Ambient Navigation Displays
Using peripheral vision for seamless navigation aid

Navigation systems are useful for helping us find our way to our destination, locating a point of interest on the way, or simply knowing our location on the map. The user inputs the destination address and the device draws him a map, often times with alternative routes to take according to the driver’s preferences: scenic views, faster routes, or toll free roads. Once the route is selected, the system displays a map and turn-by-turn directions on the screen to guide the driver. In a way, a navigation system is a passenger in the car that reads the map for you and tells you when and where to turn.
The navigation system’s turn-by-turn directions can be text-based messages on a screen, audio messages coming through the speakers in the car, or a head-up display showing a simplified version of the map. A problem with text-based onscreen directions is that they require the driver to take his eyes off the road and read the screen to see the next turn, and how close he is to it. Audio messages solve the problem of the driver repeatedly shifting his focus inside and outside the car by reading the directions for the driver. However, these audio messages interfere with the audio sphere of the car and are not very pleasant for people engaged in a conversation inside the car or while listening to music. Head up displays are directed at the driver exclusively, and help him keep his eyes on the road. My aim is to design an interface for navigation system that does not require cognitive effort from the driver and is as informative as possible.

v.1. Haptic feedback through the steering wheel

The initial idea I had was signaling turns through vibrations on the steering wheel. My assumption was that this model didn’t interfere with the driving activity and was only directed to the driver.
The navigation system sends the turn-by-turn directions to the steering wheel. The steering wheel has a vibrating motor on the right and the left side, and indicates the direction to turn by buzzing one of the motors before coming to a junction. The buzzing starts 15 seconds before the junction and the strength and duration of each
buzz increases as the junction nears. Once past the junction, the buzzing fades, regardless of the road taken.
Problems
a. Information that can be represented by pulses is limited: it is not possible to tell the user how far he is away from the next turning point.
b. Different drivers have different styles of holding the steering wheel. Some drivers use both hands to drive, some use only one hand.
c. There is a constant vibration while driving that comes from the road surface, so the vibration on the wheel has to be strong enough to be “visible.” This strong tactile experience might not be a pleasant experience for the driver.

v.2.Visual display on the steering wheel

As a second idea, I started looking at visual displays. My idea was to make use of the driver’s peripheral vision to signal turns and show progress. Peripheral vision is valuable for the display of navigation directions, since it doesn’t require the driver to shift his central vision from road to the interface and can be processed subconsciously, therefore requiring less cognitive effort from the driver.
The top of the steering wheel is the most visible part of the car interior that is in the driver’s field of peripheral vision, so I decided to place the displays on the wheel. In this model, the display is a progress bar placed on the both sides of the steering wheel, left and right. The navigation system sends turn-by-turn directions to the steering wheel, filling a progress bar with light from center to the left or right depending on the driver’s speed and distance from the turning point. The progress bar model placed on two sides of the wheel shows time and direction simultaneously, and is also perceivable without too much effort.
Problems
a. Because the displays are on the wheel and the wheel is not in the same position all the time, information may be
unusable in turns that come one after the other.
b. This interface lacks enough information in some specific turns, for example when there are 2 roads going in the same direction, it’s not possible to differentiate which one is the right direction to take.

v.3.Final Solution: Progress bars with HUD

The progress bar I designed in the second interface was a successful way of indicating time and direction, however the positioning of it was still questionable.
The movement of the display was not a very desirable or usable solution, so I looked for other ways of making an ambient interface.
For the interface to show directions, I needed to deal with look and feel, placement, signaling, and visibility of the elements that construct the interface. These elements are indicators of direction and progress bars to indicate remaining time.
Since my aim was not to recreate a map interface but to create a seamless interface
for navigation help, I decided to carry on with the head up display model. The advantage of the head up display is direct annotation on the road, which eliminates the problem of refocusing between the road and the interior of the car, and helps the driver keep his focus on the road. Current models of head-up displays are maps drawn in 2D format, from a bird’s eye view. In my design I did not want the interface to be a replication of a map, but simply an indicator that the driver can follow while driving. The display is a live image of a 3D arrow that maps itself along
the road and is located on the bottom-left side of the middle of the windshield.
While driving straight, the arrow is invisible. However, it becomes more visible while approaching a turn, and actually moves in the direction of the route to take the driver through that turn. Afterwards, the arrow goes back to the invisible mode again.