Elastic Formation Control Of Unmanned Surface Vessel And Remote Operated Vehicle Under Network Attack

To meet the needs of ocean-going deep-sea efficient operation and deal with the increasingly rampant potential network threats at sea,the network formation control system with excellent performance and supported by security strategy has become an indispensable key link in engineering operations.let the long-distance macro-control intelligent ship-underwater robot equipment group to complete the formation cooperation task with high efficiency and high quality become a reality.Based on this goal,this paper analyzes the internal relationship of control instability caused by easily concealed,low-cost and highly destructive deceptive attacks tampering with system input,at the same time,further dig into the inherent coupling characteristics of complex input anomalies caused by actuator faults and network attacks,and then analyze the different factors of different types of random disturbances that affect the effect of formation control.Finally,from the control point of view,this paper designs a formation cooperative elastic control system to solve related problems,and provides a safe and effective control strategy reference for marine equipment formation operation.This paper takes the bad degree of damage caused by network attacks to the output of the controller as the main line of research,and takes the formation scheme design of intelligent ship-underwater vehicle,equipment failure and random motion as the auxiliary line,This paper will carry out the following research work:To begin with,using graph theory to construct a communication topology network for the formation system of intelligent ships and underwater robots,a switch is designed to achieve transmission allocation regulation under mixed communication.Furthermore,the geometric relationship and motion characteristics between formation equipment are further analyzed,and the formation position geometric constraint relationship is established through geometric transformation methods,thereby transforming the formation control problem into a motion tracking control problem.Based on actual marine operating conditions,a leader-follower centralized control strategy is adopted,and the desired task trajectory is designed as the highest-level virtual leader.The heterogeneous models are established as matrix-type formation state space equations to solve the problem of difficult-to-unify control and regulation of different devices and states of different orders,which is beneficial to achieve formation control of multiple types of marine equipment.In addition,according to the high efficiency operation requirements of ocean engineering,further consideration of unknown disturbances caused by internal and external factors of equipment,as well as the problem of tampering and destruction of formation control instructions caused by covert deception attacks,the control system needs to have better adaptive elastic response adjustment capability and robust performance.Therefore,starting from the sliding mode reaching phase and sliding mode phase,based on the dynamic characteristics of sliding mode control,a sliding mode damper is designed to construct a variable damping reaching law,which realizes dynamic control of rapid convergence for large errors and decelerated convergence for small errors,reduces chattering,and improves sliding mode robust response.Thus,a new terminal sliding mode surface that can avoid singular problems is constructed to achieve finite-time convergence on the sliding mode surface.Therefore,while responding to sudden attack impacts with sliding mode robust response,the expected formation tasks can also be completed quickly.In addition,combining with event-triggered mechanism and adaptive technology considering input saturation characteristics,the design of a variable damping terminal sliding mode controller is completed.Furthermore,to address the problem of random disturbances caused by different types of ocean disturbances such as wind,waves,and currents on the sea surface and sea floor,as well as matching disturbances caused by the execution structure or communication conversion inside the equipment,an adaptive radial basis function(RBF)neural network approximator is designed based on the time correlation of disturbances to achieve online adaptive compensation.Furthermore,considering that some random problems caused by network attacks,umbilical cables,and wind,waves,and currents have Brownian motion characteristics,the random phenomena of the equipment are defined as the vibration problems generated by internal and external random excitations of the formation system based on the macroscopic system view.The generalized force analysis method is used to establish the driving source random force model,and finally,the Stratonovich stochastic dynamic model is constructed to provide a model reference for controller design.Finally,to address the problem of compound anomalies affecting the system input under the simultaneous action of coupled deception attacks and failure faults,a nonlinear saturation filtering mechanism is designed based on the saturation characteristics of the equipment to perform nonlinear fitting on the controller output.Then,using the bounded properties after fitting,an analytical standard quantity is designed and mapped to Euclidean space with the system input.An elastic saturation analytical mechanism is established using the Euclidean distance quantification method to decouple the coupled input anomalies.Furthermore,an RBF neural network is utilized in the controller design to dynamically approximate the virtual parameters of the decoupled input anomalies and achieve online estimation and compensation of the input anomalies.At the this paper,the Lyapunov stability analysis of the proposed control algorithm is carried out,and it is proved that the bounded consistent convergence can be achieved in finite time.In addition,the corresponding contrast control system is designed,and the effectiveness and superiority of the algorithm proposed in this paper are verified by simulation.