Abstract
Short-term travel time information plays a critical role in Advanced Traffic Information
Systems (ATIS) and Advanced Traffic Management Systems (ATMS). In this context, the need
for accurate and reliable travel time information sources is becoming increasingly important.
Bluetooth Technology (BT) has been used as a relatively new cost-effective source of travel
time estimation. However, due to low sampling rate of BT compared to other sensor
technologies, existence of outliers may significantly affect the accuracy and reliability of the
travel time estimates obtained using BT. In this study, the concept of outliers and
corresponding impacts on travel time accuracy are discussed. Four different estimators named
Min-BT, Max-BT, Med-BT and Avg-BT with different outlier detection logic are presented in
this paper. These methods are used to estimate travel times using a BT derived dataset. In order
to quantify the accuracy and robustness of these estimators against outliers, a comparative study
between BT and Floating Car Data (FCD) is conducted. Results show that the Min-BT and
Med-BT are more robust concerning the existence of outliers in the dataset and can provide
more accurate travel time estimates compare to Max-BT and Avg-BT.
Systems (ATIS) and Advanced Traffic Management Systems (ATMS). In this context, the need
for accurate and reliable travel time information sources is becoming increasingly important.
Bluetooth Technology (BT) has been used as a relatively new cost-effective source of travel
time estimation. However, due to low sampling rate of BT compared to other sensor
technologies, existence of outliers may significantly affect the accuracy and reliability of the
travel time estimates obtained using BT. In this study, the concept of outliers and
corresponding impacts on travel time accuracy are discussed. Four different estimators named
Min-BT, Max-BT, Med-BT and Avg-BT with different outlier detection logic are presented in
this paper. These methods are used to estimate travel times using a BT derived dataset. In order
to quantify the accuracy and robustness of these estimators against outliers, a comparative study
between BT and Floating Car Data (FCD) is conducted. Results show that the Min-BT and
Med-BT are more robust concerning the existence of outliers in the dataset and can provide
more accurate travel time estimates compare to Max-BT and Avg-BT.
| Originalsprog | Engelsk |
|---|---|
| Tidsskrift | ITS World Congress |
| Antal sider | 12 |
| ISSN | x000-0181 |
| Status | Udgivet - 2012 |
| Begivenhed | ITS World Congress: Smarter on the way - Messe Wien Congress Center, Wien, Østrig Varighed: 22 okt. 2012 → 26 okt. 2012 |
Konference
| Konference | ITS World Congress |
|---|---|
| Lokation | Messe Wien Congress Center |
| Land/Område | Østrig |
| By | Wien |
| Periode | 22/10/2012 → 26/10/2012 |