Autonomous Helicopter Hover Using an Artificial Neural Network
نویسندگان
چکیده
This paper details the developments to date of an unmanned air vehicle (UAV) based on a standard size 60 model helicopter. The design goal is to have the helicopter achieve stable hover with the aid of an INS and stereo vision. The focus of this paper is on the development of an artificial neural network (ANN) that makes use of only the INS data to generate hover commands, which are used to directly manipulate the flight servos. Current results show that networks incorporating some form of recurrency (state history) offer little advantage over those without. At this stage, the ANN has partially maintained periods of hover even with misaligned sensors.
منابع مشابه
Using an Artificial Neural Network
This paper details the developments to date of an unmanned air vehicle (UAV) based on a standard size 60 model helicopter. The design goal is to have the helicopter achieve stable hover with the aid of an INS and stereo vision. The focus of this paper is on the development of an artificial neural network (ANN) that makes use of only the INS data to generate hover commands, which are used to dir...
متن کاملFlight Control Using an Artificial Neural Network
This paper details the developments to date of an unmanned air vehicle (UAV) based on a standard size 60 model helicopter. The design goal is to have the helicopter achieve stable hover with the aid of an INS and monocular vision. The focus of this paper is on the development of an artificial neural network that makes use of this data to generate hover command, which are used to directly manipu...
متن کاملHelicopter Rotor Airloads Prediction Using CFD and Flight Test Measurement in Hover Flight
An implicit unsteady upwind solver including a mesh motion approach was applied to simulate a helicopter including body, main rotor and tail rotor in hover flight. The discretization was based on a second order finite volume approach with fluxes given by the Roeand#39;s scheme. Discretization of Geometric Conservation Laws (GCL) was devised in such a way that the three-dimensional flows on arbi...
متن کاملAdaptive Output Feedback Control of a Scale Helicopter Restricted to a 2dof Platform
Abstract: In this paper we present the application of the adaptive neural-networktechnique in (Calise, Hovakimyan and Idam, 2001) to the control of the 3DoF model helicopter in (Avila, Brogliato, Dzul and Lozano, 2003). The technique uses feedback linearization, linear (coarse) compensation and adaptive-neural-network (fine) compensation. The description of the technique is presented in a way a...
متن کاملReal-Time Scale Invariant Object Recognition Using an Artificial Neural Network
In order to satisfy a need of real-time object identification a scale invariant feature matching method using an artificial neural network (ANN) is proposed. Scale Invariant Feature Transforms (SIFT) feature detection and feature description is compared with the SIFT feature detection and a custom feature descriptor matched with an appropriately trained ANN. The proposed method requires less ti...
متن کامل