Tight Integration of 3D RISS/GPS/Map Data for Land Vehicle Navigation Utilizing Particle Filtering

Due to interruption or degradation in satellite based navigation systems in dense urban scenarios, they have to be augmented with other systems to achieve continuous and accurate vehicular navigation. Among the systems that can be integrated with satellite navigation are: (i) dead reckoning systems, such as inertial navigation systems and odometry; and (ii) geospatial information system (GIS) such as map data and road networks. For land vehicle navigation, usually the Global Positioning System (GPS) is integrated with micro-electro mechanical system (MEMS)-based inertial sensors because of their low cost, small size, light weight and low power consumption. Despite the advantages of MEMS-based inertial sensors, they provide inadequate performance in degraded GPS environments such as downtown, urban canyons or tunnels. The inadequate performance of these low-cost sensors is because of their complex error characteristics which are stochastic in nature and difficult to model. This paper proposes a positioning solution for land vehicles based on integrating low-cost MEMS-based inertial sensors, the vehicle odometer, GPS, and map data from road networks. Despite the traditional inadequate performance of MEMS-based sensors in this problem, three methods to enhance the performance are proposed in this work to enable MEMS to be used for this navigation application: (i) The use of a three-dimensional (3D) reduced inertial sensor system (RISS) that has better performance for land vehicles than traditional full-IMU solutions; (ii) The use of map information from road networks to constrain the positioning solution; (iii) The use of advanced nonlinear filtering techniques based on particle filtering (PF) to perform the integration of 3D RISS/GPS/Map data. This usage of PF also enables the utilization of sophisticated models for inertial sensor stochastic drifts. Furthermore, PF can incorporate the map information inside the filter itself. The performance of the proposed positioning system has been verified extensively on real-life road tests. The results have been examined to verify the suitability and satisfactory performance of the proposed solution even in downtown trajectories with degraded/multipath or totally denied GPS for long durations.