A Low-Order DGPS-Based Vehicle Positioning System Under Urban Environment

Abstract

The vehicle positioning system is a key component in functions such as vehicle guidance, driver alert and assistance, and vehicle automation. Since installing a low-cost global positioning system (GPS) or inertial navigation system (INS) unit is becoming a common practice in vehicle applications, its involvement in vehicle guidance and vehicle safety deserves a closer investigation. Typical vehicle applications require high reliability, low cost, and sufficient accuracy under all operational conditions. For GPS-based positioning, urban driving with its complicated maneuvers, frequent GPS blockage, and multipath, are some of the most difficult driving environments. This paper explores the feasibility of a low-order vehicle positioning system functioning under an urban environment. The equipped vehicle has a midrange differential GPS (DGPS) unit and few relatively simple in-vehicle sensors. A low-order integration is explored by utilizing a vehicle model-based extended Kalman filter (EKF) to incorporate in-vehicle motion sensors and to largely avoid direct integration of INS signals. Further, the characteristics of DGPS measurements under urban environments are investigated, and novel DGPS noise processing techniques are proposed to reduce the chances of exposing the EKF to undesirable DGPS measurements due to common DGPS problems such as blockage and multipath. A resulting fourth order EKF based positioning system is successfully implemented in the test vehicle to demonstrate the feasibility of the proposed design. Experimental results illustrate the ability of the system to meet the accuracy and robustness requirements in the presence of blockage and multipath under a typical urban driving environment