Search

This paper discusses the novel anti-disturbance control algorithm for hypersonic flight vehicle (HFV) models by using neural network (NN) identifier. Different from those existed anti-disturbance results, the unknown exogenous disturbances in HFV models are assumed to be described by the designed NNs with adjustable parameters. Furthermore, the disturbance-observer-based-control (DOBC) algorithm with adaptive regulation laws is thus presented to estimate the nonlinear disturbances. By integrating the estimated value of disturbances with the PI feedback control input, a composite controller based on convex optimization theory is generated to ensure the satisfactory stability and dynamical tacking convergence of HFV models. Finally, a numerical example for HFV models is included to illustrate the effectiveness of the theoretical results.

Multiple Structuring Element Neural Learning Via Cellular LogicLan Shao, Liren Liu, Guoqiang Li and Xuejun ZhangShanghai Institute of Optics and Fine Mechanics, Academia SinicaP.O.Box 800-211, Shanghai 201800, P.R.ChinaABSTRACTAn idea for the construction of morphological structuring elements byneural learning from multiple targets is proposed. Accordingly, neuralerosion is developed. Thus multi-character recognition and multi-to-onesymbolic substitution can be easily realized. Optical implementationbased on incoherent optical correlation is illustrated.