The minimal time change detection algorithm for intelligent control system

摘要

System parameters should be tracked online to build an intelligent control systems even though there exists an abrupt change. For this purpose, a new performance index that we are studying is the speed of adaptation-how quickly does the system determine that a change has occurred? In this paper, a new, robust algorithm that is optimized to minimize the time delay in detecting a change for fixed false alarm probability is proposed. Simulation results for the aircraft lateral motion with a known or unknown change in control gain matrices, in the presence of doublet input, indicate that the algorithm works fairly well. One of its distinguishing properties is that detection delay of this algorithm is superior to that of Whiteness Test.