Transforming In-Dash Navigation

Tuesday, June 17th, 2008

Adapted from school a project write-up from last semester — When you add several ingredients together, the result will either become nothing more than a hodgepodge of dissimilar ingredients, or something new and equal to much more than the sum of its parts. The latter case is a transformation. It’s the difference between the tacked-on motion-sensor in the PlayStation 3 controller and the motion-sensitive functions of the Nintendo Wii. Or the difference between sites developed from the ground up to foster social networking and sites which added this functionality as another bullet point in their list of features. It might be difficult to tell when a media transformation has occurred, but it’s pretty easy to tell when one has not.

Several months ago, the project team I became a part of set out to create something out of little more than a marketing phrase, a few ideas stemming from it, and a combination of several media. the “product” became called NavShield. We set out to take the head up display (HUD) technology already available in some vehicles—the Corvette has had this feature for nearly a decade—combine it with several current and upcoming vehicle technologies, and refine it into something new. We started out by thinking about how “cool” it would be to project pretty Apple-esque icons onto the HUD on your windshield. I came up with ideas by driving and having “if only I had this feature” moments. The “thinking process” of the system would be something like this:

• Navigation information is pulled from a GPS receiver (e.g. where you are and where you’re going)
• This information is augmented with location-relevant information pulled from the internet via Wi-Fi or Wi-Max (e.g. gas stations nearby, weather in the area)
• The location-based information is refined based on your preferences (e.g. only gas stations nearby that sell diesel fuel, only whether or not it is expected to rain in your destination at the time you are scheduled to arrive based on your current average speed)
• Finally, the augmented, filtered information gets placed on your HUD

The group’s first tendency was to come up with as many ideas for icons as possible, and clutter the windshield with pretty icons. Just as your first tendency upon first using Mac OS X’s Dashboard or Yahoo! Widgets or Windows Vista’s Sidebar would be to search for and add any widget that perks your fancy until your desktop becomes a mess. While no harm can be done by having too many of these widgets on your computer, having too many on your windshield would be a disaster. That’s why I came up with this process above to connect and filter the data that comes in, and only display the end result.

After showing our project in its current form at the Showcase of Undergraduate Research Excellence event at UCF, the single most frequent piece of feedback we received was the following question: “What about driver distraction?” I recall that driver distraction was an issue when BMW’s iDrive debuted because many core functions’ hardware buttons were replaced with computer-like menus and sub-menus displayed on-screen. The trend is toward displaying more information on the navigation screen, and I felt that were it backed with psychological research, the NavShield project could solve this issue.

While conducting psychology experiments pertaining to driver distraction, and then usability tests on the interface are well outside scope of this one-semester project, it’s definitely the next step. In the meantime, I decided on some measures to limit driver distraction:

• Limiting HUD to upper-third: I decided on a rule that, should something like get to working prototype stage, the HUD could only be displayed on the upper-third of the windshield. What I thought was, if the law prohibits tint below that part of the windshield because of visibility issues, then we should prohibit HUDs below that part of the windshield for the same reason.
• Prioritizing information: Information would be grouped into several types, and prioritized. For example, when at speed, only the most important information (e.g. current speed) would be displayed, but while stopped, second-tier information could also appear. Beyond that, context- or location-relevant information would appear only when needed. For example, an arrow telling you which turn to make would only appear as the turn approaches.

The dashboard screens available in many of today’s automobiles cram as much information as viable, and it seems apparent that this information is added mostly to one-up the competition in terms of feature sets. It’s not uncommon for systems which previously gathered and displayed only navigation information now connect with and display everything from your media player’s music list to your phone’s contact list. At present, the only product on the market putting this information together in a package that feels transformational and not simply tacked-on is Microsoft’s Sync. The Dash navigation system also works similarly to what I have outlined for NavShield; for example, it combines GPS information with traffic data pulled from over the internet. Therefore, I would use these two products are the benchmarks for NavShield were the project taken further.

While I’m not so deluded to say that my project team’s semester project has already reached the level of becoming a piece of transformational media, I do feel that it’s on the right track. The idea of grabbing a lot of information from many sources, intelligently putting them together, filtering them based on user preferences, and displaying only the most relevant information is key, and I feel it means the difference between transforming disparate media into a cohesive whole versus a bullet list of features. Furthermore, even if we were to ignore the application, the idea of collecting, connecting, filtering, and displaying information has applications in any field. It’s something key to the attention data and data portability movements and something that will change the way we behave as much as social networking has.