Predicting Globally and Locally: A Comparison of Methods for Vehicle Trajectory Prediction
Abstract
We propose eigen-based and Markov-based meth-ods to explore the global and local structure of pat-terns in real-world GPS taxi trajectories. Our goal is to predict the subsequent path of an in-progress taxi trajectory. The exploration of global and local structure in the data differentiates this work from the state-of-the-art literature in trajectory predic-tion methods, which mostly focus on local struc-tures and feature selection. We propose four algo-rithms: two eigen-based (EigenStrat, LapStrat), a Markov-based algorithm (MCStrat), and a fre-quency based algorithm FreqCount, which we use as a benchmark. A pairwise analysis of algorithm performance reveals the best performer FreqCount on a large real-world data set to be LapStrat, which performs better or the same as the more locally de-pendent (MCStrat). 1
Cite
Text
Groves et al. "Predicting Globally and Locally: A Comparison of Methods for Vehicle Trajectory Prediction." International Joint Conference on Artificial Intelligence, 2013.Markdown
[Groves et al. "Predicting Globally and Locally: A Comparison of Methods for Vehicle Trajectory Prediction." International Joint Conference on Artificial Intelligence, 2013.](https://mlanthology.org/ijcai/2013/groves2013ijcai-predicting/)BibTeX
@inproceedings{groves2013ijcai-predicting,
title = {{Predicting Globally and Locally: A Comparison of Methods for Vehicle Trajectory Prediction}},
author = {Groves, William and Nunes, Ernesto and Gini, Maria L.},
booktitle = {International Joint Conference on Artificial Intelligence},
year = {2013},
pages = {5},
url = {https://mlanthology.org/ijcai/2013/groves2013ijcai-predicting/}
}