Smart Real-Time Autonomous Navigation System using integration of MEMS-based Low-Cost IMU/GPS

Abstract

This paper aims to design a real time autonomous navigation system (ANS) using low cost electronics devices with significant navigation accuracy levels. This system could be practically implemented inside any vehicle to enable autonomous movement. The proposed ANS is implemented through an open loop decentralized loosely integration approach of an Inertial Measurement Unit (IMU) and a Global Positioning System (GPS). Due its modularity, the 6 degree of freedom (DOF) MPU6050 breakout board is used to be the IMU in the system. The GPS receiver is selected to be the Skylab SKM53 module. Appropriate real time navigation accuracy levels are obtained through the correction of Position and Velocity (PV) readings coming out from the GPS. Two Extended Kalman Filters (EKFs) are used to provide the corrected PV readings. The filters are implemented through two state machines that are running simultaneously in a concurrent form on an Atmel ATmega16 microcontroller which is the main controller of the proposed ANS. The system also includes 3DRobotics 915 MHz wireless telemetry module to grant the real time processing of the navigation data. The system accuracy is validated on a real vehicular trajectory. The Root Mean Square Error (RMSE) values for the implemented ANS in ENU directions, have reached 0.254, 0.337 and 0.462 meters respectively, which is notably competitive with other recent related work.