Shahed University

Robust ROV path following considering disturbance and measurement error using data fusion

M. Hosseini | S. Seyedtabaii

Date :  2016/01/01
Publish in :    Applied Ocean Research

Keywords :Robust, following

Abstract :
tROV accurate path following is challenging due to system unmodeled dynamics, disturbances and naviga-tion sensors error. The model uncertainty and disturbances are commonly treated using robust methodssuch as the sliding mode controller where by incorporating an integral action in the zero tracking erroris also guaranteed. Practically, the ROV position data is often computed using low cost inertial measure-ment unit (IMU) with outputs contaminated with bias and noise. Failure of mission is an immediateconsequence of employing such biased sensors. However, the problem can be circumvented using theconcept of redundant measurements and data fusion. In this respect, a set of 12 measurements from IMU,magnetometer and Doppler velocity log (DVL) are employed where the last two are aided sensors. Theset up is shown to be capable of providing ROV path following with zero (in average) steady state track-ing error irrespective of its dynamic parameters, environmental disturbances and erroneous data; as if itenjoys the exact values of the position of the ROV. It means that the combined DVL and magnetometer aresufficient for filtering the IMU biased measurements. Various simulations conducted confirm the results.