A simple method to improve the transients in adaptive systems

A method, based on classical optimization techniques, is proposed to improve the adaptation transients in adaptive systems. Its philosophy consists basically in using a modified adaptation mechanism which alternately works in a priori mode, in which the optimization-correction horizon is slid forward in time, and in a posteriori mode, in which, by using an equivalent system, the parameters (and the adaptation gain matrix, if time varying) are recomputed over this horizon to generate the adaptive plant input.