Absolute location of underwater robotic vehicles by acoustic data fusion

Computer assistance for the manned guidance of ROVs (remotely operated vehicles) is studied. In this approach, it is possible to estimate an optimal location by fusing data deriving from log and gyroscope or a nonoptimal location by composition of compass, depthmeter, and log data. This relative reckoning is achieved by taking into account a hypothesis on the value of the undercurrent. Therefore, a second reckoning module model is proposed to improve the ROV position close to an offshore jacket, fusing data given by an echo-sounder belt and taking into account a CAD map of the surroundings. This new estimate is independent of the undercurrent, and it is possible with the help of the operator to match this absolute model with the relative one in an asynchronous way. This matching gives a new hypothesis on the value of the undercurrent and resets the dead-reckoning module.<
>