Study Of Filtering Algorithm For A Class Of Systems With Equality State Constraints And Multiplicative Noise

The research of systems with multiplicative noise has largely promoted the theoryof data processing. With the accuracy of modeling in the field of dynamic positioningdata processing and underwater target tracking estimation theory of systems withmultiplicative noise is increasingly showing its superiority. With the swift developmentof measurement technology, we can take advantage of these priori observed informationto establish equality constraints and effectively combine with the theory of Kalman filterto improve the accuracy of the data processing. So the research of filtering algorithm forsystem with multiplicative noise and constraints is full of theoretical and practicalsignificance.The main focus of this paper is on resolving the problem of filtering for systemswith multiplicative noise and constraints. This paper presents two algorithms based ondifferent criterion for systems with multiplicative noise and constraints. At the same time,this paper proposes algorithms for nonlinear equality constraints. This paper alsoproposes filtering algorithms for nonlinear system with multiplicative noise andconstraints. This paper presents a lot of simulation and comparative experiments toverify the effectiveness of the algorithm. Here are the main works of this paper:1. Filtering algorithms for systems with equality state constraints and multiplicativenoise are proposed in this paper. First, the form of observation vector is changed. Thenbased on the projection theorem guidelines the algorithm is solved. In contrast withalgorithms are calculated by mathematical optimization theory, the algorithm proposedin this paper are revealed the validation. Computer simulation results are presented toillustrate the algorithm.2. This paper proposes filtering algorithm for a class of equality constrainedsystems with multiplicative noise based on mathematical optimization method. Thealgorithm is based on optimization theory. Combine updated variables withunconstrained variables to establish filtering algorithm with updated variables. Thisalgorithm takes updated variables into account. So it improves the algorithm accuracy. Itis verified by simulation and comparison.3. This paper proposes a filtering algorithm for system with multiplicative noise andnonlinear equality constraints. It is verified by a lot of simulation and comparison. Fornonlinear system, this paper discusses several feasible methods and filtering algorithmwith equality constraints.