Design and implementation issues of a portable navigation system
Date
2009
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Personal navigation is one of the fastest developing fields due to ever changing needs of consumers. Most recently, GPS-based systems have gained popularity in commercial navigation systems. However, a problem arises when the GPS is unable to provide positions due to interruptions in the signals. One way to solve this problem is by using inertial sensors that do not require any outside signals. The thesis deals with the issues related to the design specifications for a completely portable navigation system (PNS). A PNS is defined as a navigation module that is completely transferable from vehicle to vehicle and from one mode of transit to the other and should be able to provide navigation information even for short GPS signal outages. The PNS requires calibration and alignment before navigation can be started. These two topics require different implementation methods to accommodate for the special needs associated with portability. A new method was proposed that can perform calibration of the sensors without any requirement on the PNS shape. Alignment is not required if six sensors are part of the system, however, partial sensor system will require some alignment criteria. For low-cost commercial system, signal sampling requirement is also important as the choice of micro processor, memory card and DAQ card are all cost defining factors. A frequency of 50Hz and timing errors of up to 1 ms did not degrade the navigation results, however, 10 bits resolution produced significant degradation. It was also determined that a heading gyro and a dual axis accelerometer in the horizontal plane with velocity aiding produced comparable results to six sensors unit without any aiding, and hence can be used for low-cost navigation. As a PNS can be used either by a person on-foot or in-vehicle, one of the important issues is to identify the correct mode of transit to apply the most appropriate navigation equations. Consequently, an algorithm was developed and tested to detect mode of transit when the system was attached on the belt of the user. A levelled ground walking scenario produced the best results and the average accuracies of other cases stayed over 90%.
Description
Bibliography: p. 190-198
Many pages are in colour.
Many pages are in colour.
Keywords
Citation
Syed, Z. F. (2009). Design and implementation issues of a portable navigation system (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca. doi:10.11575/PRISM/3019