Details
| Original language | English |
|---|---|
| Pages (from-to) | 139-144 |
| Number of pages | 6 |
| Journal | International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives |
| Volume | 38 |
| Publication status | Published - 2010 |
| Event | ISPRS Technical Commission III Symposium on Photogrammetric Computer Vision and Image Analysis, PCV 2010 - Saint-Mande, France Duration: 1 Sept 2010 → 3 Sept 2010 |
Abstract
Accurate and reliable localization in extensive outdoor environments will be a key ability of future driver assistance systems and autonomously driving vehicles. Relative localization, using sensors and a pre-mapped environment, will play a crucial role for such systems, because standard global navigation satellite system (GNSS) solutions will not be able to provide the required reliability. However, it is obvious that the environment maps will have to be quite detailed, making it a must to produce them fully automatically. In this paper, a relative localization approach is evaluated for an environment of substantial extent. The pre-mapped environment is obtained using a LIDAR mobile mapping van. From the raw data, landmarks are extracted fully automatically and inserted into a landmark map. Then, in a second campaign, a robotic vehicle is used to traverse the same scene. Landmarks are extracted from the sensor data of this vehicle as well. Using associated landmark pairs and an estimation approach, the positions of the robotic vehicle are obtained. The number of matches and the matching errors are analyzed, and it is shown that localization based on landmarks outperforms the vehicle's standard GNSS solution.
Keywords
- Accuracy, Feature Extraction, Localization, Mapping, Mobile Laser Scanning
ASJC Scopus subject areas
- Computer Science(all)
- Information Systems
- Social Sciences(all)
- Geography, Planning and Development
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives, Vol. 38, 2010, p. 139-144.
Research output: Contribution to journal › Conference article › Research › peer review
}
TY - JOUR
T1 - Vehicle localization using landmarks obtained by a LIDAR mobile mapping system
AU - Brenner, Claus
PY - 2010
Y1 - 2010
N2 - Accurate and reliable localization in extensive outdoor environments will be a key ability of future driver assistance systems and autonomously driving vehicles. Relative localization, using sensors and a pre-mapped environment, will play a crucial role for such systems, because standard global navigation satellite system (GNSS) solutions will not be able to provide the required reliability. However, it is obvious that the environment maps will have to be quite detailed, making it a must to produce them fully automatically. In this paper, a relative localization approach is evaluated for an environment of substantial extent. The pre-mapped environment is obtained using a LIDAR mobile mapping van. From the raw data, landmarks are extracted fully automatically and inserted into a landmark map. Then, in a second campaign, a robotic vehicle is used to traverse the same scene. Landmarks are extracted from the sensor data of this vehicle as well. Using associated landmark pairs and an estimation approach, the positions of the robotic vehicle are obtained. The number of matches and the matching errors are analyzed, and it is shown that localization based on landmarks outperforms the vehicle's standard GNSS solution.
AB - Accurate and reliable localization in extensive outdoor environments will be a key ability of future driver assistance systems and autonomously driving vehicles. Relative localization, using sensors and a pre-mapped environment, will play a crucial role for such systems, because standard global navigation satellite system (GNSS) solutions will not be able to provide the required reliability. However, it is obvious that the environment maps will have to be quite detailed, making it a must to produce them fully automatically. In this paper, a relative localization approach is evaluated for an environment of substantial extent. The pre-mapped environment is obtained using a LIDAR mobile mapping van. From the raw data, landmarks are extracted fully automatically and inserted into a landmark map. Then, in a second campaign, a robotic vehicle is used to traverse the same scene. Landmarks are extracted from the sensor data of this vehicle as well. Using associated landmark pairs and an estimation approach, the positions of the robotic vehicle are obtained. The number of matches and the matching errors are analyzed, and it is shown that localization based on landmarks outperforms the vehicle's standard GNSS solution.
KW - Accuracy
KW - Feature Extraction
KW - Localization
KW - Mapping
KW - Mobile Laser Scanning
UR - http://www.scopus.com/inward/record.url?scp=84875202588&partnerID=8YFLogxK
M3 - Conference article
AN - SCOPUS:84875202588
VL - 38
SP - 139
EP - 144
JO - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
JF - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives
SN - 1682-1750
T2 - ISPRS Technical Commission III Symposium on Photogrammetric Computer Vision and Image Analysis, PCV 2010
Y2 - 1 September 2010 through 3 September 2010
ER -