A heading observer for ROVs under roll and pitch oscillations and acceleration disturbances using low-cost sensors. (1st December 2015)

• Record Type: Journal Article
• Authors: Le, Khoa Duy; Nguyen, Hung Duc; Ranmuthugala, Dev; Forrest, Alexander
• Is Part Of: Ocean engineering. Volume 110 Part A (2015)
• Page Start: 152
• Page End: 162
• Publication Date: 2015-12-01
• Subjects: ROV -- Kalman filter -- Heading observer -- Inertial sensor -- Magnetometer; Ocean engineering -- Periodicals; Ocean engineering; Periodicals; 620.4162
• DOI: 10.1016/j.oceaneng.2015.10.020
• Languages: English
• ISSNs: 0029-8018

Abstract

Abstract: This paper presents the development of a heading angle observer for Underwater Remotely Operated Vehicles (ROVs) during dynamic conditions in the presence of roll/pitch oscillation and acceleration disturbances. Unlike previous algorithms, which focused on deep water working environments where the external accelerations and oscillation were rather small, this paper proposes a heading observer that deals with the effects of near surface working conditions on ROVs, such as waves or unexpected water currents. These effects cause significant roll and pitch disturbances as well as high-acceleration drift motions, which affect the measurements from the incorporated sensor system. To overcome this situation, a novel heading observer is proposed, consisting of two Kalman filter (KF) steps: one for acceleration elimination and one for roll and pitch oscillation compensation. The data from the inertial measurement unit and the magnetometer are combined to estimate and compensate for external accelerations and roll/pitch disturbances by adjusting the covariance values. Free running tests, based on a physical ROV model, were conducted under various working environments to verify the performance of the proposed observer. The comparison with previous algorithms was also made to verify the effectiveness of the proposed algorithm. Highlights: The novel two-layer Kalman filter for ROV heading estimation under disturbances. Experiments based on low-cost inertial sensor and 2-D digital compass. Comparison between the two-layer Kalman filter with traditional algorithms.