Research On Adaptive Control For Multi-Agent Constraint Systems

Multi-agent systems are systems that involve multiple subsystems interacting and collaborating with each other to complete complex tasks.They are widely used in modern industrial,military,and agricultural fields,such as intelligent robots,unmanned aerial vehicles(UAVs),and power grid systems,attracting the attention of numerous scholars.In particular,multi-agent collaborative tracking control is one of the most typical hot research issues in this field.Therefore,this paper studies the theory of cooperative tracking control to solve the problems of security constraints,anti-interference and improving efficiency of multi-agent system.It is pointed out that the key technical difficulty is how to apply collaborative tracking control theory to ensure the safe completion of established tasks in plant protection drones.It is caused by the large operation area of plant protection UAVs,which will be affected by external disturbances such as wind gusts and strong light during the operation.The main contents of this paper are as follows.(1)An adaptive fuzzy output feedback control method is proposed for nonlinear multi-agent systems with unmeasurable states.Due to the immeasurability of system states,high-performance state observers are designed to accurately estimate unknown states.Fuzzy logic systems and adaptive backstepping control methods are used to design adaptive laws,and integral barrier Lyapunov functions are constructed to analyze the stability of the closed-loop system.On the one hand,it breaks through the conservative limitation of error boundaries,and on the other hand,it ensures the performance of constraint tracking control.Simulation examples verify expected tracking performance.(2)An adaptive neural network distributed tracking control strategy is proposed for multi-agent systems subject to time-varying full state constraints.The boundary of asymmetric state constraints is related to the state vector and time,which further enhances the flexibility of the constraint boundary,but increases the difficulty of controller design.Using the approximation characteristics of neural networks,unknown functions are modeled,asymmetric barrier Lyapunov functions are constructed,and adaptive distributed controllers are designed based on the backstepping method,ensuring the global boundedness of the closed-loop system signals.A simulation example verifies the effectiveness of the proposed method.(3)The research for the impact of external disturbances on plant protection drones during operation based on the preset performance formation control scheme for multiple plant protection drones with interference observers.The mathematical model of UAVs is established by calculating the conversion relationship between the spatial coordinate system machine and the volumetric coordinate system.Designing an disturbance observer based on fixed time theory achieves accurate estimation of disturbances,ensuring the convergence of the interference observer within the fixed time.Introducing a new prescribed performance index function into the joint operation scheme of multiple plant protection UAVs formation can ensure the quality of operation while improving the efficiency of collaborative operation to meet the needs of large-scale production.Simulation example illustrates the feasibility of the scheme.In this paper,for multi-agent systems with measurable/unmeasurable states,the state observer and the disturbance observer are constructed.Combining the theory of integral barrier Lyapunov function,asymmetric barrier Lyapunov function,and prescribed performance index function,the design of constraint tracking controller for multi-agent systems is completed.The research can be applied to multiple plant protection drone systems.