Robust Controller Technique Of An Autonomous Underwater Vehicle For Underwater Pole Inspection

• Main Author: Song, Yoong Siang
• Format: Thesis
• Language: English
• Published: 2018
• Subjects: T Technology; TK1-9971 Electrical engineering. Electronics. Nuclear engineering
• Online Access: http://eprints.usm.my/47410/1/Robust%20Controller%20Technique%20Of%20An%20Autonomous%20Underwater%20Vehicle%20For%20Underwater%20Pole%20Inspection.pdf; http://eprints.usm.my/47410/
• Institution: Universiti Sains Malaysia

The oil and gas offshore platform is facing unwanted marine growth problem. A periodic inspection of the submerged platform’s leg is needed. This research studies the possibility to involve an Autonomous Underwater Vehicle (AUV) in the underwater pole inspection application. A predefined inspection path is needed to increase the efficiency of the AUV in the inspection mission. Besides, a robust control technique is needed suppress the effect of unknown hydrodynamic parameters and external disturbances on the AUV system. As a solution, this research proposed a time-optimal inspection path specifically for vertical pole inspection by using grid-based Coverage Path Planning (CPP). A planar map is modelled to represent the 3D target environment of the pole. Five patterns of inspection paths are designed and compared to choose the best inspection path among them. Moreover, a robust tracking control law combined with robust filter and fuzzy logic control techniques is proposed. The robust filter control technique is used to compensate the effect of added mass, hydrodynamic damping, model nonlinearities, coupling effect, and external disturbances on the AUV system, whereas the fuzzy logic control technique is used to improve the force profile. Furthermore, an open frame box-shaped AUV prototype which is suitable for pole inspection application is developed to verify the performance of the proposed controller. The proposed inspection path is designed based on Boustrophedon’s motion with smooth turning and vertical sweep direction. Calculation results show that the proposed inspection path can cover 99 % of the pole’s surface by the shortest inspection time and shortest trajectory length compared to others. Besides, simulation results show that the force profile of the proposed controller has been improved by the added fuzzy logic controller. The maximum control force generated by the proposed controller is 8.7 N less than that of the robust filter with high natural frequency. From the simulation results, the frequency of control signal given by the proposed controller is 0.1 Hz, which is lower than 2 Hz of Model Free High Order Sliding Mode Control (MFHOSMC) and 100 Hz of Time Delay Control (TDC). Besides, the maximum steady state error of the proposed controller is 0.0023 m, which is comparable to 0.0003 m of MFHOSMC and 0.0095 m of TDC. Furthermore, experiment results showed that the control signal of proposed controller has lower average switching frequency and lower saturation percentage than the MFHOSMC and TDC.