National Technical University of Athens (NTUA)
Ιχνηλασία πολλαπλών αισθητήρων σε αυτόνομα και συνεργατικά συστήματα αντίληψης οδικού περιβάλλοντος
Abstract
dc:descriptionTraffic accidents are the 11th most common cause of death, while human factor remains the main responsible. In the beginning of 90’s information, sensor and computer technologies were first applied in the transport domain, creating the sector of the Intelligent Transportation System. In the end of the decade systems like Adaptive Cruise Control (ACC) were first introduced in commercial cars. The first chapter of the current thesis contains a presentation of the technologies that are used in Advanced Driver Assistance Systems. Additionally, the basic theory of data fusion, estimation, Kalman filtering and tracking is explained. Finally, there is a reference to techniques regarding localization and use of GPS devices. In the next chapter, the theoretical analysis and implementation of a road environment perception system with application to automated vehicle is presented Topics that are discussed include sensor and global level tracking. Moreover, ego vehicle state estimation and road geometry estimation using multiple sensors are examined and implemented The third chapter is dedicated to a Measurement Oriented Multiple Hypotheses Tracker applied in a sensor level tracking. A theoretical analysis and the practical implementation into a laser scanner tracking system are presented. Additionally, optimization techniques like clustering and hypotheses merging are investigated and applied to improve algorithm performance. Fourth chapter analyzes the use of tracking system that receives also data from a vehicular network. In this case the Track Oriented Multiple Hypotheses Tracking (TOMHT) algorithm is used, while the issues that stem from using in tracking information from remote nodes are examined. The first sections of the chapter are devoted to the challenge of centralized tracking using remote sensor measurements (radar and vehicular network). The application of TOMHT for data association of radar and vehicular network information is presented in the remaining sections
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2011
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Thomaidis, Georgios
- Θωμαΐδης, Γεώργιος
Subjects
dc:subject × 20- Ιχνηλασία πολλαπλών στόχων - πολλαπλών αισθητήρων
- Σύντηξη δεδομένων
- Συσχέτιση δεδομένων
- Ιχνηλασία πολλαπλών υποθέσεων
- Συνεργατικά συστήματα
- Αυτο - οργανούμενα δίκτυα αυτοκινήτων
- Ασύρματη επικοινωνία
- Φιλτράρισμα δεδομένων
- Multi sensor - multi target tracking
- Data fusion
- Data association
- Multiple hypothesis tracking
- Cooperative systems
- Ad hoc vehicular networks
- Wireless communication
- Filtering
- 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/26834