About the interest of Linearised Kalman Filter for low-cost GPS-based hybrid positioning system for land vehicles
Abstract
GPS-based positioning systems have been widely introduced in cars and land vehicles, but, except for comfort-oriented applications such as simple navigation, the level of availability and integrity provided by these systems are far from satisfying the requirements of most of the new location-based services that are under development for road transportation. The research is very active on the topic of data fusion between GNSS and proprioceptive sensors to improve the naturally poor GPS integrity. Among the numerous fusion techniques that can be used, the well-known Extended Kalman Filter (EKF) is still the most popular, but simpler and less computation demanding techniques can also be used such as Linearised Kalman Filter (LKF). This paper discusses the design of a hybrid localisation system with low cost sensors (odometer and gyro) and compares the efficiency of the 2 filters, especially around long GPS outages, which represent the most constraining situations. Different trials have been carried out on a real circuit and illustrate the comparison.