Showing papers by "Robert F. Stengel published in 2005"

Smooth function approximation using neural networks

Abstract: An algebraic approach for representing multidimensional nonlinear functions by feedforward neural networks is presented. In this paper, the approach is implemented for the approximation of smooth batch data containing the function's input, output, and possibly, gradient information. The training set is associated to the network adjustable parameters by nonlinear weight equations. The cascade structure of these equations reveals that they can be treated as sets of linear systems. Hence, the training process and the network approximation properties can be investigated via linear algebra. Four algorithms are developed to achieve exact or approximate matching of input-output and/or gradient-based training sets. Their application to the design of forward and feedback neurocontrollers shows that algebraic training is characterized by faster execution speeds and better generalization properties than contemporary optimization techniques.

Robust nonlinear flight control of a high-performance aircraft

Abstract: This paper considers probabilistic robust control of nonlinear uncertain systems. A combination of stochastic robustness and dynamic inversion is proposed for general systems that have a feedback-linearizable nominal system. In this paper, the stochastic robust nonlinear control approach is applied to a highly nonlinear complex aircraft model, the high-incidence research model (HIRM). The model addresses a high-angle-of-attack enhanced manual control problem. The aim of the flight control system is to give good handling qualities across the specified flight envelope without the use of gain scheduling and also to provide robustness to modeling uncertainties. The proposed stochastic robust nonlinear control explores the direct design of nonlinear flight control logic. Therefore, the final design accounts for all significant nonlinearities in the aircraft's high-fidelity simulation model. The controller parameters are designed to minimize the probability of violating design specifications, which provides the design with good robustness in stability and performance subject to modeling uncertainties. The present design compares favorably with earlier controllers that were generated for a benchmark design competition.