Abstract
The covariance estimation of dynamic system control models has applied both to estimator design and controller performance monitoring. Many algorithms has been proposed to estimate the unknown noise covariance of dynamic systems, such as maximum likelihood estimation (MLE), Bayesian estimation, covariance matching, correlation technique. The MLE method that maximizes likelihood estimation of the noise covariance matrix for the given observation sequence has a larger time overhead. This paper solves this problem by proposing MLE based on golden section. Each iteration of this algorithm will reduce the convergence interval to 0.618 times of the previous one. The length of convergence interval will be exponentially reduced. Simulation results show that the proposed algorithm has a more stable and faster convergence speed than the gradient-based MLE in both linear and nonlinear examples.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
