Research On Robust Cooperative Control Of Swarm Systems For Uncertain Target Tracking

The swarm cooperative control of motion carriers such as UAVs and robots,plays an important role in the fields of industrial and agricultural production,disaster relief,and geological surveys.Consensus is the fundamental of swarm tracking,formation and other cooperative control behaviors.In practical applications,the dynamics of the vehicles in different media are generally modeled in the form of fractional differential equations,and their models are affected by various factors and exist uncertainty.Therefore,it is of great significance to study the swarm consensus under the fractional-order model and the influence of uncertainty.This thesis first designs a robust consensus tracking control protocol for integer-order leaders with model uncertainty,and the followers are in the presence of model uncertainty and external disturbances.Then,the relevant results are extended to the tracking of the fractional-order leaders.Finally,the theoretical results are applied to design the consensus control protocol of the quadrotor swarm,and the relevant interference suppression technology is effectively integrated to realize the robust cooperative control of the swarm.The main work and innovative results of this thesis are summarized as follows:1.Study on the problem of swarm robust consensus tracking control under the condition of model uncertainty.Based on the neural network(NN)estimation technology,by constructing the Lyapunov function,the rate of change of the relevant estimated parameters is generated online to realize the estimation of the uncertainty components in the leader's dynamic equation;in addition,uncertainty and disturbance estimator(UDE)and neural network estimation technology are used to estimate the model uncertainty and external disturbances appearing in the follower agents' model.These methods effectively suppress the negative impact of the system uncertainty components on the control performance,and improve the system's robustness.In particular,the cascade UDE(CUDE)structure designed in this thesis can estimate and compensate the disturbances in the follower agents' position-loop and velocity-loop,respectively.Then the consensus tracking controller is designed based on the backstepping method.Finally,the simulation experiments are designed to verify the effectiveness of the control protocol.Compared with the non-cascade UDE method,the proposed method can eliminate the influence of uncertain components of each order on the robustness of the system.2.Study on the swarm robust consensus tracking control problem of fractional leaders.The fractional leader's model is uncertain,so we use the properties of fractional differential equation and neural network estimation technology to estimate the uncertain components in the leader's dynamics;considering there exists the uncertainty of the follower agent's model itself.There are uncertain components both in the velocity-loop and the position-loop,and the corresponding control protocol is designed based on the neural network estimation technology;in addition,the consensus tracking problem under the unknown disturbances is considered,and the UDE technology is used to estimate and compensate the impact of the unknown disturbances on the robustness of the system.The related conclusions of integer-order system are promoted,and the consensus tracking control protocol for fractional-order leaders is designed.Finally,the control effects and applicability of the two methods are verified through simulation experiments.3.Study on the problem of consensus tracking control of the quadrotor swarm under the condition of disturbances.First,the six-degree-of-freedom dynamic model was established,and then based on the feedback linearization method,the nonlinear model was simplified to the double integral model to apply the theoretical research results in the previous sections in the consider of the channel being subjected to unknown disturbances.The robust consensus control method of the quadrotor swarm is studied when a certain channel is disturbed or certain channels are disturbed.UDE is added to the control model of the inner and outer ring of the quadrotor to compensate the impact of the unknown disturbances on the system.Finally,the controller of inner-loop is designed based on the PD control and the inverse module.