Research On Adaptive Neural Network Tracking Control For Several Nonlinear Time-delay Systems

In real social life,most systems have uncertain nonlinear characteristics.Uncertain non-linear systems are much more complicated than linear systems and general non-linear systems,because they may contain uncertain interference terms or other terms that cannot be detected and controlled at present.The output tracking problem of the system is an important research topic in the comprehensive problem of control theory.The nonlinearity of the actual system and the inevitable uncertainty factors have a great impact on the output tracking of the system.Thence,adaptive neural network tracking control for uncertain nonlinear systems has become a research hotspot.Adaptive neural network tracking control methods have been applied in many practical aspects,such as autonomous formation flight and F-16 adaptive pitch rate tracker.Neural networks have powerful function approximation capabilities which can be used to approximate unknown functions,and it effectively handles uncertain nonlinear systems with unknown functions through combining with adaptive backstepping control technology.Uncertain factors such as unmodeled dynamics,time delay,output interference,and state constraints often appear in actual systems.Therefore,it is becoming more and more important that the adaptive neural network tracking control for uncertain nonlinear systems with unmodeled dynamics,output disturbances,state constraints and time delays.Based on this,the following aspects are mainly studied in this paper.First,for a class of uncertain nonlinear systems with output disturbance and unknown time delay,an adaptive neural network tracking control method based on backstepping technology is proposed.Neural network approximation is introduced as a very effective estimation technique to approximate unknown functions.Appropriate Lyapunov-Krasovskii function is constructed,and the unknown time delay is compensated by the organic combination of Young's inequality.The Nussbaum function is used to handle unknown virtual control directions.A practical robust control method is proposed to deal with controller singularity.This method does not require prior knowledge.In this way,all signals reach the semi-globally uniform ultimate boundedness(SGUUB),and it is proved that the tracking error will eventually converge to the area around the origin.Simulation results verify the feasibility and effectiveness of the method.Secondly,in practically complex systems,the impact of unmodeled dynamics on system stability cannot be ignored.Therefore,based on previous research,further research is performed on a class of uncertain nonlinear time-delay systems with unmodeled dynamics and output disturbances.An adaptive tracking control strategy based on neural network is proposed.To mitigate the main effects caused by output disturbances and unknown time delays,appropriate Lyapunov-Krasovskii functions and backstepping techniques are used.To obtain the required adaptive state feedback controller,assumptions related to dynamics signals are used to handle unmodeled dynamics.By designing adaptive methods,unmodeled dynamics,output disturbances and unknown time-varying virtual coefficients can be handled.During the design process,neural networks are used to approximate unknown functions.In the end,all signals achieved semi-global uniform ultimate boundedness,and verified by simulation.Finally,based on the previous studies that considered uncertain nonlinear systems with uncertainties such as time delay and unmodeled dynamics,the state constraints were further considered,which is very common in practical systems,such as autonomous formation flight,ship operation systems,etc.Therefore,for a nonlinear state-constrained time-varying delay system with unmodeled dynamics,an adaptive neural network tracking control method is studied.The hyperbolic tangent function can effectively deal with the effects of unmodeled dynamics.At the same time,using the Lyapunov Krasovskii function and backstepping technology,some effective methods have been designed to effectively deal with time-varying delays.In addition,based on the barrier Lyapunov function(BLF),the state never violates the complete state constraint.In the end,all signals reach semi-global uniform ultimate boundedness,and the tracking error finally converges near the origin.Simulation results verify the effectiveness of the control method.