Accounting for Performance Limitations Inside the Model Following Control System of an Unmanned Helicopter

Abstract

In order to prevent generating reference signals that cannot be followed by an unmanned helicopter, a method which considers thrust limitations inside a reference model control scheme was developed and is presented in this paper. Contrary to methods that consider only small accelerations around hover flight, the proposed method is designed to permit highly agile maneuvers. The focus of the paper is on the description and the explanation of the proposed limiting method. Each component of the method is thus detailed and justied. The achieved results, benets and constraints of this method are discussed and simulation test scenarios were specically designed to throw the behaviors resulting from the proposed method into relief. It is shown that limitation of the command values inside the reference model in an open-loop control architecture is possible and leads to very small errors on the real plant compared to other approaches even though this approach is signicantly less conservative. Flight tests were carried out on DLR's unmanned helicopter ARTIS. They illustrate the applicability of the proposed method and conrm its potential for helicopter envelope protection enhancement when using a reference model based control architecture.