Research On Filtering Problems Of State Constrained Systems:An Event-Triggered Scheme

With the rapid development of network technology,the study of the theory related to state constrained systems has become a hot issue,and the demand for its performance in the practical engineering field is increasing rapidly.However,the state,output,input and stability of the system are often affected by the imperfect equipment,the immaturity of the environment and the technology,which makes it difficult to meet the actual production requirements.Compared with the defects of traditional control theories in this respect,the control theory of state constrained systems have higher research value.In addition,in the field of practical engineering,it is often necessary to know the state information of the system for specific purposes such as optimization and approximation.However,due to the constraints imposed on the system state by real situations,we often do not have direct access to the relevant state information within the system.In this case,an effective estimation of the system itself based on the measured data is necessary.Considering this situation,this paper will take a specific relevant system as the research object,and start from the relevant requirements,conduct an in-depth study in the system state constrained estimation and specific constrained situations,so this paper considers the filtering problem of the state constrained system based on the event-triggered scheme.In this dissertation,we firstly analyze the characteristics of state constrained systems in depth,and takes typical state constrained systems as the research object,and establish a specific state constrained systems(discrete time-varying parameter uncertainty systems under equality constraints),then analyzes the state estimation problem of state constrained systems.On the other hand,in order to save resources,we adopted eventtriggered mechanism.The filter is analyzed by using the state information after triggering,which makes the result more concrete and reliable.Subsequently,based on the previous theory,the filtering problem of stateconstrained systems with outliers is studied by using the recursive idea.Similarly,an event-triggered mechanism is introduced and an in-depth analysis in terms of system control and filter performance is carried out.Synthesizing the above discussed scenarios,the main theoretical results of this paper are obtained as follows:1.For the discrete time-varying parameter uncertainty system,we consider the state constraint in the parameter uncertainty system,and study the state estimation problem of the system with parameter uncertainty based on the event-triggered mechanism.By constructing constrained resilient filters,different from the traditional Lyapunov method,a simple recursive method is used to obtain sufficient conditions for the upper bound matrix of the estimation error covariance.By considering the situation that the system state is constrained,it can be more consistent with engineering practice and can effectively reduce the conservativeness of the system analysis.2.In this dissertation,the filtering problem of discrete time-varying systems with abnormal measurements under event triggered-mechanism is studied for state constrained systems.By adopting a new saturation function,the measured outliers are treated effectively,and the adaptive threshold of saturation level is adopted to deal with the influence of outliers,a more conservative conclusion can be obtained.At the same time,the sensitivity of the filter to outliers is significantly reduced.3.In view of the discrete time-varying system in the actual operation process,it is often constrained by equality,the performance of the filter could not achieve satisfactory results.Based on this,we construct an equality constraints filter,so that it can solve the problem of inaccurate estimation.Considering the discrete time-varying system under the eventtriggered mechanism,we establish a filter model that can meet the needs of actual production.Finally,the simulation algorithm verifies that the constrained state estimation can accurately reflect the actual state of the system.