Urban and Indoor Vehicular Navigation using IMU, GNSS, LiDAR, and Radar

Vehicles must be able to localize themselves in all environments (unmapped and mapped) including urban and indoor areas where Global Navigation Satellite Systems (GNSS) performance may degrade. The research and development in this thesis cover three major localization techniques that use an assortment of sensors to achieve this. In urban environments, an Inertial Measurement Unit (IMU) and GNSS fusion using the Extended Kalman Filter (EKF) is developed. For indoor environments, Light Detection and Ranging (LiDAR) Simultaneous Localization and Mapping (SLAM) and Radio Detection and Ranging (radar) SLAM systems are devised. Novel techniques are developed to tune EKF parameters using a Genetic Algorithm (GA) approach and to apply radar in a Rao-Blackwellized particle filter. The thesis presents in-depth explanations of experimental approaches as well as results that demonstrate a variety of localization systems performing high accuracy estimations in several experiments.