Font Size: a A A

Three-Dimensional Control For High-speed Trajectory Tracking Of Underactuated UUVs

Posted on:2021-02-16Degree:DoctorType:Dissertation
Country:ChinaCandidate:M WangFull Text:PDF
GTID:1362330605480326Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
In recent years,the topics of navigation,guidance and control of unmanned underwater vehicles(UUVs)have attracted considerable attention from the researchers around the world.The main reasons are not only the increasing applications in the field of ocean engineering,but also the theoretical challenges of multiinput-multioutput(MIMO)nonlinear systems.Among these practical applications,the first core problem that need to be resolved is the task for accurate tracking and control of UUVs.Especially for the trajectory tracking in three-dimensional(3-D)space,the controller design presents strict constraints of the task states on the relations of time and space,which includes the full-state time-varying control of position,attitude and velocity.Therefore,this dissertation considers a mission scenario that an UUV performs the maneuver tasks in high speed.But the underactuated vehicle is equipped with only stern propellers,steering and diving rudders.As a result,the sway and heave motions cannot be controlled,and the roll motion can be neglected.Then,the control methodology of 3-D trajectory tracking is investigated.The main works are given as follows:(1)3-D trajectory tracking control of underactuated UUV with non-diagonal system matricesFor a general class of underactuated UUVs with non-diagonal inertial and damping matrices in the case of only starboard-port symmetry,two control strategies that consist of command filtered backstepping control(CFBC)and adaptive fast nonsingular terminal sliding mode control(AFNTSMC)are proposed.Different from the traditional assumptions that the UUVs have three symmetrical planes,here the system inertial and damping matrices can not satisfy the diagonal conditions.Firstly,In CFBC,the UUV guidance system is revised by control input and state transformations;then the kinematic and dynamic model of underactuated UUV is derived into the conventional diagonal normal form;to avoid the “complexity expansion” problem in backstepping control design,a low-pass filter is introduced in dynamic surface control;in the end,using the Lyapunov stability theorem,the closed-loop control system is proven to be uniformly ultimately bounded.Secondly,in AFNTSMC,the system state transformation and output redefinition are introduced,which not only achieve the diagonal normal form for the mathematic model of underactuated UUVs,but also change the control structure of the nonminimum phase;in order to improve the convergence and response speed of trajectory tracking errors,the combination of integral sliding mode and terminal sliding mode is conducted;then the relations of parametric model uncertainties,bounded disturbances and UUV velocities are analyzed such that the design of compensation based robust adaptive sliding mode control is presented.Finally,the simulations sufficiently demonstrate the effectiveness of the proposed control strategies,which guarantee the precise tracking in 3-D space even for an underactuated UUV with non-diagonal system matrices.(2)3-D trajectory tracking control of underactuated UUV with system modeling uncertaintiesFor the 3-D trajectory tracking problem of underactuated UUVs subject to system parametric uncertainties and external environmental disturbances,the designs of robust adaptive sliding mode control,backstepping and finite-time control based stochastic time-varying disturbance theorem are presented,respectively.Firstly,in robust adaptive sliding mode control,the dual closed-loop integral sliding mode controller is utilized to enhance the robustness of the system,which takes advantage of the insensitivity to the perturbation of model parameters and bounded disturbances;to save the time of online learning,the direct adaptive neural network controller is revised with only one online adaptive parameter;and then the conditional integrator of bounded disturbance estimate is introduced,which well inherits the advantages of proportional-integral(PI)control and sliding mode control.Secondly,in stochastic time-varying control,the nonlinear stochastic dynamical equations of underactuated UUVs are derived using Winner process;then with the combination of Lyapunov theorem and backstepping technique,the 3-D trajectory tracking controller is proposed;to ensure more fast response and convergence speed,aforementioned backstepping control is extended to the finite-time control such that the closed-loop system of all trajectory tracking errors is UUB.Finally,the simulations sufficiently demonstrate the effectiveness of the proposed controllers,which guarantee the precise tracking in 3-D space even for an underactuated UUV with system modeling uncertainties.(3)3-D trajectory tracking control of underactuated UUV without the velocity measurementsThis part aims at the control problem of an underactuated UUV with only position and attitude measurements,but without velocity feedback.The proposed nonlinear observer-controller schemes are respectively based on UUV dynamical model,adaptive neural networks and bio-inspired model,which not only guarantees the online real-time velocity estimation of underwater vehicles,but also achieves the precise trajectory tracking in 3-D space.Firstly,to address the autonomous underwater homing and docking problem of underactuated UUVs,a novel guidance system for initial homing to generate a feasible reference trajectory is presented;then based on UUV dynamical model without disturbances,the velocity observer and output feedback controller are designed such that the underactuated UUV can accurately track the 3-D desired trajectory in initial homing.Secondly,to work out the problem of modeling parametric uncertainties and bounded disturbances,an extended state observer-controller scheme based adaptive neural network is presented.Moreover,a novel state observer based bio-inspired model is proposed for underwater moving target tracking,which is effective to eliminate the drawback that the vehicle velocity estimations rely heavily on dynamical model;the rigorous system stability analysis using Lyapunov method and simulations also validate the uniformly ultimately bounded conclusion of the closed-loop trajectory tracking control system.Finally,the dessertation respectively considers the practical applications of initial homing and underwater moving target tracking as mission scenarios,and thus designs the simulations of 3-D trajectory tracking control.It sufficiently illustrates the effectiveness of the proposed nonlinear observer-controller schemes,which can achieve 3-D trajectory precise tracking of underactuated UUVs even without the velocity measurements.In a word,the dessertation investigates and solves the 3-D control for high-speed trajectory tracking of underactuated UUVs mainly from the following three viewpoint: UUV system parametric model,time-varying stochastic disturbances,and system state observation.
Keywords/Search Tags:underactuated unmanned underwater vehicle, three-dimensional trajectory tracking, filtered backstepping techniques, adaptive sliding mode control, neural network
PDF Full Text Request
Related items