Improvement of terrain referenced navigation using a Point Mass Filter with grid adaptation

Abstract

We propose a Point Mass Filter (PMF) with a grid adaptation algorithm suitable for Terrain Referenced Navigation (TRN) systems having linear system models and highly nonlinear measurement models. The TRN systems are once again attracting attention, with the development of related hardware and sensors that can replace the Global Positioning System (GPS) in GPS denied environments. However, conventional TRN systems implemented by the Extended Kalman Filter (EKF) have some limitations, when operated in a featureless terrain area. A global nonlinear Bayesian estimation method referred to as a PMF has been applied, to overcome these drawbacks. This method, however, also has some limitations and unclear steps. Thus, we propose an adaptive PMF method for TRN systems. This algorithm can overcome the divergence problem, when a vehicle having large initial position error flies over flat or repetitive terrain for a long time. The effectiveness and the performance improvement of the proposed PMF are verified through computer simulation for TRN, compared to the EKF and two conventional PMF methods.