GPS/INS/Odometer/DR Integrated Navigation System Aided with Vehicular Dynamic Characteristics for Autonomous Vehicle Application

Abstract

Accurate positioning and orientation are of vital importance for autonomous vehicles. Global Navigation System(GPS) and Inertial Navigation System(INS) are two popular types of navigation systems and are usually integrated together to overcome each other’s defects. Based on the GPS/INS system, many other elements are introduced into it for accuracy improvement such as odometer, kinematic or dynamic models and so forth. This paper utilizes the dynamic characteristics of vehicles by constructing tire dynamic model and vehicle dynamic model. In this way, an integrated navigation system of GPS/INS/Odometer/DR(dead reckoning) is established by means of extended Kalman Filter(EKF), a practical and popular channel for data fusion. Field test is conducted to prove the algorithm feasible and valid. Test results indicate that this proposed GPS/INS/Odometer/DR integrated navigation system is able to provide accurate position and orientation information thus qualified for autonomous vehicles navigation. The test of GPS signal outage is also held and experiment results show that the proposed algorithm can also remain stable and reliable in case of GPS signal unavailability.