A real-time navigation system for autonomous underwater vehicle

This paper focuses on the study and implementation of a navigation system in order to estimate position, velocity and attitude of an autonomous underwater vehicle, AUV. The extended Kalman filter, EKF, is investigated for the fusion of the sample data from different sensors: the strapdown inertial measurement unit, magnetic compass, Doppler velocity log, depth sensor, and an acoustic positioning system. Results are applied to the development of a navigation system for the Pirajuba AUV, an autonomous underwater vehicle that is being developed at the mechatronics department of the Politechnic School of the University of Sao Paulo. The navigation system is composed by off the shelf components integrated in a CAN based network. On the hardware platform, a software architecture is implemented based on free and largely known tools, like C language, and the GNU compiler. The real-time performance of the filter is validated through laboratory and field tests. The last one includes experiment using an automobile vehicle. Results in the field tests indicate the correct choice for the system model assumed in the EKF, and the good performance of the navigation algorithm in real-time. During the simulation, the accuracy obtained in the estimation of the AUV position and attitude are satisfactory.