مجله کنترل
سال چهاردهم شماره 4 (زمستان 1399)

Bearings only tracking is a challenging issue with many applications in military and commercial areas. In distributed multi-sensor multi-target bearings only tracking, sensors are far from each other, but are exchanging data using telecommunication equipment. In addition to the general benefits of distributed systems, this tracking system has another important advantage: if the sensors are sufficiently spaced apart, the target state is observable and the maneuver is not necessary by sensors. In this work, Multi-sensor multi-target bearings only tracking with decentralized architecture and asynchronous measurements is newly proposed. In this study, with the help of the idea of ​​composite measurements and taking into account the time of measurements in the calculations, while overcoming the asynchronous of the measurements, the nonlinear effects in the measurement equation are also eliminated. Also, diffusive filtering is used to exploit the information of neighboring sensor agents to improve the estimates. The simulations show that the system designed in this research can well detect targets and track them with acceptable accuracy.

To accelerate the learning process in high-dimensional learning problems, the combination of TD techniques, such as Q-learning or SARSA, is usually used with the mechanism of Eligibility Traces. In the newly introduced DQN algorithm, it has been attempted to using deep neural networks in Q learning, to enable reinforcement learning algorithms to reach a greater understanding of the visual world and to address issues Spread in the past that was considered unbreakable. DQN, which is called a deep reinforcement learning algorithm, has a low learning speed. In this paper, we try to use the mechanism of Eligibility Traces, which is one of the basic methods in reinforcement learning, in combination with deep neural networks to improve the learning process speed. Also, for comparing the efficiency with the DQN algorithm, a number of Atari 2600 games were tested and the experimental results obtained showed that the proposed method significantly reduced learning time compared to the DQN algorithm and converges faster to the optimal model.

Designing the estimator that can determine the attitude with a single sensor is vital due to the limited weight and volume in the nano-satellite, the problems caused by the limited lifetime of the mechanical gyroscope in the long term and the eclipse phenomenon. To compensate for data deficiency, a two-nested filter has been utilized in this paper. To this end, the attitude in the second filter is estimated using the sensor data and the magnetic field derivative estimation from the first filter by the extended Kalman filter. Two stochastic algorithms named as multiplicative extended Kalman filter and square-root unscented quaternion estimator are compared with the proposed symmetry-preserving nonlinear observer in order to obtain an appropriate accuracy for determining the attitude of the nano-satellite, which has only a three-axis magnetometer. The proposed method is based on invariant observers under the action of the Lie group. The moving frame approach has been used so that the observerchr('39')s parameters can be adjusted through the invariant error dynamic equations. Simulation results confirm an acceptable accuracy in all three algorithms for both time and frequency response analyses. However, the root mean square error of the attitude error with a nonlinear observer is much less than the stochastic algorithm in case of a larger initial estimation error. Furthermore, this approach guarantees convergence by the Lyapunov stability proof owing to setting the parameters with periodic differential Riccati equations.

This paper addresses the asymptotic stability, position and force tracking problem in the nonlinear bilateral teleoperation system in the presence of non-passive interaction forces, varying time-delay in communication channel and actuators fault occurrence. For this aim, a passive Fault Tolerant Control (FTC) law is presented which uses the joint positions and velocities of local and remote manipulators to reach control ends. Using the Lyapunov-Krasovskii theorem, sufficient conditions for asymptotic stability and position tracking are derived in terms of Linear Matrix Inequalities (LMIs), to tune controller parameters. The main contribution of the proposed method is that can compensate the bias fault and loss of effectiveness of actuators in nonlinear teleoperation system. Also the asymptotic stability of positions errors in the system with non-passive interaction forces is assured using the integral control. Simulation results of teleoperation system with 2 and 3 degrees of freedom manipulators with proposed method and comparison to some rival methods show the effectiveness and advantages of the proposed method.

Nowadays, collaborative multi-agent systems in which a group of agents work together to reach a common goal, are used to solve a wide range of problems. Cooperation between agents will bring benefits such as reduced operational costs, high scalability and significant adaptability. Usually, reinforcement learning is employed to achieve an optimal policy for these agents. Learning in collaborative multi-agent dynamic environments with large and stochastic state spaces has become a major challenge in many applications. These challenges include the effect of size of state space on learning time, ineffective collaboration between agents and the lack of appropriate coordination between decisions of agents. On the other hand, using reinforcement learning has challenges such as the difficulty of determination the appropriate learning goal or reward and the longtime of convergence due to the trial and error in learning. This paper, by introducing a communication framework for collaborative multi-agent systems, attempts to address some of these challenges in herding problem. To handle the problems of convergence, knowledge transfer has been utilized that can significantly increase the efficiency of reinforcement learning algorithms. Cooperation and Coordination and between the agents is carried out through the existence of a head agent in each group of agents and a coordinator agent respectively. This framework has been successfully applied to herding problem instances and experimental results have revealed a significant improvement in the performance of agents.

Nowadays, one of the most fundamental processes for realization video of contents is the object tracking, in which the process of location the moving object is performed in each video frame. In tracking process, the target must be described by a feature. In this paper, for the purpose of describing the target and removing the appearance sensitivity, the weighted color histogram is used as a target feature in order to reduce the effect of edge pixels on the target feature. This reduces the sensitivity of the algorithm to change deformation, scale variation and rotation, as well as the occlusion on the description of target feature. In the proposed method, particle swarm optimization algorithm has been used for search process. Maximization of the similarity function and calculating the minimum Bhattacharyya distance are used to determine target location. The fuzzy control parameters are used for the particle swarm optimization algorithm, which provides a novel method, which can regulate each control parameter and update according to the different states of each particle in each generation. The improved particle swarm algorithm is evaluated with 11 benchmark functions. The obtained results by improved algorithm show that appropriate convergence in a low number of iterations. The proposed method compared to state-of-the-art methods provides high performance in the success and precision rate on the OTB50 dataset.

: The Robust Cooperative Output Consensus (RCOC) in continuous time Heterogeneous Multi-Agent Systems with the directed graph is addressed. In the standard solution of the RCOC, the p-copy internal model method is used. This method requires dynamical equations of the agents and the leader. In the present paper, based on the equivalent auxiliary system method, a new auxiliary system is obtained. Then, the RCOC is transformed to a control problem. Moreover, a model-free algorithm is proposed to solve the Robust Algebraic Riccati Equation using the Integral Reinforcement Learning (IRL) method. It is shown that the proposed method satisfies the output regulation equations. A simulation example verifies the effectiveness of the proposed method.

Alarm systems are essential in safe operation of industrial plants. Since many process variables are interacting with each other, so in this paper, a new method are introduced to design and analysis of a multivariate alarm system. In this method, the alarm system is designed base on joint indices. The Joint FAR and Joint MAR are defined for a m-variable alarm system thanks to multivariate Markov scheme. In proposed method, the alarm joint indices are defined by solving a Linear Programing (LP) optimization problem. By defining joint indices, tuning of the alarm parameters (like, threshold and etc.) can be done by these indices instead of correlation analysis. In this paper, penalty scenario and Genetic algorithm are used for alarm generation, and parameter optimization in Tennessee Eastman (TE) Process. The results of proposed method are compared with other methods.

This paper proposes robust  output feedback control stabilization for uncertain Takagi–Sugeno (T-S) fuzzy systems via linear matrix inequalities (LMIs). In order to reduce the conservatism associated with T-S fuzzy system, a new form of non-monotonic Lyapunov functions is used. In the non-monotonic approach, the monotonic decrease of the Lyapunov function is relaxed which enables it to increase locally but vanish eventually. Based on the non-monotonic Lyapunov function approach, sufficient conditions for the existence of robust  output feedback control stabilization are derived. The proposed design technique is shown to be less conservative than the existing non-monotonic approach, namely, -samples variations of Lyapunov function. The effectiveness of the proposed approach is further illustrated via numerical example.

Micro grids are regarded to be crucial due to the development of communication facilities, production equipment, power storage and high utilization of renewable energy. However, micro grids are faced with the problem of frequency fluctuations in island mode because of high fluctuations in the production power of renewable sources. In this paper, the super capacitor has been used to increase the virtual inertia of the network along with the control system for energy storage and production systems in order to overcome the challenge mentioned earlier. Meanwhile, the optimal value of communication systems delay is also considered in the frequency settings of the micro grid.The system frequency behavior is simulated against load and power generation variations to evaluate the performance of the proposed strategy. Given the increased cost of the system due to increased super capacitor capacity and the reduction of communication  delay against the improvement of micro grid frequency fluctuations, a multi-objective optimization method is used to regulate the load frequency (LFC) controllers parameters and  achieve minimum cost. The simulations of the network in question have been carried out in MATLAB / SIMULINK software. In this article, the cost reduction of operation of the microgrid due to the low capacity of the installed supercapacitor unit and communication systems with acceptable delay achievement are the most important innovational aspects of the current research. Furthermore, no battery units are required in the current paper thanks due to the presence of the supercapacitor. Batteries may cause problems for the network due to their low life and high maintenance costs. Simulation results have demonstrated that the system with optimal control parameters values, super capacitor capacity and communication s system delay has been able to overcome load and power generation disturbances, and the system frequency behavior is significantly improved in comparison with the non-optimal state

This paper, studies the controller design for containment problem for a class of multi-agent systems with identical time-invariant continuous-time nonlinear dynamics and fixed directed communication graph. In this problem, the lyapunov stability theorem, the graph theory and matrix linear inequality are used. The agents are divided into two groups of leaders and followers. In containment problem, all the followers are controlled under which will asymptotically converge to the convex hull spanned by the leaders so distributed communication protocol with fixed time delay is considered and four-step algorithm is proposed for obtaining parameters and gain matrix. The above case is proved to be sufficient condition under theorem. To illustrate the reliability and efficiency of the proposed method, numerical example with simulations are presented.

The increasing expansion of automation has led to increasing the use of electric motors. that makes the main power of many instruments. The Switched Reluctance Motor, as a kind of synchronous motors, has many advantages and can be used instead of other motors to eliminate their problems. speed control of this motor is very difficult due to nonlinearities, time variant, and uncertainties. In this article, the speed control of SRM is considered by using an optimal sliding mode controller. Using the cascade structure, the biggest defect in SRM, the torque ripple, is reduced. The proposed controller, by converting the first-order sliding mode control problem to the optimization problem, and with its real-time solution by projection recurrent neural network, produces an optimal control signal that does not have chattering, but satisfies the sliding condition. evaluation of The proposed controller with other controller is carried out by simulation and its effectiveness is shown.