cat brain.log | less

Getting it down on `paper`

Yellow Light Database

Here’s an idea for an Android/iPhone App, or some other in-dash car after-market add-on.

There is a camera attached to your car facing forward. It has computer vision to detect stop-lights. There is a GPS device on-board to geo-tag these stop-lights. Associated with the stop-light location is its yellow-light duration. Ideally, whenever possible, the app would synchronize with other apps, sharing location and duration information as well. Using this information and GPS data, an app would be able to calculate how long it will take the driver to reach the light. If the light turns yellow, the driver will know whether he/she must stop or may continue through the yellow without running a red light.

An added feature would be a local database of start/stop acceleration data. This data could be used to calculate how must faster or slower the driver would/could travel to still make the light before it turns red.

Finally, a moisture sensor would need to be present, or other traction-like sensor, to determine road conditions and stopping distance. Correlation between historic acceleration data and moisture could be used to statistically compute stopping distance.

Future modifications might include a car detection system indicating how much space is needed between cars to stop in the event of a sudden incident.

Synchronization of traffic speed data is also achievable using this same architecture, thereby providing an alternative to in-ground traffic sensors.

Disclaimer: I know of at least one company using cell phone GPS data to measure traffic speed.

Edit: Waze – Real-time maps and traffic information based on the wisdom of the crowd
Edit: Same idea, different wording, published later



No comments so far.

(comments are closed)