Title :
Non-Identifier-Based Adaptive Tracking Control for a Two-Mass System
Author :
Schuster, Hans ; Westermaier, Christian ; Schröder, Dierk
Author_Institution :
Inst. of Electr. Drive Syst., Tech. Univ. Munich
Abstract :
High-gain-control is a well known control strategy which is non-identifier-based and needs no identification algorithm therefore. The outcome of this is a rather simple control concept which enables to control plants with uncertain values of the system parameters. The most severe disadvantage however is its nondecreasing gain function. This inhibits the use of high-gain-control especially if tracking control is required. For this reason, funnel-control is introduced which is an advancement of high-gain-control and contains a non-monotone time-varying gain function. Because of the proportional control law, a steady state error must be accepted as well as a tracking error. Although the error can be forced acceptable small if no sensor noise is present, this is not longer true in practical applications. To get an asymptotically vanishing control error, the funnel-controller is extended with a dynamic model, generating the reference signal, and with a dynamic feedback structure. Both extensions are introduced such that the closed loop stability is maintained. It is shown that a two-mass flexible servo system can be controlled with this strategy
Keywords :
adaptive control; closed loop systems; control system analysis; feedback; gain control; servomechanisms; stability; time-varying systems; tracking; asymptotically vanishing control error; closed loop stability; dynamic feedback structure; funnel-control; funnel-controller; high-gain-control strategy; nonidentifier-based adaptive tracking control; nonmonotone time-varying gain function; proportional control law; steady state error; tracking error; two-mass flexible servo system; Adaptive control; Control systems; Error correction; Feedback; Programmable control; Proportional control; Servomechanisms; Signal generators; Stability; Steady-state;
Conference_Titel :
Intelligent Engineering Systems, 2006. INES '06. Proceedings. International Conference on
Conference_Location :
London
Print_ISBN :
0-7803-9708-8
DOI :
10.1109/INES.2006.1689367