Increase the aviation efficiency of UAVs using artificial neural networks (original) (raw)
Related papers
Neural networks-based sensor validation for the flight control system of a B777 research model
2002
Shows the results of the analysis of a scheme for sensor failure, detection, identification and accommodation (SFDIA) using experimental flight data of a research aircraft model. Conventional approaches to the problem are based on observers and Kalman filters while more recent methods are based on neural approximators. The work described in the paper is based on the use of neural networks (NNs) as online learning nonlinear approximators. The performances of two different neural architectures are compared. The first architecture is based on a multi layer perceptron NN trained with the extended backpropagation algorithm. The second architecture is based on a radial basis function (RBF) NN trained with the extended minimal resource allocating networks (EMRAN) algorithms. The experimental data for this study are acquired from the flight-testing of a 1/24th semi-scale B777 research model designed, built, and flown at West Virginia University.
Extended Minimal Resource Allocating Neural Networks for Aircraft Sfdia
2001
This paper presents an Adaptive Neural Network (ANN) based tool for the modeling, simulation and analysis of aircraft Sensor Failure, Detection, Identification and Accommodation (SFDIA) problem. The tool is based on a SFDIA scheme in which learning NNs are used as on-line non-linear approximators of the analytically redundant portion of the system dynamics. This can provide validation capability to measurement
Sensitivity Analysis of a Neural Network based Avionic System by Simulated Fault and Noise Injection
2018
The application of virtual sensor is widely discussed in literature as a cost effective solution compared to classical physical architectures. RAMS (Reliability, Availability, Maintainability and Safety) performance of the entire avionic system seem to be greatly improved using analytical redundancy. However, commercial applications are still uncommon. A complete analysis of the behavior of these models must be conducted before implementing them as an effective alternative for aircraft sensors. In this paper, a virtual sensor based on neural network called Smart-ADAHRS (Smart Air Data, Attitude and Heading Reference System) is analyzed through simulation. The model simulates realistic input signals of typical inertial and air data MEMS (Micro Electro-Mechanical Systems) sensors. A procedure to define the background noise model is applied and two different cases are shown. The first considers only the sensor noise whereas the latter uses the same procedure with the operative flight n...
International Journal of Intelligent Unmanned Systems, 2021
Purpose This paper aims to design an adaptive nonlinear strategy capable of timely detection and reconstruction of faults in the attitude’s sensors of an autonomous aerial vehicle with greater accuracy concerning other conventional approaches in the literature. Design/methodology/approach The proposed scheme integrates a baseline nonlinear controller with an improved radial basis function neural network (IRBFNN) to detect different kinds of anomalies and failures that may occur in the attitude’s sensors of an autonomous aerial vehicle. An integral sliding mode concept is used as auto-tune weight update law in the IRBFNN instead of conventional weight update laws to optimize its learning capability without computational complexities. The simulations results and stability analysis validate the promising contributions of the suggested methodology over the other conventional approaches. Findings The performance of the proposed control algorithm is compared with the conventional radial basis ...
International Journal of Robust and Nonlinear Control, 2002
This paper focuses on the analysis of a scheme for sensor failure, detection, identification and accommodation (SFDIA) using experimental flight data of a research aircraft model. Recent technical literature has shown the advantages of time-varying estimators and/or approximators. Conventional approaches are based on different versions of observers and Kalman filters while more recent methods are based on different approximators based on neural networks (NNs). The approach proposed in the paper is based on the use of on-line learning nonlinear neural approximators.The characteristics of three different neural architectures were compared through different sensor failures. The first architecture is based on a multi layer perceptron (MLP) NN trained with the extended back propagation algorithm (EBPA). The second and third architectures are based on a radial basis function (RBF) NN trained with the minimal resource allocating network (MRAN) and extended-MRAN (EMRAN). The MRAN and EMRAN algorithms have recently been developed for RBF networks and have shown remarkable learning capabilities at a fraction of the memory requirements and computational effort typically associated with conventional RBF NNs.The experimental data for this study are flight data acquired from the flight-testing of a th semi-scale B777 research model designed, built, and flown at West Virginia University (WVU). Copyright © 2002 John Wiley & Sons, Ltd.
Mathematical Method of Artificial Neural Networks in Aircraft Maintenance, Repair and Overhaul
TEM Journal, 2020
Aircraft Maintenance, Repair and Overhaul (MRO) is one of the major components of the Aircraft Life Cycle Cost (LCC). Increasing the efficiency of MRO, as well as reducing MRO cost, is one of the main ways to reduce LCC. In modern aviation technology complexity of Avionics and its maintenance increase. Traditional methods of failure prediction are difficult to apply in complex technical systems which make it necessary to reduce MRO interval. This research proposed the mathematical method of Artificial Neural Networks (ANN) as a possible solution to this problem. The avionics of Unmanned Aerial Vehicle (UAV) is the research object. The reliability and forecasting of failures by traditional and ANN methods have been analyzed, and results comparison are received. The study suggests that the method used is suitable for solving this problem. The obtained results show a high degree of reliability. Further research is proposed to scale to more complex avionics aircraft. The introduction of...
Redundancy requirements for UAV (Unmanned Aerial Vehicle) are hardly faced due to the generally restricted amount of available space and allowable weight for the aircraft systems, limiting their exploitation. Essential equipment as the Air Data, Attitude and Heading Reference Systems (ADAHRS) require several external probes to measure significant data as the Angle of Attack or the Sideslip Angle. Previous research focused on the analysis of a patented technology named Smart-ADAHRS (Smart Air Data, Attitude and Heading Reference System) as an alternative method to obtain reliable and accurate estimates of the aerodynamic angles. This solution is based on an innovative sensor fusion algorithm implementing soft computing techniques and it allows to obtain a simplified inertial and air data system reducing external devices. In fact, only one external source of dynamic and static pressures is needed. This paper focuses on the benefits which would be gained by the implementation of this s...
Unmanned aerial vehicle (UAV) modelling based on supervised neural networks
Proceedings 2006 IEEE International Conference on Robotics and Automation, 2006. ICRA 2006.
This paper proposes the utilization of hybrid models of supervised neural networks for the modelling of dynamic systems. Particularly, as an example of a system, a autonomous helicopter or UAV is identified in both attitude and position systems. The evaluation of the model is done by comparing the radial basis and multilayer perceptron with the real system.
Flight Control System using Neural Networks
2021
Despite all the research being done in an attempt to bridge the gap between control systems and artificial intelligence, there is still an immense risk of failure and instability that exists. One particular application that this research will look into and expand on is aircraft control mechanisms. This paper will examine the existing uncertainties within these systems that could be suspected as the cause of failure in the artificial control operation of an aircraft. This study will act as a further extension of research on the feedback linearization of an aircraft’s control architecture using adaptive neural networks to decrease the probability of an uncontrolled error resulting from the nonlinearity of the aircraft’s dynamic characteristics. The stability of previously implemented mechanisms to control aircraft systems will also be investigated. This research will require a thorough approach and understanding of various possible areas of malfunction and instability caused by multip...
Method of the Intelligent System Construction of Automatic Control of Unmanned Aircraft Apparatus
Radio Electronics, Computer Science, Control, 2019
Context. Military conflicts of the late XX-early XXI centuries are characterized by the using of a large number of new weapons, which allowed the warring parties to distance themselves as far as possible from the direct collision with each other. Unmanned aircraft apparatus (UAA) have become one of the latest weapons on the battlefield, which during military conflicts were proven to be more effective than manned planes, in conducting air reconnaissance and other combat tasks, as well as strike at the enemy. One of the ways to increase the efficiency of UAA is to increase the level of technical excellence of their control systems. Creating new approaches for designing navigation systems for unmanned aerial vehicles particular, based on a free-form inertial navigation system, is an urgent task, as it will allow automatic control of the UAA flight route in the absence of corrective signals from the global satellite navigation system. Objective. The purpose of this work is to develop a methodology for managing an unmanned aerial apparatus using an intelligent automatic control system. This technique will minimize the error of a free inertial navigation system due to the using of a fuzzy neural network system. The algorithm of the proposed method of constructing the intellectual system of automatic control of UAA navigation system using the fuzzy neural network apparatus in the MatLab 7 software environment was developed. A neural network training was conducted in the Python 3.6 software environment (Jupyter-notebook), as well as testing the UAA model in the robot operational system (ROS) simulator environment for comparison with existing methods. Method. To achieve this goal, the following methods were used: intelligent systems, the theory of automatic control, pseudospectral method; methods based on the genetic algorithm and apparatus of the fuzzy neural network. Results. The method of constructing the intelligent system of automatic control of an unmanned aerial apparatus for minimizing the error of a free-form inertial navigation system due to the application of the neural network has been developed. The work of the intellectual system of automatic control of the UAA navigational system using the neural network in the MatLab software environment based on the proposed implementation algorithm were tested. The possibility of practical application of the obtained results and comparison with traditional methods were investigated.