This work is a joint research effort between USC/ISI and USC School for Policy, Planning and Development.
The purpose of this research is to demonstrate the feasibility of an accurate, low cost and rapidly deployable vehicle traffic classification system, a Network Of Traffic Sensors (NOTS). This system will consist of a number of small, low cost computer nodes, each with one or a few sensors (such as pneumatic tubes or adhesive magnetic sensors), connected with a wireless network. Although individual sensors may be relatively inaccurate, we expect this research to develop algorithms that allow the combination of individual sensor readings in the sensor network to provide highly accurate vehicle classification. Our work is complementary with research into new individual sensors, rather than build a new individual sensor, we will show how existing sensors can work together to improve accuracy. Similarly, while we will use current research in radios and energy-conserving sensor networking, the key contribution of our work is developing new algorithms to apply sensor networks to vehicle classification.
The convergence of inexpensive small computers, sensors and wireless networking creates the possibility for sensor networks: collections of many small computers equipped with sensors and radios, providing collaborative and distributed ability to sense, process, and communicate. Sensor networks have several crucial advantages:
This work will study NOTS systems through simulation and simple experiments.
This work issupported by USC/CSULB METRANS 2003-04 grant #65A0047, SURE-SE, and by USC/CSULB METRANS 2005 grant #tbd, SURE-FT.
For a more complete list of related publications, see the I-LENSE publications page.
Data collected as part of the SURE-SE and SURE-FT projects is made freely available on request to interested researchers. Please contact John Heidemann for requests.
The primary dataset is a collection of 1500 detections taken at three sites at USC on August 6, 2004. Sensor data was supplemented with human observers and videotape to provide ground truth data. We selected three locations on internal campus streets to get a mix of low- and moderate-speed traffic. Vehicles include the USC shuttle bus, construction traffic, including cement mixers and 18-wheel trucks, and general automobile traffic. In addition to general traffic, we selected two passenger cars and ran them over each sensor 10 or more times to provide a baseline known vehicle to evaluate re-identification and sensor consistency.
Details about this dataset are available in the SURE-SE Final Report.