Challenges of adaptive control–past.pdfVIP

Annual Reviews in Control 32 2008 123–135 /locate/arcontrol Challenges of adaptive control–past, permanent and future§ Brian D.O. Anderson a,b, Arvin Dehghani a,* a Department of Information Engineering, Research School of Information Sciences and Engineering, The Australian National University, Canberra, ACT 0200, Australia b National ICT Australia Ltd., Locked Bag 8001, Canberra, ACT 2601, Australia Received 1 November 2007; accepted 5 June 2008 Abstract This paper reviews three different types of challenges to adaptive control. The ?rst group comprises challenges met in the subject’s development. They include dif?culties associated with the MIT rule, bursting, the Rohr’s counterexample and unplanned instability in iterative identi?cation and control. An understanding of these phenomena and mitigating strategies are now available. The second group comprises dif?culties that are intrinsic to virtually any adaptive control algorithm, and that have frequently been overlooked. For example, if a plant is unknown, and a control objective is set, the objective may in practical terms be unachievable, and any adaptive control algorithm needs to deal with that possibility. The third group comprises some issues to which researchers are currently devoting signi?cant attention, including multiple model adaptive control and model-free design. # 2008 Elsevier Ltd. All rights reserved. Keywords: Adaptive control; Adaptive systems; Multiple model adaptive control; Model-free design 1. Introduction speci?c dif?culties became apparent only sometime after generation of the initial ideas. In this group, we include the Adaptive control has been with us for some decades. Many MIT rule, bursting, the Rohr’s counterexample, and iterative books, for example Anderson et al., 1986; Astrom control and identi?cation. The second set of problems we Wittenmark, 1989; Goodwin Sin, 1984; Narendra describe