Project Description:: Realtime path planning and guidance system using adhoc networks.
Existing mapping technology: The traffic signs form a path from point A to the Holland Tunnel entrance at
point D . |
Providing next-step directions to
users attempting to find a path is a well-known technique. Often, the path
planning is static, occurring before the user starts off to their
destination. Unfortunately, the predefined route cannot take into
account delays like traffic accidents, spontaneous road hazards, and
congestion. Even worse, such systems are no good at planning alternate
routes in real-time. This project will direct drivers through the city to a specific destination using real-time status information in an adhoc network. The real-time status information will allow more flexibility and adaptivity to changing traffic conditions. Sensors will be embedded into signs located throughout the city. Each sign guides the driver to the next sign, which in turn provides the next turn or guidance to the next step until the driver reaches his final destination. |
|
Clockwise from upper left: Traffic sign directs
westbound drivers to turn left at the corner of Lafayette St. and Kenmare
St.
The next sign directs drivers to make a right onto Broome Street. Third, drivers are directed to bear left at the Broome St./Watts St. split. Finally, at Watts St. and Varick St., drivers see the entrance to the Holland Tunnel. Idea: The electronic sign below can be changed, depending on conditions. If there's a traffic jam and I need to get to the airport, the system should be able to direct me to a shortcut. |
|
| Visualization: Visual cues such as lights are also used to provide incremental directions about paths, such as in the case of runway lights. Each pair of lights on either side of the runway indicates the next step in the path for the landing plane. |
 |
| With the exception of
the electronic sign, all of these cues are static, and their cues are all
placed and directed by central coordinators, such as traffic engineers.
This project explores dynamic real-time traffic coordination. In an emergency, what if signal lights could direct people inside a building to the exits, even when available exits change and even when congestion through different passageways blocks certain hallways. The solution here is intelligent networks of sensors to create dynamic, adaptive pathfinding directions. |
 |
| Hardware: Proposal to use the Pushpin sensor platform.
|
 Pushpin sensor
|
|
Bibliography Christakos, Constantine K., Using Interactive Sensor Networks to Analyze Distributed Phenomena, MIT Department of Media Arts & Sciences, Ph.D. Thesis Proposal, June 2004 Li, Q., De Rose, M., Rus, D., "Distrbiuted Algortihsm for Guiding Navigation Across a Sensor Network", Proceedings of the 9th annual international conference on Mobile computing and networking, San Diego, CA, USA, pp. 313-325 |
|
| Home | | | Biography | | | MIT Media Lab | | | Useful Sites | | | Reading Journal | | | Interesting Quotes | | | Contact |
Path
planning visualization