Implementation of a ROV navigation system using acoustic/Doppler sensors and Kalman filtering

Hydro-Que/spl acute/bec developed a new underwater ROV which is use to inspect its dams. Since 2000, the ROV has allowed Hydro-Que/spl acute/bec to save 5M$ in inspection costs. This paper presents the study of a Kalman filter and its performances. Kalman filtering is used to merge data from fiber gyros, DVL, accelerometers and global positioning system in order to have a better estimation of the position and orientation of the ROV. In real systems, sensors do not send data at the same rate and data acquisition is often intermittent. One innovative aspect of this approach is that it accepts asynchronous information and delay from the sensors. Experimental results in a real environment are presented.