Cooperative Path Following Control For Autonomous Surface Vehicles Subject To Limited Communication Resources

The ocean is one of the most valuable resources on the earth,and it plays an crucial role in the survival and sustainable development of humanity.Autonomous surface vehicles(ASVs),as tools for exploring and exploiting the ocean,have been widely used in both civilian and military fields in recent years due to their high autonomy,good flexibility,and low cost.Multiple ASVs can interact and collaborate through a communication network,greatly improving work efficiency,reducing operation costs,and enhancing system fault tolerance.They can adapt to complex working environments and task requirements,making them the future development trend.Cooperative path following control refers to multiple ASVs tracking given paths and maintaining a desired formation.It has broad prospects in practical applications such as resource exploration,maritime search and rescue,and ocean environment monitoring.This thesis focuses on the underactuated ASV subject to limited communication resources,and studies the cooperative path following control problem for the ASVs in the context of nonperiodic communication,non-periodic sampling,non-periodic control,and the presence of humans in the loop.The main works of this thesis are stated as follows:Firstly,this thesis investigates the cooperative path following control problem of the ASVs subject to limited communication network bandwidth resources,onboard computing resources,and actuator resources.An static event-triggered cooperative path following control method featured with aperiodic communication,sampling,and control is proposed.In the coordination layer,an event-triggered path updating law is designed to reduce the network traffic by introducing an event-triggered mechanism.A local path variable predictor is designed to estimate the path variables of the neighboring ASV in the communication interval.In the guidance layer,an event-triggered guidance law is designed by introducing the event-triggered sampling mechanism to reduce the sampling and computation.In the control layer,an eventtriggered extended state observer and an event-triggered kinetic control law are designed.The event-triggered extended state observer can estimate the total disturbances of the ASV by using the aperiodic sampling data.Based on the estimated disturbances,the event-triggered kinetic controller achieves aperiodic disturbance rejection control,reducing sampling,computation,and execution.Cascade stability analysis and Zeno analysis prove that the closed-loop control system is input-to-state stable(ISS),and Zeno behavior is excluded.Simulation and experimental results verify the effectiveness of the proposed static event-triggered cooperative path following control method for the ASVs.Furthermore,this thesis further investigates methods for reducing the steady-state network traffic,as well as the cooperative path following control problem of the ASVs subject to internal uncertain dynamics and external disturbances caused by wind,waves,and currents.A dynamic event-triggered cooperative path following control method is proposed.In the coordination layer,by introducing a dynamic variable,a dynamic event-triggered mechanism is designed,and then a dynamic event-triggered path updating law is proposed to reduce network traffic both in transient and steady states.The path variable predictor is also applied to estimate the path variables of the neighboring ASVs in the communication interval.In the guidance layer,a guidance law based on line-of-sight guidance is designed to guide the ASV to follow the desired path.In the control layer,a super-twisting observer is designed to estimate the total disturbances within a finite time.Based on the estimated disturbances,a super-twisting dynamic control law is designed to achieve finite-time control of the vehicle kinetics.Via cascade system theory and Zeno analysis,it is proved that the closed-loop control system is ISS,and Zeno behavior is avoided.Simulation results show that the proposed dynamic event-triggered method triggers fewer events in constrast to the static event-triggered method,thereby occupying less network bandwidth resources,and it verifies the effectiveness of the proposed dynamic event-triggered cooperative path following control method.Subsequently,this thesis investigates the cooperative path following control problem of the ASV in the presence of limited communication network bandwidth,unmeasured velocities,and static/moving obstacles in the waterway.A self-triggered output-feedback collision-free cooperative path following control method featured with aperiodic network listening is proposed.In the coordination layer,by introducing a self-triggered communication mechanism,a self-triggered path updating law is designed to reduce network transmission and listening times,simultaneously.A path variable predictor is also designed to estimate the path variables of the neighboring ASVs in the communication interval.For the control of per single vehicle,a robust exact differentiator(RED)based observer is designed to estimate the velocity and the total disturbances,simultaneously.Then,output-feedback super-twisting control laws are designed to control the position and heading of the vehicle,respectively.In the position control loop,an artificial potential field based on the relative distance and velocity is designed to avoid collisions between the vehicles and the dynamic/static obstacles.Unnecessary collision avoidance actions can be reduced by taking the relative velocity into consideration,additionally.By using cascade system theory and Zeno analysis,it is proved that the closed-loop control system is ISS,and the proposed self-triggered communication method does not cause Zeno behavior.Experimental results verify the effectiveness of the proposed self-triggered outputfeedback collision-free cooperative path following control method.Finally,this thesis investigates methods to reduce the network listening times,as well as the path following control problem of the ASVs in the presence of unmeasured states and human operator’s supervision in the network.A human-in-the-loop output-feedback cooperative path following control method is proposed.In the coordination layer,by introducing an operator’s speed commands,a human-in-the-loop virtual leader is designed.The operator can change the path updating speed of the virtual leader to adjust the sail speed of the ASVs,indirectly,which realizes the human-machine integration.This allows the operator to adjust the formation mission or to respond to emergencies,avoiding collisions with dynamic obstacles.Then,the control model of the ASV is transformed,and the path following control problem is transformed into a control problem of a second-order along-track error dynamic system and a third-order cross-track error dynamic system.Next,a second-order and a thirdorder RED observers are designed to estimate the unmeasured states based on the position information of the vehicle.Finally,based on the estimated states,output-feedback continuous twisting control laws are designed to achieve path following control by using only position information.Through cascade system theory analysis,it is proved that the closed-loop control system converges in finite time.Simulation results verify the effectiveness of the proposed human-in-the-loop output-feedback cooperative path following control method.