NavLab, a generic simulation and post-processing tool for navigation

Abstract

The ambition of getting one common tool for a great variety of navigation tasks was the background for the development of NavLab (Navigation Laboratory). The main emphasis during the development has been a solid theoretical foundation with a stringent mathematical representation to ensure that statistical optimality is maintained throughout the entire system. NavLab is implemented in Matlab, and consists of a simulator and an estimator. Simulations are carried out by specifying a trajectory for the vehicle, and the available types of sensors. The output is a set of simulated sensor measurements. The estimator is a flexible aided inertial navigation system, which makes optimal Kalman filtered and smoothed estimates of position, attitude and velocity based on the available set of measurements. The measurements can be either from the simulator or from real sensors of a vehicle. This structure makes NavLab useful for a wide range of navigation applications, including research and development, analysis, real data post-processing and as a decision basis for sensor purchase and mission planning. NavLab has been used extensively for mass-production of accurate navigation results (having post-processed more than 5000 hours of real data in four continents). Vehicles navigated by NavLab include autonomous underwater vehicles (AUVs), remote operated vehicles (ROVs), ships and aircraft.