Street Smart Traffic

Tuesday, April 04, 2006

In-car navigation and other relevant research

This is an excellent survey of Existing Systems that use In-car navigation.
In-Car Navigation Usage: An End-User Survey on Existing Systems

A group at the University of Nottingham is conducting research to understand the:

• Implications of new technologies in a driving context (e.g. for safety, efficiency, comfort/pleasure)
• Design issues for novel user-interfaces (e.g. augmented reality, speech, gesture-driven input devices)
• Methods for use in design and evaluation (e.g. the use/validity of driving simulators, setting performance standards)

Link to the lab's homepage

Their publications can be found here.

Survey Says

People realy value traffic information in driving aids, but the information is of little value. Check ou the survey results:
http://www.viktoria.se/%7Efresva/documents/survey_results.pdf

I will cut out the most relevant charts and post them directly here later.

Wednesday, March 15, 2006

Early Stage Documentation

I have created a few techinical documents at this point.

My first paper on this subject.
A presentation on early experimental results.

Thesis Abstract

Automobile traffic is a major problem in developed societies. We collectively waste huge amounts of time and resources traveling through traffic congestion. Drivers choose the route that they believe will be the fastest; however traffic congestion can significantly change the length of a trip. Drivers that know where clusters of slow traffic exist can choose other, more efficient routes. We could save significant amount of wasted time if traffic congestion patterns could be effectively discovered and disseminated to the general public. Currently most people use a centralized system that is over 50 years old. This system is fairly effective, but it has significant problems.

We propose a system that uses a standard GPS driving aid, augmented with peer-to-peer wireless communication. This system should provide more complete traffic monitoring than existing systems, and do so at almost no cost to the service provider. The project will be evaluated in a simulation. The system uses a combination of clustering and epidemic communication to find and disseminate traffic information. To work in realistic setting the system must be able to accommodate dynamic traffic patterns. To efficiently use the communication medium the system must impose a hierarchy on the peer-to-peer network. To accurately test the system we will use sophisticated automobile traffic simulators along with realistic communication simulators. This project could become a very useful system, saving millions of human hours and dollars.

Tuesday, March 07, 2006

Won 2nd Place

This idea won second place at a umbc business plan competition. I will get $100.

I was disappointed that I got second place. The business that won is OpenPosting, which is basically Craig's List without the user base. My plan does not is not a current commercial competitor. They did have a 5 page business plan and mine was just 2 pages I threw together one afternoon. There are other academics and commercial research labs that are working on this idea. However none of them have taken the same approach as me, hopefully my approach works best.

I'll beat them out at the real competing at the MoshPit. Silly name, but decent money at stake: $30k+ with $10k+ going to first place.

I have a while till then and I will have solid business plan by then.

Found Previous Work

I found out today that I am not the first one to come up with this idea. I kind of thought this was an original idea. Well, those are few and far between. So the bar was set by TrafficView. I think that I can do better. I will lift thier best ideas and keep going.

I already have a higher page rank on google for "peer to peer traffic automobile".
Got to love a good page rank for ebiquity.