredibly complex, requiring gigabytes of software. AUTOSAR is trying to become a standard for these new, complex systems with AUTOSAR Adaptive, and it may be part of the solution, but requirements and technology are moving at a rapid pace.

In addition, to unlock and deliver the full potential value and functionality of application software in the vehicle it becomes something that needs to be constantly updated. Think about it, right now, when you buy a vehicle, it is as good as it will ever be when you drive it off the dealer lot. But that’s not true for cell phones or computers because they are constantly updated. When you walk out of the store with a new cell phone, it’s only going to get better as the software is improved. As the automotive industry moves into a world where the car is continually updated after delivery, the value will be in the software.

The challenge

The challenge is to have compatibility between systems, taking one piece of functionality across multiple vehicles or brands. But for it to truly work from a cost/value scale, it needs to be more than just compatibility across one OEM’s brands, we need broad compatibility across vehicles. It may sound impossible, but it can be done. In fact, it has been done.

Let’s look at cell phones again. There are two operating systems (iOS and Android), so an application can be built to work on all versions of an iPhone or Android phone. You don’t need individual apps for each phone brand or version. If that was the case, it would create an unbalanced cost/value scale because each manufacturer would be investing resources to create their operating system and there would be no market to make a return on investment for the creation applications – just like the cell phone market and its lack of a vibrant application marketplace before iOS and Android began to dominate. If the auto industry doesn’t address this standardized operating system challenge, the cost of development will be too high to drive strong consumer adaption. So, the more compatibility of software throughout the industry, the less expensive it will be for the end consumer, resulting in higher demand and adoption – creating a virtuous cycle as we have seen in the PC and cell phone markets when value shifted to software.

Where do we go from here?

So, what does the automotive industry do? First off, all the industry players – OEMs and suppliers – have to accept and participate into this standardized approach.  Hopefully once they see the business case for certain AD features doesn’t make sense until they reach that value scale point, they will start to adapt. This will probably mean more collaboration throughout the industry because it’s not sustainable or cost-effective – for the OEMs or consumers – for each automaker to build their own operating system. In essence, like the cell phone industry, we are going to have to streamline to just a few operating systems. But the big question is, who makes them? It’s a fair question and ultimately, we think market influence will make the decision. Let’s go back to cell phones again. Remember Palm? They had a good phone, but the functionality wasn’t – over time you simply couldn’t get the applications you could on other phones. So, despite being a good phone, the value shifted to software applications and since it couldn’t offer that functionality, people didn’t want to buy it.

OK, so we need a standardized operating system that provides an environment for a variety of vehicle functions to safely operate. What are some of the must-haves for this operating system?

• An environment that is easy for developers to create applications.
• The ability to support and compartmentalize applications running safety features (e.g., automated valet parking) as well as those that are simpler and less critical (e.g., remote controlled air conditioning).
• A root of trust that the system and applications are safe, secure and won’t interfere with one another.
• Provide connectivity to the cloud and manage data coming from the cloud to applications.

The ETAS piece of the puzzle

Though we’ve been more involved in AUTOSAR Classic for years – our RTA solutions are in more than 2 billion ECUs today – we also are heavily involved in solutions for AUTOSAR Adaptive. In fact, we’re working beyond the standards to support OEMs by providing more pieces to their operating system initiatives. We’re working to take the functionality of AUTOSAR Adaptive a step higher to develop functions around autonomous driving to make it easier and faster to build these features. Because having AD features in the operating system will make it easier to add AD applications along the line.

To discuss your particular use case or to learn more about how we can help you with your AUTOSAR Classic or Adaptive needs, contact us.