This paper proposes three novel neural network controllers for dual adaptive control of a class of functionally uncertain, nonlinear, multiple-input/multiple-output stochastic systems. Both Gaussian radial basis function and sigmoidal multilayer perceptron neural networks are considered for approximation of the unknown dynamic functions. Control and estimation are effected through optimization of a stochastic cost function which elicits the dual effects of caution and probing, resulting in control laws that take into consideration the interactions between estimation and control, leading to improved control performance. The Kalman filter is used for estimation of the weights of the radial basis function network, while the extended and unscented Kalman filter are used for the multilayer perceptron case. The performances of the three schemes are compared and evaluated through extensive Monte Carlo simulations and statistical significance tests.
Bar-ShalomY (1981) Stochastic dynamic programming: Caution and probing. IEEE Transactions on Automatic Control26(5): 1184–1195.
3.
BugejaMKFabriSG (2007) Dual adaptive control for trajectory tracking of mobile robots. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA 07), April, Rome, Italy, pp. 2215–2220.
4.
BugejaMKFabriSGCamilleriL (2009) Dual adaptive control of mobile robots using neural networks. IEEE Transactions on Systems, Man and Cybernetics - Part B39(1): 129–141.
5.
FabriSGBugejaMK (2013a) Kalman Filter-based Estimators for Dual Adaptive Neural Control: A comparative analysis of execution time and performance issues. In Proceedings of the 10th International Conference on Informatics in Control, Automation and Robotics (ICINCO), July, Reykjavik, Iceland, pp. 169–176.
6.
FabriSGBugejaMK (2013b) Unscented transform-based dual adaptive control of nonlinear MIMO systems. In Proceedings of the European Control Conference 2013, July. Zurich, Switzerland: EUCA, pp. 392–397.
7.
FabriSKadirkamanathanV (1996) Dynamic structure neural networks for stable adaptive control of nonlinear systems. IEEE Transactions on Neural Networks7(5): 1151–1167.
8.
FabriSKadirkamanathanV (1998) Dual adaptive control of nonlinear stochastic systems using neural networks. Automatica34(2): 245–253.
9.
FabriSGKadirkamanathanV. Functional Adaptive Control: An Intelligent Systems Approach. London: Springer-Verlag2001.
10.
Fel’dbaumAA (1960) Dual control theory I–II. Automation and Remote Control21: 874–880, 1033–1039.
11.
Fel’dbaumAA (1961) Dual control theory III–IV. Automation and Remote Control22: 1–12, 109–121.
12.
Fel’dbaumAA (1965) Optimal Control Systems. New York: Academic Press.
FilatovNUnbehauenH (2004) Adaptive Dual Control: Theory and applications. Berlin: Springer.
15.
HaykinS (1999) Neural Networks: A Comprehensive Foundation, 2nd edn.Upper Saddle River, NJ: Prentice-Hall.
16.
HaykinS (ed.) (2001) Kalman Filtering and Neural Networks. New York: John Wiley and Sons.
17.
HerzallahRLoweD (2008) A Bayesian perspective on stochastic neurocontrol. IEEE Transactions on Neural Networks19(5): 914–924.
18.
JagannathanSLewisFL (1996) Multilayer discrete-time Neural-Net controller with guaranteed performance. IEEE Transactions on Neural Networks7(1): 107–130.
19.
JulierSJUhlmannJK (2004) Unscented filtering and nonlinear estimation. Proceedings of the IEEE92(3): 401–422.
20.
KrálLŠimandlM (2008) Functional adaptive control for multi-input multi-output systems. In Proceedings of the 17th IFAC World Congress, Vol. 17. Seoul, South Korea: IFAC, pp. 11263–11268.
21.
MilitoRPadillaCSPadillaRACadorinD (1982) An innovations approach to dual control. IEEE Transactions on Automatic Control27(1): 133–137.
22.
Murray-SmithRSbarbaroD (2002) Nonlinear adaptive control using non-parametric Gaussian Process prior models. In 15th Triennial World Congress of the International Federation of Automatic Control, July, Barcelona, Spain.
23.
RadenkovićMS (1988) Convergence of the generalized dual control algorithm. International Journal of Control47(5): 1419–1441.
24.
SannerRMSlotineJJE (1992) Gaussian networks for direct adaptive control. IEEE Transactions on Neural Networks3(6): 837–863.
25.
SbarbaroDFilatovNMUnbehauenH (1998) Adaptive dual controller for a class of nonlinear systems. In Processings of the IFAC Workshop on Adaptive Systems in Control and Signal Processing, August, Glasgow, Scotland, pp. 28–33.
26.
SpoonerJTPassinoKM (1996) Stable adaptive control using fuzzy systems and neural networks. IEEE Transactions on Fuzzy Systems4(3): 339–359.
27.
ŠimandlMDuníkJ (2005) Sigma point Gaussian sum filter design using square root Unscented filters. In Preprints of the 16th IFAC World Congress, July, Prague, Czech Republic.
28.
WanEAvan der MerweR (2001) The Unscented Kalman Filter. In HaykinS (ed.), Kalman Filtering and Neural Networks (Adaptive and Learning Systems for Signal Processing, Communications and Control). New York: John Wiley and Sons, pp. 221–280.
29.
WangLX (1994) Adaptive Fuzzy Systems and Control: Design and Stability Analysis. Englewood Cliffs, NJ: Prentice-Hall.
