Welcome to the National Transport Library Catalogue

Normal view MARC view

Human vehicle interaction : Drivers' body and visual behaviour and tools and process for analysis Hanson, Lars

By: Series: Publication details: Lund Lund University, 2004; Department of Design Sciences, ; Ergonomics and Aerosol Technology, ; Publication 9, Description: 131 sISBN:
  • 9162861530
Subject(s): Bibl.nr: VTI 2004.0607Location: Dissertation note: Diss. Lund : Lund University, 2004 Abstract: Vehicle design is characterized by a high number of compromises. The special role of human factors in the process is to ensure that the compromises do not occur at the expense of the driver and passengers. Within the automotive industry there is a need for the development of inexpensive and efficient ergonomic tools, facilitating a fast and cost effective car development process as well as discussions between different professionals. The general research objectives were to develop and evaluate: 1) tools for the evaluation and visualization of human vehicle interaction and 2) a process for the use of human simulation tools. To meet these objectives, the aim was to investigate and describe drivers' behaviour and attitudes when interacting with the vehicle cockpit. Driver behaviour, preferred posture and visual time off road in new interior designs were tested, examples of which include flexible pedals and buttons placed close to the gear stick. The resulting preferred minimum and maximum posture intervals correspond well with other experimental data and are larger than intervals for comfort, strength and alertness based on theoretical reasoning. The flexible pedals may have affected lower body angle distributions, which disagrees with other studies. No gender differences in preferred posture were found. Furthermore, it was demonstrated that button location had a significant effect on visual time off road: those buttons close to the steering wheel and close to the road scene were perceived as safe to handle and required the lowest visual time off road. One surprising exception was for the button close to the gear stick. Although placed furthest away from the road scene, a comparatively short visual time off road was required compared to the other buttons placed on the centre stack. There are three possible explanations for this: 1) low discrimination level, 2) perception of risk, and 3) motor skill. Fuzzy logic and neural networks were feasible methods for modelling human behaviour and human perception. Four models were developed for illustration: one for prediction of upper body movement when pushing a button, one for prediction of visual time off road, and two for prediction of driver perception in terms of comfort and safety. The models developed can be integrated in human simulation tools. A combination of results from the methods for modelling driver behaviour and the methods for modelling driver perception along with the analyzer's own judgment of an animation may lead to well-founded decisions as well as comfortable and safe vehicle design. A tool needs a working process. Such a process for usage of human simulation tools was developed and evaluated. The process was presented on a web site and linked to a database. The subjects taking part in the evaluation appreciated the formalization of the human simulation process as it decreased confusion by providing a structured guide for how to work. The most relevant feature of the web based system was the database that facilitates searches for, and the reuse of, results or methods from previous studies that human simulation users within the company had performed.
Item type: Dissertation
Holdings
Current library Call number Status Date due Barcode
Statens väg- och transportforskningsinstitut Available

Diss. Lund : Lund University, 2004

Vehicle design is characterized by a high number of compromises. The special role of human factors in the process is to ensure that the compromises do not occur at the expense of the driver and passengers. Within the automotive industry there is a need for the development of inexpensive and efficient ergonomic tools, facilitating a fast and cost effective car development process as well as discussions between different professionals. The general research objectives were to develop and evaluate: 1) tools for the evaluation and visualization of human vehicle interaction and 2) a process for the use of human simulation tools. To meet these objectives, the aim was to investigate and describe drivers' behaviour and attitudes when interacting with the vehicle cockpit. Driver behaviour, preferred posture and visual time off road in new interior designs were tested, examples of which include flexible pedals and buttons placed close to the gear stick. The resulting preferred minimum and maximum posture intervals correspond well with other experimental data and are larger than intervals for comfort, strength and alertness based on theoretical reasoning. The flexible pedals may have affected lower body angle distributions, which disagrees with other studies. No gender differences in preferred posture were found. Furthermore, it was demonstrated that button location had a significant effect on visual time off road: those buttons close to the steering wheel and close to the road scene were perceived as safe to handle and required the lowest visual time off road. One surprising exception was for the button close to the gear stick. Although placed furthest away from the road scene, a comparatively short visual time off road was required compared to the other buttons placed on the centre stack. There are three possible explanations for this: 1) low discrimination level, 2) perception of risk, and 3) motor skill. Fuzzy logic and neural networks were feasible methods for modelling human behaviour and human perception. Four models were developed for illustration: one for prediction of upper body movement when pushing a button, one for prediction of visual time off road, and two for prediction of driver perception in terms of comfort and safety. The models developed can be integrated in human simulation tools. A combination of results from the methods for modelling driver behaviour and the methods for modelling driver perception along with the analyzer's own judgment of an animation may lead to well-founded decisions as well as comfortable and safe vehicle design. A tool needs a working process. Such a process for usage of human simulation tools was developed and evaluated. The process was presented on a web site and linked to a database. The subjects taking part in the evaluation appreciated the formalization of the human simulation process as it decreased confusion by providing a structured guide for how to work. The most relevant feature of the web based system was the database that facilitates searches for, and the reuse of, results or methods from previous studies that human simulation users within the company had performed.

Powered by Koha