On a learning algorithm of feedback control for an aircraft in windshear: a case study

摘要

An algorithm for learning feedback control is presented. Based on the gradient method, a set of formulas for online updating of the weights of a neural controller is given. To demonstrate the algorithm, the proposed controller is used in a practical problem for an aircraft encountering windshear during take-off. The proposed controller is shown to be feasible for many nonlinear systems, and seems to be robust in the presence of windshear.