In control and signal processing, adaptation is a natural tool to cope with real-time changes in the dynamical behaviour of signals and systems. In this area, strongly connected with prediction and identification, there has been an increasing interest in switching and supervising methods. Moreover in recent years, special attention has been paid to the ideas evolving round the theory of statistical learning as a potential tool of improved adaptation. The IFAC workshop on Adaptation and Learning in Control and Signal Processing in 2001 gathered together experts in the field and interested researchers from universities and industry to present a full picture of the area. This proceedings volume presents papers covering the following subjects: Model reference and predictive control; Multiple model control; Adaptive control I/II; Adaptive control and learning; Learning; Adaptive control of nonlinear systems I/II; Supervisory control; Neural networks for control; PID design methods; Sliding mode; Adaptive filtering and estimation; Identification methods I/II.Fig. 22. Asynchronous motor coupled to the combustion engine 9. CONCLUSIONS The demands for improved engine emissions, performance and efficiency require the application of advanced ... Fast neural networks with local linear models can approximate multidimensional problems within small training times of mostly less than a minute. ... Die Abgasreini- gung der FSI-Motoren von Volkswagen.
|Title||:||Adaptation and Learning in Control and Signal Processing 2001|
|Publisher||:||Pergamon - 2002|