Application of Maximum Likelihood Identification with Multisensor Fusion to Time-varying Stochastic System

Abstract

The maximum likelihood parameter identification of time-varying sensor measurement system is considered. In order to identity the basic underlying system models, maximum likelihood functional analysis is suggested. The problem is then formulated as an Optimization problem. The data fusion problem is also studied and it is shown that the confidence of produced estimates could be improved by combining several similar estimates. An illustrative robot sensing example is given to demonstrate the effectiveness of proposed algorithm.