Abstract of Thesis presented at COPPE/UFRJ as a partial fulfillment of the requirements for the degree of Master of Science (M.Sc.)

Inertial Navigation Unit Aided by GPS Using Kalman Filter

Paulo Sérgio Pastore

March/2004

Advisors:

Liu Hsu Ramon Romankevicius Costa

Department:

Eletrical Engineering

      This thesis presents one of the possible position information fusion process of an Inertial Nayigation Unit (INU) and a GPS receiver, to estimate the INU position deviations. The focus of this thesis is on vessels applications. The stochastic Kalman filtering method is used to estimate the INU deviations trough the position informa­tion fusion process. A trajectory generation algorithm is used to simulate two different trajectories. The position coordinates generated by this algorithm are the basis to sim­ulate the GPS coordinates. White noises corrupt the simulated inertial sensors signals, which are used to generate the INU trajectories related to the reference trajectories. Feedforward and feedback configurations are implemented to correct the INU devia­tions. Based on the results, it can be concluded that the position information fusion process improves the INU autonomy comparing with INU unaided.