State Estimation Algorithm for Nonlinear System with Unknown Parameter Based on Minimum Perturbation Principle

For the state estimation problem of nonlinear systems with unknown parameters, based on the principle of minimum perturbation, the state equation with unknown parameters and both the state equation and measurement equation with unknown parameters are considered, respectively. By converting to a constrained optimization problem, and using the alternate estimation of the system state and parameters, we design the filters in two cases, and give the calculation form when the Jacobian matrix of the system is singular. In order to prove the effectiveness of the proposed algorithm, a nonlinear model with unknown parameters is used, and the simulation results show the superiority of the proposed algorithm.