Marginalized particle filtering for reliable land vehicle navigation in global navigation satellite system-denied environments

Abstract

Accurate localization in land vehicle navigation systems is highly dependent on the global navigation satellite system (GNSS). However, GNSS signal outages are common in urban areas due to obstacles such as tall buildings and tunnels. To mitigate these issues, digital road maps and dead reckoning sensors, like odometers, are often integrated to provide continuous vehicle localization. This paper presents a robust estimation method to solve the fusion problem of GNSS, odometer, and digital road map measurements in the presence of GNSS out- ages. The proposed solution utilizes a marginalized particle filter (MPF), which combines the robustness of particle filtering with the efficiency of a Kalman filter to handle the linear and non-linear parts of the state and/or measurement equations, respectively. When GNSS signals are unavailable, the MPF fuses all available pseudo-range data with odometric and map information to enhance vehicle positioning. The effectiveness of the proposed method is demonstrated using real-world data in an urban transportation scenario, highlighting signifi- cant performance improvements and real-time application potential.