Research On Guidance And Control Alogrithms Of Autonomous Unmanned Surface Vechile

Unmanned surface vehicle(USV)has lots of advantages such as better concealment,smaller volume,stronger survivability,and others,which can be applied widely in the exploration of marine resources and maintenance of maritime security.However,there is a requirement for USV to further improve the autonomic performance and achieve a completely autonomic application,since the use of USV has been in semi-autonomous.Guidance and control are both the key technologies in motion control of USV.Therefore,in view of the specific application of guidance and control technologies in the autonomous navigation system of USV,this thesis mainly takes researches on the path planning,global path tracking and control,as well as obstacle avoidance trajectory planning algorithms.The main contributions are as follows:The horizontal three degrees of freedom for USV motion model and the perturbation model under the disturbance of ocean current are established,which lay the foundation and theoretical supports to study the specific guidance and control algorithms of autonomous USV in the following chapters.For the global path planning problem of USV,an algorithm based on the particle swarm optimization(PSO)and Fermat's Spiral(FS)is proposed.Considering the navigation features of USV,the PSO global path planning algorithm is first used to search the waypoint sequence and connect it in order to obtain feasible,shortest,and absolutely safe polyline path.The strategy for constructing heading and curvature-continuous path is then designed by using the properties of FS has a zero curvature at its origin and its curvature varies continuously,so that the obtained path can be applied to the motion control such as accurate path following.Furthermore,taking into account the maneuverability limitation of USV,the minimum radius of rotation constraint is added to the FS transition curve design,and the effectiveness of the proposed algorithm is finally verified by simulation experiments.For the path following problem of the underactuated USV under external disturbances such as ocean currents,the control strategy is designed using cascaded theorem,which comprises the line of sight(LOS)guidance principle and adaptive techniques combined with a sliding mode.Through the coordinate transformation of the asymmetric motion model,a typical underactuated system model is first obtained,making convenient to the controller design and stability analysis.Considering the global stability problem of the system,the sliding angle is then redefined and the dynamic of the path-following errors are described under the non-singular SF coordinates,which is further arranged as a cascade system.Next,taking the underactuation of USV into account,the path-parameter update law is used as the additional control input,and the position-following control law is designed based on the time-varying distance LOS guidance law;for the ocean current disturbance,the adaptive techniques combined with a sliding mode are used to design the heading-and speed-following control law;based on the cascaded system theory,the uniform semiglobal exponential stable(USGES)and uniform globally asymptotic stable(UGAS)of the path-following control system are analyzed and obtained.Finally,the simulation experiment is presented to verify these theoretical results.For an obstacle-avoidance trajectory-planning problem of USV,a multi-phase trajectory-planning algorithm based on adaptive ph-Radau pseudospectral method is proposed.Considering the coordination and effectiveness of obstacle avoidance,and following the International Regulations for Preventing Collisions At Sea(COLREGs),the avoidance waypoints and forbidden zones are fist set up.Then,the optimal trajectory-planning model for continuous-time obstacle avoidance containing constraints such as forbidden zones,maneuverability and state equations is established,and the resolving strategy based on multi-interval Radau pseudospectral discretization method is given.Furthermore,the adaptive ph-Radau pseudospectral discretization resolving strategy is given when considering the balance on the solution accuracy and velocity.Finally,the avoidance waypoints are considered and the concept of interior point is introduced.The avoidance waypoints are then transformed into interior point constraints,and then the obstacle-avoidance trajectory planning is transformed into multi-phase trajectory-planning problem and solved by the adaptive ph-Radau pseudospectral discretization strategy.The effectiveness of the proposed algorithm is further verified by simulation experiments.