A framework for sensor fusing using fuzzy logic for navigation systems

Navigational sensors are evolving both on a commercial and research level. However, the limitation still lies in the accuracy of the respective sensors. For a navigation system to reach a certain accuracy, multi sensors or fusion sensors are used. In this paper, a framework of fuzzy sensor data fusing is proposed to obtain an optimised navigational system. Different types of sensors without a known state of inaccuracy can be fused using the same method proposed. This is demonstrated by fusing compass/accelerometer and GPS signal. GPS has evolved as a choice of navigation method for outdoor autonomous system. Despite it emerging trend and application, one problem remains is that it is still prone to inaccuracies due to environmental factors. These factors are available and evaluated in the GPS receiver architecture. These inaccuracies are available in the extracted NMEA(National Maritime Electronics Association) protocols as SNR (Signal to noise ratio) and HDOP (Horizontal Dillusion of Precision). Dead reckoning sensors on the other hand does not depend on external radio signal coverage and can be used in areas with low coverage. Unfortunately, the errors are unbounded and has an accumulative effect over time.