Journal
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
Volume 60, Issue 12, Pages 3747-3755Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TIM.2011.2147670
Keywords
Geographic information system (GIS); global positioning system (GPS); Kalman filter (KF); trajectory prediction
Funding
- United States Department of Transportation through the University of Vermont Transportation Research Center
- National Science Foundation through the Faculty Early Career Development [1054333]
- Div Of Electrical, Commun & Cyber Sys
- Directorate For Engineering [1054333] Funding Source: National Science Foundation
Ask authors/readers for more resources
This paper proposes several extensive methods to predict the future location of an automobile. The goals of this paper are to find a more accurate way to predict the future location of an automobile by 3 s ahead, so that the prediction error can be greatly reduced with the innovative idea of merging global-positioning-system (GPS) data with geographic-information-system (GIS) data. The improvement starts by applying existing techniques to extrapolate the current GPS location. Comprehensive Kalman filters (KFs) are implemented to deal with inaccuracy in the different identified possible states an automobile could be found in, which are identified as constant locations, constant velocity, constant acceleration, and constant jerks. Then, the KFs are set up to be part of a interacting-multiple-model (IMM) system that provides the predicted future location of the automobile. To reduce the prediction error of the IMM setup, this paper imports an iterated geometrical error-detectionmethod based on GIS data. The assumption that the automobile will remain on the road is made; therefore, the predictions of future locations that fall outside are corrected accordingly, making a great reduction to the prediction error. The actual experimental results validate our proposed system by reducing the prediction error to around half of what it would be without the use of GIS data.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available