DATA-ORIENTED ALGORITHM FOR ROUTE CHOICE SET GENERATION IN A METROPOLITAN AREA WITH MOBILE PHONE GPS DATA
Nowadays, for the estimation of traffic demand or people flow, modelling route choice activity in road networks is an important task and many algorithms have been developed to generate route choice sets. However, developing an algorithm based on a small amount of data that can be applied generally...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2012-07-01
|
| Series: | ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://www.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net/I-2/111/2012/isprsannals-I-2-111-2012.pdf |
| Summary: | Nowadays, for the estimation of traffic demand or people flow, modelling route choice activity in road networks is an important task
and many algorithms have been developed to generate route choice sets. However, developing an algorithm based on a small amount
of data that can be applied generally within a metropolitan area is difficult. This is because the characteristics of road networks vary
widely.
<br><br>
On the other hand, recently, the collection of people movement data has lately become much easier, especially through mobile
phones. Lately, most mobile phones include GPS functionality. Given this background, we propose a data-oriented algorithm to
generate route choice sets using mobile phone GPS data. GPS data contain a number of measurement errors; hence, they must be
adjusted to account for these errors before use in advanced people movement analysis. However, this is time-consuming and
expensive, because an enormous amount of daily data can be obtained. Hence, the objective of this study is to develop an algorithm
that can easily manage GPS data.
<br><br>
Specifically, at first movement data from all GPS data are selected by calculating the speed. Next, the nearest roads in the road
network are selected from the GPS location and count such data for each road. Then An algorithm based on the GSP (Gateway
Shortest Path) algorithm is proposed, which searches the shortest path through a given gateway. In the proposed algorithm, the road
for which the utilization volume calculated by GPS data is large is selected as the gateway. Thus, route choice sets that are based on
trends in real GPS data are generated.
<br><br>
To evaluate the proposed method, GPS data from 0.7 million people a year in Japan and DRM (Digital Road Map) as the road
network are used. DRM is one of the most detailed road networks in Japan. Route choice sets using the proposed algorithm are
generated and the cover rate of the utilization volume of each road under evaluation is calculated. As a result, the proposed route
generation algorithm and GPS data cleaning process work well and a huge variety of routes that have high potential to be used in the
real world can be generated. |
|---|---|
| ISSN: | 2194-9042 2194-9050 |