3D space mapping using inertial navigation with GA-KF

Abstract

Dead reckoning module is a form of inertial navigation system which uses the combination of accelerometer, gyro and compass to work similarly as Global Positioning System (GPS). This study integrates a hybrid of Genetic Algorithm and Kalman Filter to a dead reckoning module for accuracy improvement. This study successfully developed an experimental device using accelerometer, gyro and compass, and a microcontroller for embedded implementation of Kalman Filter and Genetic Algorithm. Simultaneously, a three-dimensional coordinates was taken from the experimental device and the GPS when it is navigating under water. Granted that the objective is for the device to operate and give systematic accurate results, measuring and mapping a certain area undeterred by interferences or obstructions. Based on the conducted experiments under given circumstances and environment where obstructions such as buildings, walls, body of water, the device presented that utilizes Genetic Algorithm Based Dead Reckoning and Kalman filter can be accepted as a way to map places with obstructions aforementioned with an accuracy of 2meters, therefore presenting the high possibility of 3D Space Mapping.