Kagan Koray Ayten

Single image dehazing based on multiscale product prior and application to vision control

In this paper, a novel dehazing algorithm based on multiscale product (MSP) prior is presented. First, the observed hazy image is decomposed into its approximation and detail subbands by undecimated Laplacian decomposition. Then the MSPs of the approximation subbands for each band of the image are calculated to obtain the MSP prior. This prior keeps the significant information of the image, whereas it is capable of detecting the haze in the image. By the use of this prior and the hazy image model, a fast and robust dehazing algorithm is obtained. The proposed method is compared with commonly used methods. The results demonstrate that the proposed algorithm is better than the former methods. Being a fast and robust algorithm, the proposed method has also been applied to a real-time robotic vision control system.

A comparative study of two model-based control techniques for the industrial manipulator

In this paper, design, analysis and real-time trajectory tracking control of a 6-degree of freedom revolute spherical-spherical type parallel manipulator, actuated by six hybrid stepper motors, has been studied. Two different control approaches have been used to improve the trajectory tracking performance of the designed manipulator. The first approach considered a single input-single output (SISO) linear quadratic regulator (LQR) for trajectory tracking control of the manipulator. Another controller type based on a nonlinear sliding mode controller method has been utilized to take decoupled dynamic approximation model of the manipulator into account and to improve tracking performance of the manipulator. Real-time experimental results for the two different control techniques have been verified. Finally, according to the results, the nonlinear sliding mode controller method has improved the tracking performance of the designed manipulator.

Real time optimum trajectory generation for redundant/hyper-redundant serial industrial manipulators

This article presents an optimization technique to develop minimum energy consumption trajectories for redundant/hyper-redundant manipulators with predefined kinematic and dynamic constraints. The optimization technique presents and combines two novel methods for trajectory optimization. In the first method, the system’s kinematic and dynamic constraints are handled in a sequential manner within the cost function to avoid running the inverse dynamics when the constraints are not satisfied. Thus, the complexity and computational effort of the optimization algorithm is significantly reduced. For the second method, a novel virtual link concept is introduced to replace all the redundant links to eliminate physical impossible configurations before running the inverse dynamic model for the trajectory optimization. The method is verified on a three-degree of freedom redundant manipulator and the result is also demonstrated with computer simulations based on an 8-link planar hyper-redundant manipulator.

A combined control approach for industrial process systems using feed-forward and adaptive action based on second order sliding mode controller design

This paper presents a new method that uses a feed-forward adaptive second order sliding mode control (FF-ASSMC) for a coupled tank process system that is often used in industry. This proposed method is to combine a feedforward action with adaptive feedback for robust performance of the couple tank system for the first time. The reason for using this feedforward action with the feedback sliding mode control (SMC) is to reduce the effect of the measured disturbance on the coupled tank system. In addition to this, an adaptive second-order SMC proposes to track the desired tank’s level trajectories under disturbances and uncertainties. In order to illustrate the efficiency of the proposed FF-ASSMC method, the simulation and real time experimental studies have been realized. The simulation and experimental outcomes strongly verified that the proposed controller gives a quite successful trajectory tracking response and smaller magnitude overshot under external disturbances.

Real-time implementation of self-tuning regulator control technique for coupled tank industrial process system

This article presents the self-tuning regulator control technique for a coupled tank liquid level system that often used in industry. An autoregressive with exogenous model has been used as the liquid process model with the self-tuning control implementation in order to track the desired tank level trajectories with disturbances and uncertainties of the system dynamics. The designed self-tuning controller has been sensitive to parameter variations of the nonlinear coupled tank system. The parameters of the proposed controller are periodically updated themselves during the process by means of online recursive least square method with the forgetting factor algorithm. In this way, the parameter variations and unwanted disturbances of the system are eliminated in real-time application. In order to demonstrate the efficiency of the self-tuning regulator control technique, the real-time studies have been executed. The obtained experimental results demonstrated that the proposed controller gives the better trajectory tracking performance and smaller magnitude in overshot and undershot than the designed classical proportional–integral and sliding mode controllers.

Bir Otonom Elektrikli Tekerlekli Sandalyenin Yörünge Kontrolünde Hesaplamalı Tork Kontrol ve PID Kontrol Yöntemlerinin Karşılaştırılması

Bu calismada bir otonom elektrikli tekerlekli sandalyenin (OETS) analizi, tasarimi ve gercek zamanli yorunge takip kontrolu gerceklestirilmistir. Oncelikle tasarlanan OETS’ nin kinematik analizi yapilmis ve ardindan sistemin tam dinamik modeli Lagrangian tekniginden yararlanilarak elde edilmistir. Sistemin gercek zamanli yorunge izleme performansini artirabilmek icin, bu calismada farkli kontrol yontemlerinden yararlanilmistir. Ilk olarak sistemin yorunge izleme kontrolu PID kontrolu uzerinden gerceklestirilmistir. Ardindan, sistemin yorunge izleme performansini artirabilmek icin OETS’ nin tam dinamik modelinin goz onune alindigi Model Tabanli Hesaplamali Tork Kontrol yonteminden yararlanilmistir. Yapilan gercek zamanli yorunge izleme performans deneylerinden elde edilen sonuclara gore, sistem icin yapilan matematiksel analizlerin dogru oldugu kanitlanmis ve ozellikle hesaplamali tork yonteminin kullanilmasi ile sistemin yorunge takip kontrolu guclendirilmistir.

Real-time trajectory tracking control for electric-powered wheelchairs using model-based multivariable sliding mode control

In this paper, real-time trajectory tracking control of an electric-powered wheelchair (EPW), actuated by two DC motors, has been designed, analysed and studied. Two different control approaches such as nonlinear SMC and the classical proportional–integral–derivative (PID) are employed to increase the tracking performance of the electric-powered wheelchair. A nonlinear SMC technique has been presented in order to consider the complete dynamic model of the EPW and so as to increase trajectory tracking performance of the EPW. In this study, realtime experimental application of the two different control methods has been realized. Lastly, in terms of the results, the nonlinear SMC technique has improved the trajectory tracking performance of the designed EPW.

Sensorless model reference adaptive system control for mobile robot

In this paper, a robust and an efficient tracking controller based on model reference adaptive system (MRAS) technique is performed for mobile robot in order to get the accurate trajectory tracking performance. The model reference adaptive system (MRAS) technique has been presented so as to achieve the estimated value of systems speed information. To estimate the rotor position and speed of the mobile robot system, the MRAS technique is one of the best methods which can be applied to the mobile robot system due to its performance in terms of decrease computational complexity, increase robustness, easy to implement to the systems. In addition, the control scheme of the mobile robot also performs the perfect speed tracking during the motion. All the simulations have verified that the proposed control algorithm is indeed feasible and effective.