Hydrodynamics Test And Research On Motion Control For Deep Sea Work-class Remotely Operated Vehicle

Ocean is the origins of human being, which contains fruitful natural resources andenergy. All kinds of underwater vehicles have been playing more important roles in deepsea resources exploration and exploitation. Among which, ROV (Remotely OperatedVehicle) has been applied in many fields, such as oceanographic survey, seafloor geographymapping, pipeline inspection and offshore structure protection. It has become an importantassistant tool to accomplish different underwater missions for the sustainable developmentof newly emerging ocean industry.This dissertation originates from4500m Deep Sea ROV Operating Systems, supportedby the National High Technology Research and Development Program (863, Grant No.2008AA092301). The research is inspired by abundant difficulties existing in operation,guidance and control of underwater vehicles in that they operate in some uncertain andhazardous environments. Based upon designing data obtained by the staffs, this researchsubject is concerned with nonlinear robust control design of the deep-sea work-class ROV,which includs hydrodynamics test of scaled model, mathematics modeling of a dynamicsystem, motion stability analysis, nonlinear robust motion controller design, thrust controlallocation algorithm and space trajectory tracking control.The hydrodynamic characteristics of open-framed ROV are distinguished from otherunderwater vehicles such as AUV, HOV and underwater glider, due to lots of differencesexisting in principal dimensions, configuration and general arrangement. The maincontributions of the present work are listed as follows:1. Accurate hydrodynamics coefficients measurement of ROV are significant for themaneuverability and control algorithm. Accordingly, the scaled model of ROV wasconstructed by1:1.6, and hydrodynamic tests were carried out through VPMM andLAHPMM. Resistances along longitudinal, transversal and normal axises under differentvelocity are figured out based on the real size of ROV, respectively. Linear and nonlinearhydrodynamic coefficients are figured out by using least square method.2. Nonlinear mathematical models with5DOF (Degree-of-Freedom) of ROV areestablished according to the hydrodynamic test results and motion hypothesis. Motionstabilities on both horizontal and vertical plane are analysed based on the coefficients,which are of great significance for the structure designing and parameter selections for the ROV systems.3. The external forces acted on ROV are analysed systematically. To weaken theumbilical cable influence on ROV, the design of buoyancy balls attached on the cableterminal is introduced for the coupling effects of umbilical cable and ROV. Through thisdesigning scheme, ROV has a better maneuverability and could be controlled more easilyboth in vertical and horizontal plane.4. The problem of spacial trajectory tracking control for ROV is addressed despite ofthe constant ocean currents and parametric modelling uncertainty. A nonlinear adaptivecontroller is presented that steers ROV move along a sequence of way-points consisting ofdesired positions in Earth coordinated system. The controller is first derived at thekinematic level, and then integrator self-adaptive method and backstepping algorithm arecombined to extend the kinetics case and to deal with model parameter uncertainty, finally,the stability of the closed-loop system trajectory is proved through using Lyapunovfunctions. Several simulation cases are discussed, and the results illustrates the robustnessof the proposed controller.5. The object oriented technology and Petri net(OOPN) are novelly adopted to designthe formalized model for the soft architecture of control systems, which enable the systemsto have a better extendability, hierarchical and coherence. The3D trajectory trackingcontrol simulation platform is developed by using virtual visual technology and digitalmodelling method, including chart data processing, seafloor condition generating, vehiclemodeling and trajectory tracking displaying. The platform can also be applied into manypractical fields, such as simulation research of seafloor environment, motion andmaneuvering simulation of underwater vehicle, debugging of control system andmanipulating training. It provides an effective approach for the R&D of underwatervehicle.The original works in the present dissertation can be summarized as following:1. Hydrodynamic tests are novelly designed and researched under special operatingconditions, including test of purely yawing motion on site, oblique towing test under lowspeed and large drift angle. The oblique towing tests are conducted through LAHPMM atdesignated low speed and the whole range drift angles. Multiple regression method isadopted to address the testing data and obtain the relating hydrodynamic coefficients. Anddynamics model is derived for the next phase of controller design.2. Nonlinear robust controller is proposed to force the5DOF overactuated ROV tomove according to the given control commands despite of the presence of environmentaldisturbances, strong coupling and uncertain physical parameters. The coupling design ofcontroller is accomplished by using the so-called block strict feedback form, and the controller synthesis is based on the adaptive strategy and backstepping algorithem. Thelocally asymptotically stability of the control system with system parameter uncertainties isproved by Lyapunov direct method, and the closed loop of tracking errors can be madearbitrarily small. The nonlinear numerical simulation results show the effectiveness androbustness of the proposed multivariable coupling control algorithm. Additionally, thesimulation results of the multivariable robust controller compared to the integrationseparated PID controller are given, which further proves the advantages of the novellydesigned control strategy.3. Propulsion system is critical not only in control system, but also in maneuveringapplications, since the consequences of loss of maneuverability maybe serious. Accordingly,thruster vector configuration is designed to overcome the singularity problems which cannot produce forces/moments in every direction, such as surge force, yaw moment. Theemployment of primal dual interior point method is presented for the thrust controlallocation technology. The control allocation algorithm is formulated as optimizationproblems, where the objective is to minimize the thrust output subjected to physicalconstraints. Logarithmic barrier functions are adopted to match the bound constraints tosolve this nonlinear programming problems. Furthermore, a global convergence of theproposed algorithm is analysed via the scheme of line search methods. The simulationresults demonstrate that the developed thrust control allocation algorithm can avoid theemergence of over saturation of thrust outputs compared to the pseudo inverse method.Last but not least, the main research contents of the dissertation are summarized, andthe future research direction is presented. The algorithms proposed in the dissertation cannot only be applied into ROV being developed, but embedded in other underwater vehiclessuch as AUV, HOV and underwater glider.