Search results

The purpose of this paper is to introduce an artificial bee colony-based Kalman filter algorithm along with an extended objective function to ensure the optimality of the estimator of the quadrotor in the presence of unknown measurement noise statistics.

Six degree-of-freedom mathematical model of the quadrotor is derived. Position controller for the quadrotor is designed. Kalman filter-based estimation algorithm is implemented in the sensor feedback loop. Artificial bee colony-based hybrid algorithm is used as an optimization method to handle the unknown noise statistics. Existing objective function is extended with a penalty term. Mathematical proof of the extended objective function is derived. Results of the proposed algorithm is compared with de facto genetic algorithm-based Kalman filter.

Artificial bee colony algorithm-based Kalman filter and extended objective function duo are able to optimize the measurement noise covariance matrix with an absolute error as low as 0.001 [m²]. Proposed method and function is capable of reducing the noise from 2 to 0.09 [m] for x-axis, 3.4 to 0.14 [m] for y-axis and 3.7 to 0.2 [m] for z-axis, respectively.

The motivation behind this paper is to bring a novel optimization-based solution for the estimation problem of the quadrotor when the measurement noise statistics are unknown along with an extended objective function to prevent the infeasible solutions with mathematical convergence analysis.

The purpose of this paper is to present fault tolerant control of a quadrotor based on the enhanced proportional integral derivative (PID) structure in the presence of one or more actuator faults.

Mathematical model of the quadrotor is derived by parameter identification of the system for the simulation of the UAV dynamics and flight control in MATLAB/Simulink. An improved PID structure is used to provide the stability of the nonlinear quadcopter system both for attitude and path control of the system. The results of the healty system and the faulty system are given in simulations, together with motor dynamics.

In this study, actuator faults are considered to show that a robust controller design handles the loss of effectiveness in motors up to some extent. For the loss of control effectiveness of 20 per cent in first and third motors, psi state follows the reference with steady state error, and it does not go unstable. Motor 1 and Motor 3 respond to given motor fault quickly. When it comes to one actuator fault, steady state errors remain in some states, but the system does not become unstable.

In this paper, an enhanced PID controller is proposed to keep the quadrotor stable in case of actuator faults. Proposed method demonstrates the effectiveness of the control system against motor faults.

An aircraft engine control system consists of a large scale of control parameters and variables because of the complex structure of aero-engine. Monitoring and adjusting control variables and parameters such as detecting, isolating and reconfiguring the system faults/failures depend on the controller design. Developing a robust controller is based on an accurate mathematical model.

In this study, a small-scale turboprop engine is modeled. Simulation is carried out on MATLAB/Simulink for design and off-design operating conditions. Both steady-state and transient conditions (from idle to maximum thrust levels) are tested. The performance parameters of compressor and turbine components are predicted via trained Neuro-Fuzzy model (ANFIS) based on component maps. Temperature, rotational speed, mass flow, pressure and other parameters are generated by using thermodynamic formulas and conservation laws. Considering these calculated values, error calculations are made and compared with the cycle data of the engine at the related simulation conditions.

Simulation results show that the designed engine model’s simulation values have acceptable accuracy for both design and off-design conditions from idle to maximum power operating envelope considering cycle data. The designed engine model can be adapted to other types of gas turbine engines.

Different from other literature studies, in this work, a small-scale turboprop engine is modeled. Furthermore, for performance prediction of compressor and turbine components, ANFIS structure is applied.

This study aims to present a method for the conceptual design and simulation of an aircraft flight control system.

The design methodology is based on particle swarm optimization (PSO). PSO can be used to improve the performance of conventional controllers. The aim of the present study is threefold. First, it attempts to detect and isolate faults in an aircraft model. Second, it is to design a proportional (P) controller, a proportional derivative (PD) controller, a proportional-integral (PI) controller and a fuzzy controller for an aircraft model. Third, it is to design a PD controller for an aircraft using a PSO algorithm.

Conventional controllers, an intelligent controller and a PD controller-based PSO were investigated for flight control. It was seen that the P controller, the PI controller and the PD controller-based PSO caused overshoot. These overshoots were 18.5, 87.7 and 2.6 per cent, respectively. Overshoot was not seen using the PD controller or fuzzy controller. Steady state errors were almost zero for all controllers. The PD controller had the best settling time. The fuzzy controller was second best. The PD controller-based PSO was the third best, but the result was close to the others.

This study shows the implementation of the present algorithm for a specified space mission and also for study regarding variation of performance parameters. This study shows fault detection and isolation procedures and also controller gain choice for a flight control system. A comparison between conventional controllers and PD-based PSO controllers is presented. In this study, sensor fault detection and isolation are carried out, and, also, root locus, time domain analysis and Routh–Hurwitz methods are used to find the conventional controller gains which differ from other studies. A fuzzy controller is created by the trial and error method. Integral of squared time multiplied by squared error is used as a performance function type in PSO.

Condition monitoring and health management of an aircraft engine is of importance due to engine’s critical position in aircraft. Missions require uninterrupted and safer conditions during the flight or taxi operations. Hence, the deviations, abnormal situations or failures have to be under control. This paper aims to propose a cascade connected approach for an aircraft engine fault tolerant control.

The cascade connected structure includes a full-order unknown input observer for fault detection and eliminating the unknown disturbance effect on system, a generalized observer scheme for fault isolation and a Boolean logic mechanism for decision-making in reconfiguration process, respectively. This combination is simulated on a linear turbojet engine model in case of unknown input disturbance and under various sensor failure scenarios.

The simulation results show that the suggested fault detection isolation reconfiguration (FDIR) approach works effectively for multiple sensor failures with various amplitudes.

Different from other studies, the proposed model is sensitive to unknown input disturbance and failures that have unknown amplitudes. One another notable feature of suggested FDIR approach is adaptability of structure against multiple sensor failures. Here, it is assumed that only a single fault is to be detected and isolated at a time. The simulation results show that the proposed structure can be suggested for linear models especially for physical redundancy-based real-time applications easily, quickly and effectively.

This paper aims to discuss engine health monitoring for unmanned aerial vehicles. It is intended to make consistent predictions about the future status of the engine performance parameters by using their current states.

The aim is to minimize risks before they turn into problems. In accordance with these objectives, temporal and financial savings are planned to be achieved by contributing processes such as extending the engine life, preventing early disassembly-reassembly and mechanical wears and reducing the maintenance costs. Based on this point of view, a data-based software is developed in MATLAB (Matrix Laboratory) program for the so-called process.

The software is operated for the performance parameters of the turbojet engine that is used in a small unmanned aerial vehicle of Tusas Engine Industry. The obtained results are compared with the real data of the engine. As a result of this comparison, a fault that may occur in the engine can be detected before being determined.

It is clearly demonstrated that the engine operation in adverse conditions can be prevented. This situation means that the software developed operates successfully.

Stable approach concept has great importance for the safe operation of an airline during the approach and landing phases. The purpose of this study is to analyse the unstabilized approaches with bow-tie method and determine the threats that may cause risk in an unstable approach.

In this study, risk assessment of the unstabilized approaches is carried out by using fuzzy bow-tie method and Bayesian networks. Bow-tie method is the combination of event tree analysis and fault tree analysis. Bayesian network is used in the analysis to see interrelationship of basic and intermediate events as well as to update posterior probabilities. Finally, analysis results are verified by the safety performance indicator values.

In this study, the probabilistic values of the numerical model presented by the risk assessment system for risks were calculated using the fuzzy bow-tie method. Thus, the risk assessment system has been transformed into a structure that can be expressed in a probabilistic manner, and the relationship of the risks within the system has been examined and the effect of a possible change on the risk value has been found to be prevalent.

The bow-tie model is widely applied to assess the risks in aviation. Obtaining prior probabilities is not always possible in the risk assessment process. In this paper, innovative fuzzy bow-tie method is used to assess the risks to overcome the lack of prior probability problem in aviation operations.

This paper aims to present a nonlinear mathematical model of a small-scale turbojet aeroengine and also a speed controller design that is conducted for the constructed nonlinear mathematical model.

In the nonlinear mathematical model of the turbojet engine, temperature, rotational speed, mass flow, pressure and other parameters are generated using thermodynamic equations (e.g. mass, energy and momentum conservation laws) and some algebraic equations. In calculation of the performance parameters, adaptive neuro fuzzy inference system (ANFIS) method is preferred in related components. All calculated values from the mathematical model are then compared with the cycle data of the turbojet engine. Because of the single variable control need and effect of noise factor, modified proportional–integral–derivative (PID) controller is treated for speed control. For whole operation envelope, various PID structures are designed individually, according to the operating points. These controller structures are then combined via gain-scheduling approach and integrated to the nonlinear engine model. Simulations are performed on MATLAB/Simulink environment for design and off-design operating points between idle to maximum thrust levels.

The cascade structure (proposed nonlinear engine aero-thermal model and speed controller) is simulated and tested at various operating points of the engine and for different transient conditions. Simulation results show that the transitions between the operating points are found successfully. Furthermore, the controller is effective for steady-state load changes. It is suggested to be used in real-time engine applications.

Because of limited data, only speed control is treated and simulated.

It can be used as an application in the industry easily.

First point of novelty in the paper is in calculation of the performance parameters of compressor and turbine components. ANFIS method is preferred to predict performance parameters in related components. Second novelty in the paper can be seen in speed controller design part. Because of the single variable control need and effect of noise factor, modified PID is treated.

The purpose of this paper is to improve risk assessment processes in airline flight operations by introducing a dynamic risk assessment method.

Fuzzy logic and Bayesian network are used together to form a dynamic structure in the analysis. One of the most challenging factors of the analyses in aviation is to get quantitative data. In this study, the fuzzy data quantification technique is used to perform dynamic risk assessment. Dynamic structure in the analysis is obtained by transforming the bow-tie model into a Bayesian network equivalent.

In this study, the probability of top-event from fault tree analysis is calculated as 1.51 × 10⁻⁶. Effectiveness of the model is measured by comparing the analysis with the safety performance indicator data that reflects past performance of the airlines. If two data are compared with each other, they are at the same order of value, with small difference (0.6 × 10⁻⁷).

This study proposes a dynamic model to be used in risk assessment processes in airline flight operations. A dynamic model for safety analysis provides real-time, autonomous and faster risk assessment. Moreover, it can help in the decision-making process and reduce airline response time to undesired states, which means that the proposed model can contribute to the efficiency of the risk management process in airline flight operations.

The paper aims to address the tracking algorithm based on deep learning and four deep learning tracking models developed. They compared with each other to prevent collision and to obtain target tracking in autonomous aircraft.

First, to follow the visual target, the detection methods were used and then the tracking methods were examined. Here, four models (deep convolutional neural networks (DCNN), deep convolutional neural networks with fine-tuning (DCNNFN), transfer learning with deep convolutional neural network (TLDCNN) and fine-tuning deep convolutional neural network with transfer learning (FNDCNNTL)) were developed.

The training time of DCNN took 9 min 33 s, while the accuracy percentage was calculated as 84%. In DCNNFN, the training time of the network was calculated as 4 min 26 s and the accuracy percentage was 91%. The training of TLDCNN) took 34 min and 49 s and the accuracy percentage was calculated as 95%. With FNDCNNTL, the training time of the network was calculated as 34 min 33 s and the accuracy percentage was nearly 100%.

Compared to the results in the literature ranging from 89.4% to 95.6%, using FNDCNNTL, better results were found in the paper.