In long-range predictive control schemes, multi-step ahead predicted values can be arrived at by concatenating the unit step predictions. In doing this, the missing outputs are replaced by their previously predicted values. This approach is much simpler than the approach commonly adopted in the literature which is based on the solution of an identity. The paper concludes with the formulation of the scheme for the trajectory control of a servo-system.
CutlerC. R.RamakerB. C.Dynamic matrix control: A computer control algorithm. Proceedings of Joint Automatic Control Conference, 1980, San Francisco, paper WP5-B.
4.
De KeyserR. M. C.Van CauwenbergheA.A self-tuning multi-step predictor application. Automatica, 1979, 17, 167–174.
5.
ClarkeD. W.MohtadiC.TuffsP. S.Generalized predictive control: Parts I and II. Automatica, 1987, 23, 137–160.
KaynakO.An overview of predictive approaches for advanced motion control (invited paper). Proceedings of 2nd IEEE International Workshop on Advanced motion control, Nagoya, 1992, pp. 26–33.
8.
KaynakO.HashimotoH.KuroyanagiH.HarashimaF.Robust tracking control based on long range prediction. Proceedings of 16th Annual Conference of IEEE Industrial Electronics Society, IECON'90, San Francisco, 1989, pp. 296–300.
