Mechanical Engineering Commons^™

Online Identification And Control Of Aerospace Vehicles Using Recurrent Networks, Zhenning Hu, S. N. Balakrishnan

Mechanical and Aerospace Engineering Faculty Research & Creative Works

Methods for estimating the aerospace system parameters and controlling them through two neural networks are presented in this study. We equate the energy function of Hopfield neural network to integral square of errors in the system dynamics and extract the parameters of a system. Parameter convergence is proved. For control, we equate the equilibrium status of a "modified" Hopfield neural network to the steady state Riccati solution with the system parameters as inputs. Through these two networks, we present the online identification and control of an aircraft using its nonlinear dynamics.