Showing papers by "Farzad Hashemzadeh published in 2017"

Surrounding control of nonlinear multi-agent systems with non-identical agents

Abstract: In this paper, the surrounding control problem of a group of non-identical agents is considered, where a team of followers achieves an equidistant distributed formation to surround a team of moving leaders. An adaptive design method is presented for multi-agent systems where the dynamics of agents are supposed to be nonlinear with unknown parameters. First, an estimator for the center of the leaders is introduced. Then, two distributed adaptive controllers based on the estimated center are proposed for each follower. The stability and parameter convergence of the proposed protocols are shown by using algebraic graph theory and Lyapunov theory. Finally, a numerical example is provided to validate the theoretical results.

Dual-user nonlinear teleoperation subjected to varying time delay and bounded inputs

Abstract: A novel trilateral control architecture for Dual-master/Single-slave teleoperation system with taking account of saturation in actuators, nonlinear dynamics for telemanipulators and bounded varying time delay which affects the transmitted signals in the communication channels, is proposed in this paper. In this research, we will address the stability and desired position coordination problem of trilateral teleoperation system by extension of (nP+D) controller that is used for Single-master/Single-slave teleoperation system. Our proposed controller is weighted summation of nonlinear Proportional plus Damping (nP+D) controller that incorporate gravity compensation and the weights are specified by the dominance factor, which determines the supremacy of each user over the slave robot and over the other user. The asymptotic stability of closed loop dynamics is studied using Lyapunov-Krasovskii functional under conditions on the controller parameters, the actuator saturation characteristics and the maximum values of varying time delays. It is shown that these controllers satisfy the desired position coordination problem in free motion condition. To show the effectiveness of the proposed method, a number of simulations have been conducted on a varying time delay Dual-master/Single-slave teleoperation system using 3-DOF planar robots for each telemanipulator subjected to actuator saturation.

On distance-based formation control: A nonlinear mapping approach and its applications

Abstract: This paper introduces a novel decentralized distance-based formation controller for multi-agent systems by imposing constraints on inter-agent distance errors. Using a Nonlinear Mapping Technique (NMT) for dealing with the distance errors constraints, the problems of connectivity maintenance and collision avoidance between neighboring agents are handled. Furthermore, the proposed control scheme increases formation robustness and prevents convergence of the formation to incorrect shapes when external effects such as disturbances exist. Using Lyapunov based analysis the results are proved for both single and double integrator agent models where the target formation is minimally and infinitesimally rigid in 2-D or 3-D space.

Scaled consensus of descriptor multi-agent systems

Abstract: Multi-agent systems have been studied from different perspectives. In this paper, the problem of scaled consensus is considered for descriptor multi-agent systems. In fact, each agent is considered as a descriptor system. This is significant due to the fact that descriptor systems describe a larger group of systems. Furthermore, an algorithm is proposed to drive the agents to different consensus points, which is denoted as scaled consensus. This can also have very important applications in areas such as multi-robot systems where each robot needs to perform a different task. Moreover, in contrast with the works in the literature on scaled consensus, higher order systems are allowed in this paper. Also, as practical systems may suffer from time delay, these systems under a constant delay are considered. Finally, numerical examples are provided to further demonstrate the effectiveness of the proposed algorithms.

Distributed fault tolerant control and estimation for leader-follower consensus of multi-agent systems with actuator faults

Abstract: Leader-follower consensus tracking problem in multi-agent systems in the presence of some types of faults named actuator faults are considered in this paper. Distributed control theory has been used to develop control strategy and main challenge of this paper is to design fault tolerant control (FTC) system based on fault estimation to minimize the effects of the actuator faults. For this issue, an appropriate luenberger type fault detection filter is used to detect the occurance of the fault. Fault estimations are obtained by recursive least square (RLS) method and using virtual actuator technique, the effects of faults are minimized and it is shown that the tracking error of each agent converges to zero in fault-free case and the tracking error remains bounded in faulty case. The effectiveness of our method has been verified through simulations.

Delay-independent Li controller for linear systems with time-varying delay

Abstract: It is a well-stablished theory that stability analysis and design of time-delay systems are not an easy task to deal with especially when the delay is time varying. In this article, a new delay-independent controller is introduced. The sub-optimal criterion proposed in this paper is independent of delay volume and its variation rate. The method proposed in this article takes advantage of L1 small gain theory and its applications to stabilize linear time-delay systems. In this document, the delay modeled as an uncertainty where its 1-norm is lower than one. Reducing the L1 norm of the closed loop system is transformed into a matrix inequality problem and the controller is evaluated using LMI method.

A Fast and Simple Drowsiness Detection System Based on ARM Microcontrollers

Abstract: Drowsiness and fatigue of the drivers are responsible for sever accidents and large human life losses and monetary losses. This paper presents a method based on the image processing for drowsiness detection. Fast and simple algorithms are proposed for face and pupil detection. The proposed method is implemented using a cheap and commercial microcontroller. The implemented device can be calibrated easily for each driver in their first usage. The proposed method is simulated using MATLAB and after getting the satisfactory simulation results, the proposed method is implemented and experimental results are obtained. Using the proposed method, the drowsiness detection system can be used widely in larger communities.

A new approach on studying the stability of evolutionary game dynamics for financial systems

Abstract: ‎Financial market modeling and prediction is a difficult problem and drastic changes of the price causes nonlinear dynamic that makes the price prediction one of the most challenging tasks for economists‎. ‎Since markets always have been interesting for traders‎, ‎many traders with various beliefs are highly active in a market‎. ‎The competition among two agents of traders‎, ‎namely trend followers and rational agents‎, ‎to gain the highest profit in market is formulated as a dynamic evolutionary game‎, ‎where‎, ‎the evolutionary equilibrium is considered to be the solution to this game‎. ‎The evolutionarily stablity of the equilibrium points is investigated inspite of the prediction error of the expectation‎.

A Novel Method of Modeling Dynamic Evolutionary Game with Rational Agents for Market Forecasting

Abstract: Gold price modeling and prediction is a difficult problem and drastic changes of the price causes nonlinear dynamic that makes the price prediction one of the most challenging tasks for economists. Since gold market always has been interesting for traders, many of traders with various beliefs were highly active in gold market. The competition among two agents of traders, namely trend followers and rational agents, to gain the highest profit in gold market is formulated as a dynamic evolutionary game, where, the evolutionary equilibrium is considered to be the solution to this game. Furthermore, genetic algorithm is being used to find the unknown parameters of the model, so that we could maximize the fitness of the proposed multi agent model and the gold market daily price data. Besides the evolutionary game dynamic, we proposed a new method for modeling rational expectations using recurrent neural network. The evolutionarily stable strategies is proven despite the prediction error of the expectation. The empirical results show the high efficiency of the proposed method which could forecast future gold price precisely.

Position and velocity synchronization in bilateral teleoperation in presence of stochastic disturbances in control inputs

Abstract: The problem of synchronization of bilateral teleoperators in the presence of stochastic disturbances in control inputs is considered in this paper. It is clear that the mechanical systems are often subjected to random disturbances and it can influence the performance of the control system in an uncertain manner. To cope with this, the new adaptive controller is proposed. This technique uses the exponential practical stability concept which guarantees that the tracking error and its derivative converge to an arbitrarily small neighborhood of zero by appropriate tuning of the controller's parameters. It is noteworthy that, the proposed method does not need information, such as the physical parameters of the master and slave robots. Finally, the simulation results are given to show the effectiveness of proposed technique.

An analytical eigenvalue assignment of linear time-delay systems using Lambert W function

Abstract: A new method is presented in this article to control linear time-delay systems using Lambert W function. The absence of a sufficient proof for using matrix Lambert W function in controlling time-delay systems has frustrated researchers. It is a well-established fact that unlike Lambert W Function, the matrix version still lacks the proof, which would guarantee the validity of finding the right-most eigenvalue of linear time-delay systems using this function. This paper presents a new approach in which, a system from any order is forced to use scalar Lambert W function instead of unreliable matrix version. This article studies systems with input delay and state delay respectively and for both of them, sufficient conditions for eigenvalue assignment are proposed.

Controller design to enlarge the domain of attraction for a class of nonlinear systems

Abstract: Estimation and enlarging the domain of attraction of nonlinear systems is one of the difficult and important challenges in the control of dynamical systems. This case efficiently influences the performance and reliability of controlled nonlinear mechatronic systems. In this paper a wide range of affine nonlinear systems has been considered and based on Zubov Theorem, a controller has been designed in which choosing suitable coefficients for the Lyapunov function stabilizes and enlarges the domain of attraction of closed loop system. The proposed method has been simulated on Van der Pol oscillator and on a non-globally stabilizable system to show the efficiency of the method.

Hybrid driven based robust stabilization for networked control systems

Abstract: This paper considers robust stabilization controller design for networked control systems with communication delay and bounded system uncertainties under a hybrid framework. Hybrid-driven scheme is used to reduce the transmission data. The closed-loop system with a networked state feedback controller is modeled as a system with delayed control input. Criteria for exponential stability in the mean-square sense of the closed-loop system is derived in terms of Linear Matrix Inequalities (LMIs) and a feedback controller to reach a robust stability is designed. Finally, the effectiveness of the proposed method is demonstrated by simulation results.

Bounded Adaptive Control of Uncertain Nonlinear Teleoperation with Actuators Saturation and Asymmetric Varying Time Delays

Abstract: In this paper, adaptive controller is proposed for bilateral teleoperation systems with parametric uncertainties, asymmetric time-varying delays in their communication channel and sandwich linearity in their actuators. The proposed bounded adaptive approach is a nonlinearProportional plus nonlinear Damping (nP+nD) controller which nonlinear adaptive terms are also employed locally to cope with the parametric uncertainties. The designed controller has the ability to deal with parameter variations in the dynamics of the master and slave robots . The proposed controller also guarantees asymptotic stability and position tracking and avoiding the inputs to reach their usual actuator bounds. The asymptotic stability and tracking performance of the teleoperation system are proved using a Lyapunov-Krasovskii functional. The effectiveness of the proposed method is corroborated through simulation results.

Enlarging Domain of Attraction for Affine Nonlinear Systems

Abstract: In this paper, a new method based on Zubov theorem is proposed to enlarge the domain of attraction for a class of affine nonlinear systems. The proposed controller provides the ability to stretch the domain of attraction of affine nonlinear system in any desired dimension of state space. Not only that but the controller makes it possible to stretch the domain of attraction along some directions and compress it along others. Also the proposed approach reduces some constraints of the previous works and is capable to be applied to the larger class of affine nonlinear dynamic system. These constraints are based on the fact that the controller value must be bounded. The Simulation results on the Van der pol system and a three-dimensional system, demonstrate the efficiency of the proposed method.