Hardware implementation of a pedestrian navigator

Abstract

The ability to accurately estimate the position of a moving body has become an essential feature in a wide range of fields, particularly transportation and exploration. In recent years, several fields of navigation have enjoyed success brought about by Global Positioning System (GPS) technology. However, other fields, such as pedestrian navigation, are still proving to be a challenge. This is due to the fact that pedestrian navigation is required to operate in places such as dense urban areas, large buildings and underground shopping centres, where GPS signals are blocked or degraded. In light of this, pedestrian navigation systems typically include inertial motion sensors in order to obtain positional data in the absence of GPS. Inertial navigation systems (INS), however, suffer from drift errors and noise, which often severely restricts the obtainable accuracy of the system. Many pedestrian navigation systems make use of a GPS receiver to periodically correct these errors. That being said, such systems still suffer greatly when used in GPS-degraded environments for long periods of time since the GPS receiver is unable to correct the accumulated errors until GPS data becomes available. Therefore, the scope of this project is to develop a MEMS-based inertial pedestrian navigation system which can provide accurate positional data even when used in GPSdegraded environments for extended periods of time. The development of the system includes both software and hardware designed specifically for this application. A number of tests were performed both indoors and outdoors. From these tests, average percentage errors of the total distance and of the end point deviation were calculated, having values of 5.5% and 3.5% respectively. These results were obtained from a 200 metre route around a residential circular block walking with a medium gait and a medium pace. The accuracy of the results obtained for the implemented system were satisfactory, hence showing that such an implementation is effective in providing reasonably accurate positional information for relatively long distances regardless of GPS data availability.