Aristotle University Of Thessaloniki (AUTH)
Προηγμένο σύστημα εξατομικευμένης πληροφόρησης και προειδοποίησης των μετακινουμένων
Abstract
dc:descriptionSeveral studies on the use of web services and specifically those related to infomobility services, prove their fast penetration to the market and foresee their high increase over the next years. Such services are offered through the mobile phone or PDA, where the travelers receive location-based information. On the other hand, that gradual difusion of Advanced Driver Information Systems (ADAS) and their integration by the vehicle manufacturers aims at the enhancement of traffic safety. However, the benefits of such systems to the traffic safety may be significantly reduced by the inappropriate behaviour of the drivers towards the relevant technologies, e.g. overconfidence to the system that leads to the reduction of attention on the road or behavioural adaptation that compensates the additional safety aspects. Systems with automatic adaptation capabilities, according to the ‘needs and wants’ of each driver, constitute an optimal solution, reducing the driver workload by using ADAS and increasing his/her comfort while driving. The aim of the current PhD dissertation is the optimization of infomobility services for travelers, as welll as drivers warning systems regarding danger stemming both from the longitudinal and lateral road axes, through their personalization. Thus, particular algorithms are proposed, based upon a number of relevant parameters (static, semidynamic and dynamic), leading to a holistic personalization. The drivers’ warning algorithms are based on dynamic parameters that determine their personal driving style, being the reaction time, the distance to line crossing (or alternatively the time to line crossing) and the time to collision or time headway. The infomobility personalisation algorithms for travelers are based on the user’s previous selections and are distinguished as follows: route selection information, points-of-interest (POIs) selection information and pushed information (by the system itself, without prior request by the user). The parameters that constitute the route selection algorithm are the maximum accepted walking distance per transportation means connection, the accepted travel means and the maximum accepted number of transportation mean changes during a multimodal trip. Secondary parameters are also taken into account, which are the shortest or the cheapest route. The parameters used for the POI selection personalization are the POI type (hotel, restaurant, etc.), its sub-type (e.g. restaurant type: Italian, traditional, Chinese, etc.), the day of the week (either weekday or weekend) and of course the user’s past selections. The last parameter is also used for the personalization of the “pushed” service (service proposed to the user without explicitly asking for them). Following, a short description of the dissertation contents is given. Chapter 1 presents the market situation and trends in relation to the location-based infomobility services through the mobile devices, as well as the increasing use of ADAS and the need for their adaptation and personalisaiton for the enhancement of traffic safety. The methodology used throughout the whole document and the interconnection among its chapters are analysed in Chapter 2. The clustering of travelers, along with the analysis of their characteristics, are included in Chapter 3. The 4th Chapter starts by presenting existing clustering of ITS (ADAS, In-Vehicle Information and Communication Systems - IVICS and road users information systems), to conclude to a new, more systematic clustering. Additionally, a description of the functionalities of the two ADAS for which innovative algorithms are developed, is provided (namely Lane Deviation Warning and Frontal Collision Warning systems). Within the same chapter, best practices on Human Machine Interaction (HMI) concepts and elements for providing the information and warning of the above-mentioned ADAS to the driver, are included. The various HMI types are analysed and the need for different HMI by various drivers categories is highlighted, through the presentation of past evaluation results of relevant systems. ..................................
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2007
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Panou, Maria
- Πάνου, Μαρία
Subjects
dc:subject × 14- Μοντελοποίηση συμπεριφοράς οδηγού
- Προσωποποιημένοι αλγόριθμοι ΣΣΥΟ προειδοποίησης οδηγών
- Προσωποποιημένοι αλγόριθμοι πληροφόρησης μετακινουμένων
- Εξατομικευμένο σύστημα προειδοποίησης οδηγών
- Εξατομικευμένο σύστημα πληροφόρησης μετακινουμένων
- Driver behaviour modeling
- Personalised drivers warning algorithms
- Personalised travellers information algorithms
- Personalised drivers warning system
- Personalised travellers information system
- Επιστήμες Μηχανικού και Τεχνολογία
- Επιστήμη Πολιτικού Μηχανικού
- Engineering and Technology
- Civil Engineering
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/20593