Inertial navigation with dead reckoning

Abstract

Positioning is gaining importance not only in military activities, but also in people’s daily life. However, GPS is not always available. The satellite signals could get blocked or degraded by buildings, trees, and canyons. To counter the problem faced in environments with poor GPS signals, it is desirable to have a small module using techniques other than GPS to provide timely and accurate positioning information.

An inertial navigation device has built in sensors and can provide positioning services using dead reckoning. It relies on accelerometer to calculate the distance travelled, magnetometer to get the direction using the earth’s magnetic field as a reference. The incorporation of gyroscope could identify the orientation of the device as well as sense any turnings made by the user. By combining this information with the starting position known, one could calculate one’s current position using dead reckoning technique. One of the most significant advantages of inertial navigation device is that it could function normally in areas with no external signal reception since it relies solely on internal data.

These sensors are now commonly built into the smart phones. Understanding the characteristics of these sensors could help us estimate the possibility of incorporate positioning functions into the smart cellphones using dead reckoning. Furthermore, if we could use Wi-Fi hotspots as reference to calibrate the positioning information, we could develop a system to provide positioning services and even improve the positioning accuracy for massive users.

Starting the project from scratch, the first step of the project is to study the characteristics of a typical inertial navigation device. In addition, multiple experiments are carried out to analyze its performance under different environments. In this thesis, the characteristics of the device are studied using a statistical approach and factors affecting it performance is analyzed.