Is Auto the Next Android?

| | by

Each day, when I get in my car and drive to work I think about the software business I’m involved in. Yet it seldom occurs to me that there are over 1 million lines of code running in the car that I’m riding in -And it just works! No bugs, log files, no visible defects – it just works.

I’ve been in the software business for over 30 years now and I’ve written and debugged thousands of lines of code, none of which I can honestly say were bug free. But I never wrote software for an automobile or an aircraft, so maybe I just never had an appreciation for what it means to write software for a truly “mission critical” application. And while the software for my PC and Smartphone operate bug-free most of the time, I still get application crashes at least once a week.

At LinuxCon a few weeks ago, I presented a session called “Is Auto the next Android?” The session was intended to explore the up and coming world of software in the auto industry. While all of us have spent hundreds or thousands of hours in automobiles as drivers or passengers, like me, you have likely not spent much time thinking about all the software that has crept into running the mechanics, information and entertainment portions of our cars.

As a basis for my LinuxCon session, I made several comparisons to the Android mobile OS world to draw parallels to a familiar environment.  I’m truly impressed with all that I can do with my smartphone, and recently have begun exploring what it would mean to port many of the apps that I love on my smartphone into my car.

So let’s start with some of the basics. Turns out that over 60% of the software in a car is in what’s called the “headunit.” The headunit is a separate computer that is responsible for operating all of the “infotainment” functions for a car. Infotainment describes a combination of the entertainment, utility and info functions that a car provides. Together, they are described as the “in-vehicle infotainment” center or IVI. Building an IVI system that includes the rich feature set that is shipping with today’s cars is very costly and complex for the auto industry. With all the cool capabilities that are shipping on consumer electronic devices, there is huge pressure on the auto industry to deliver equally cool features and more.

But the auto industry has a big challenge. While consumer electronics (CE), like smartphones, have development cycles  as short as three months, automotive IVI development cycles are commonly three years or more. This leaves the auto manufacturers  chasing the CE industry, constantly struggling to deliver and keep up. The result: frustrated auto customers. Frustrated in that 1) the same feature set in an automobile costs up to 10x more and 2) as soon as the car is purchased, the feature set is out of date. With over 1 million lines of code required to build the IVI unit, how can the auto industry possible respond?

Ok, I’m sure you’re already thinking that the auto industry needs to get their head around the world of open source to be able to compete with the CE industry. And you are right. To reduce the complexity of the software stack required to run today’s IVI systems, there needs to be a big commitment to leveraging huge amounts of open source and creating standards. Each original equipment manufacturer (OEM) wants to differentiate and offer an IVI experience that outpaces its competition, but losing to the CE industry is an even larger threat. Change needs to occur sooner rather than later.

Just as handset manufactures joined together through the Open Handset Alliance to rally around Android and help it compete with the iPhone, a similar alliance called GENIVI has been created in the auto industry. With over 160 companies already participating, OEMs and their supply chains are joining together to define a common IVI stack that they can leverage to move faster through open source and standards. This will allow more time to be spent higher up in the stack, delivering differentiating choices in dramatically shorter time frames. GENIVI was started in 2009 and has been working hard to organize and lay out an architecture that will support the demanding requirements for IVI.

And if you are thinking “gee, why can’t we just use Android and run the same software that we run on smartphones,” there are additional requirements that are very important for the auto industry. Apps for cars need to be aware of the many different states of the vehicle, including running, moving, stopping, reversing and sitting idle for long periods of time. If you already have a GPS built into your car, you know that the later models don’t allow entering a new destination when the car is in drive. Makes sense, right? It takes a fair amount of focus to enter an address, which is a major distraction to the driver. This is only one example, but you can imagine the many other apps and activities that would be highly distractive to the driver. While it’s fun to have lots of cool motion and graphics in phone apps, these can be distracting to a driver, causing accidents and in some cases death.

And then there is power management. Ok, so you’re probably thinking that a car battery is WAY bigger than a cell phone battery. But people take trips and leave the car in the garage for a week or two, so the system has to be aware of long-term idol periods. Additionally, when a vehicle is in reverse, it needs to be able to activate the rear camera. And on top of this, there are safety and regulatory organizations that are monitoring what goes on in your car that need to be considered. There aren’t yet formal policies and guidelines around what exactly should be restricted in an IVI, but research is underway and we can expect formal guidelines to be developed in coming years.

IVI units will quickly become sophisticated middleware platforms that include connection to the cloud, peer to peer connection between multiple vehicles, connections to social networks and more. With all this on the horizon, auto IVI is sure to develop into a powerhouse of opportunity for app developers. App stores specific to the auto industry can’t be far away, so if you missed out early on in the mobile app world, here’s a chance for you to get in on the ground floor and become part of the next big opportunity. I’m sure watching this one closely and I hope you do too. I spend upwards of 20 hours a week in my car, so I’m pretty excited about the possibilities ahead!

Tags: , , , , , , , , , , , , , , , , , ,

No comments yet.

Leave a Reply