Research On Multivariable Robust Position And Attitude Control Of Work-class ROV

Underwater vehicles expand the abilities of people in marine resources development and marine scientific research greatly. Therefore, the development of underwater vehicles has got great attentions by scientists, engineers and military experts. In recent years, the deep-sea work-class ROVs have got more and more attentions for their strong working abilities. In this dissertation, we choose the robust control of work-class ROVs'dynamic positioning and trajectory tracking as the research subject, and choose the HW-I work-class ROV which is developing in ZHEJIANG university as the research object. We systematically solove the problems of multivariable nonlinear robust control of work-class ROV. The method proposed in the dissertation can not only be used in work-class ROVs, but can also be used in other types of underwater vehicles.The dissertation is divided into eight chapters, each chapter is summarized as follows:In first chapter, the history and current situation of underwater vehicles, especially the work-class ROVs, are introduced, and the advantages and disadvantages of navigation technology of underwater vehicles are also been analyzed. Then the history of control methods of underwater vehicles are reviewed detailed. At last, the research significances, research difficulties and research contents are presented.In chapter two, the development backgrounds, overall structure design and each subsystems of HW-I work-class ROV are introduced.In chapter three, for the performance of hydraulic propulsion system affects the control performance of ROVs significantly, so studying the nonlinear properties of hydraulic propulsion system is needed. The nonlinear properties of hydraulic propulsion system are obtained by theoretical and experimental studies, and based on this, system can be compensated and adjusted to reduce its influences to ROV control performance. In this chapter, different types of hydraulic propulsion system of ROVs are introduced and contrasted firstly. Then the hydraulic propulsion system of HW-I work-class ROV are presented. Based on this, the nonlinear properties of system are analyzed. And then, the steady-state and dynamic model of the system are obtained by experiments. The auto heading and auto depth control performances of HW-I ROV are obtained by tank test to prove that the basic control performances of HW-I ROV can meet the requirement of advanced control method.In chapter four, according to the requirements of high accuracy position, attitude, speed and angular speed measureing of work-class ROVs in small region dynamic positioning control, a passvie arm (PA) position and attitude measuring sensor system is introduced. Firstly, the background of PA is reviewed, and the structure design method of PA is proposed. The6-DOF position and attitude measuring recursive calculation method of PA is derived. Then the6-DOF speed and angular speed measuring recursive calculation method is also derived. At last, the measuring accuracy of PA can meet the requirements of work-class ROVs'dynamic positioning has been proved by a series of tests.In chapter five, the movement and force of underwater vehicles are analyzed to estimated a general nonlinear mathematical model. And then, by reasonable assumptions, the nonlinear mathematic model of work-class ROV which can be used to design the controller is established. And the6-DOF nonlinear simulation model of HW-I ROV is built to prove the effectiveness of control algorithm.In chapter six, for the properties of complex disturbances, parameter uncertainties and strong nonlinear coupling of work-class ROVs, a multivariable nonlinear robust controller design method for work-class ROVs is proposed. This method is based on adaptive backstepping and sliding-mode control method. By Lyapunov stability theory, the locally asymptotically stability of system with system parameter uncertanties and unknown bounded disturbances have been proved. To simplify the realization of controller, the system model is reduced and a simplified4-DOF model is obtained by considering the characteristics of work-class ROVs at dynamic positioning. The simulation results proved that the multivariable nonlinear robust controller has better control and robust performances than traditional PID controller. In tank tests, high accuracy small region dynamic positioning control were realized. In the tests, the system states information that controller needed were measured by PA sensor system. And the effectiveness of the controller was further proved to be compared with the test results of PID controller.In chapter seven, for the6-DOF position and attitude control of work-class ROV with large parameter uncertainties and only position and angle measurement, a novel nonlinear robust control algorithm based on adaptive gain smooth sliding observer and multivariable integrated backstepping controller is designed to realize the robust output feedback control of underwater vehicles. By Lyapunov stability design method, the stability of observer-controller system is guaranteed. The adaptive gain smooth sliding observer eliminates the high frequency switching that usually appeared in traditional sliding observer, and obtains a smooth velocity estimation. And in the presence of model uncertainties and bounded unknown disturbances, the velocity estimation error will converge into a small ball field by exponential speed. The multivariable integrated backstepping controller can also guarantee the tracking error converge into a small ball field. For the6-DOF control of HW-I ROV, some comparative simulation results were performed using new proposed controller and traditional PID cotnroller respectively. The simulation results proved that in the presence of parameter uncertainties, unknown disturbances and measurement noise, the new proposed controller has strong robustness, and the underwater vehicle can track the desired trajectory with good dynamic performance and steady accuracy at the same time. Its performance is better than PID controller obviously. And the restriction that the set point of heading can not greater than90degrees in PID controller has been removed. In tank tests, high accuracy small region dynamic positioning control were realized and further proved the effectiveness of controller.In chapter eight, the main work results and innovations of the dissertation are summarized. Future work and the research orientation of work-class ROVs are prospected.