Abstract
For continuous probability density functions (PDFs) approximated by the B-spline basis functions, the relationships between the B-spline weights, the entropy and the mean have been analyzed in detail. It shows the different characteristics of the entropy with and without mean constraint. A minimum entropy controller subjected to mean constraint is developed by taking the performance function as a Lyapunov function and ensuring the negativeness of its first-order derivative. Simulation examples are included to validate the analysis results and evaluate the closed-loop control performance. Copyright © 2005 IFAC.
| Original language | English |
|---|---|
| Title of host publication | IFAC Proceedings Volumes (IFAC-PapersOnline)|IFAC Proc. Vol. (IFAC-PapersOnline) |
| Pages | 93-98 |
| Number of pages | 5 |
| Volume | 16 |
| Publication status | Published - 2005 |
| Event | 16th Triennial World Congress of International Federation of Automatic Control, IFAC 2005 - Prague Duration: 1 Jul 2005 → … |
Conference
| Conference | 16th Triennial World Congress of International Federation of Automatic Control, IFAC 2005 |
|---|---|
| City | Prague |
| Period | 1/07/05 → … |
Keywords
- B-spline neural network
- Dynamic stochastic systems
- Mean constraint
- Minimum entropy
- Probability density function (PDF)