National Technical University of Athens (NTUA)
Ανάπτυξη συστήματος αντίληψης οδικού περιβάλλοντος με σύντηξη δεδομένων από πολλαπλές πηγές πληροφορίας
Abstract
dc:descriptionIn recent years the on-going research, for decreasing road accidents, focus in using various sensors onboard vehicles for perceiving the road environment. This effort aims in detecting and recognizing the current situation evolving around the vehicle in order to inform or warn the driver or even take control of the vehicle. The importance of such systems has already started being recognized and nowadays the first generation of such systems is equipping heavy trucks or luxurious vehicles by using one or two different sensors. The functions that are developed are collision avoidance, lane keeping, speed assistance, intersection support and stop & go. Currently, their market penetration is very low, as those systems are under research and development, but it is anticipated their extended use in the near future. The main purpose is the field of view of all sensors to include the whole area of the vehicle. Thus, for example, the driver assistance system (ADAS) should be able to cover a possible collision with another vehicle moving in the adjacent lane during a lane change, or the collision with an obstacle in front of the vehicle during bad weather. Typical sensors used in ADAS systems are radars, cameras, laser scanners, infrared cameras, digital maps, GPS and inertial sensors for estimating the vehicle dynamics. This thesis includes a short introduction in advanced driver assistance systems, the presentation of the architecture of data fusion systems and the proposal of some new algorithms or the extension of previous ones for the processing of data provided by the sensor network installed on the vehicle. The main work focuses at the road geometry estimation using digital maps and the lane estimation using multiple sources of information. It is also focusing in the development of algorithms regarding situation assessment and particularly at estimating the risks involved with the current situation. It is also deals with the problem of detecting the performed maneuver by the driver and the intention associated with the corresponding maneuver. Finally, a software platform for testing all the proposed algorithms is developed and presented.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2008
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Tsogas, Manolis
- Τσόγκας, Εμμανουήλ
Subjects
dc:subject × 16- Οδική ασφάλεια
- Σύντηξη δεδομένων
- Γεωμετρία δρόμου
- Θεωρία αποδείξεων
- Ασαφής λογική
- Αξιολόγηση κατάστασης
- Traffic safety
- Data fusion
- Road geometry
- Evidence theory
- Fuzzy logic
- Situation assessment
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Engineering and Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/24781