Connection admission control of ATM network using integrated MLP and fuzzy controllers

This paper presents a new approach to the problem of call admission control (CAC) of variable bit rate (VBR) traffic in an asynchronous transfer mode (ATM) network. Our approach employs an integrated neural network and fuzzy controller to implement the CAC controller. This scheme capitalizes on the learning ability of a neural network and the robustness of a fuzzy controller. Experiments show that this scheme is able to achieve high throughput and low cell loss while achieving fairness among different classes of VBR traffic. For comparison, we have also implemented four other CAC schemes: (1) peak bandwidth method, (2) equivalent bandwidth method, (3) average bandwidth method and (4) neural network quality of service (QoS) predictor. Results of these experiments are presented in this paper.     

Designing a hierarchical fuzzy logic controller using the differential evolution approach

In conventional fuzzy logic controllers, the computational complexity increases with the dimensions of the system variables; the number of rules increases exponentially as the number of system variables increases. Hierarchical fuzzy logic controllers (HFLC) have been introduced to reduce the number of rules to a linear function of system variables. However, the use of hierarchical fuzzy logic controllers raises new issues in the automatic design of controllers, namely the coordination of outputs of sub-controllers at lower levels of the hierarchy. In this paper, a method is described for the automatic design of an HFLC using an evolutionary algorithm called differential evolution (DE).The aim in this paper is to develop a sufficiently versatile method that can be applied to the design of any HFLC architecture. The feasibility of the method is demonstrated by developing a two-stage HFLC for controlling a cart–pole with four state variables. The merits of the method are automatic generation of the HFLC and simplicity as the number of parameters used for encoding the problem are greatly reduced as compared to conventional methods.     

Multiple network fusion using fuzzy logic

Multiplayer feedforward networks trained by minimizing the mean squared error and by using a one of c teaching function yield network outputs that estimate posterior class probabilities. This provides a sound basis for combining the results from multiple networks to get more accurate classification. This paper presents a method for combining multiple networks based on fuzzy logic, especially the fuzzy integral. This method non-linearly combines objective evidence, in the form of a network output, with subjective evaluation of the importance of the individual neural networks. The experimental results with the recognition problem of on-line handwriting characters show that the performance of individual networks could be improved significantly.     

A multiregion fuzzy logic controller for nonlinear process control

Although a fuzzy logic controller is generally nonlinear, a PI-type fuzzy controller that uses only control error and change in control error is not able to detect the process nonlinearity and make a control move accordingly. In this paper, a multiregion fuzzy logic controller is proposed for nonlinear process control. Based on prior knowledge, the process to be controlled is divided into fuzzy regions such as high-gain, low-gain, large-time-constant, and small-time-constant. Then a fuzzy controller is designed based on the regional information. Using an auxiliary process variable to detect the process operating regions, the resulting multiregion fuzzy logic controller can give satisfactory performance in all regions. Rule combination and controller tuning are discussed. Application of the controller to pH control is demonstrated     

Design of a force reflection controller for telerobot systems using neural network and fuzzy logic

This paper presents a new method for selecting the force-reflection gain in a position-force type bilateral teleoperation system. The force-reflection gain greatly affects the task performance of a teleoperation system; too small gain results in poor task performance while too large gain results in system instability. The maximum boundary of the gain guaranteeing the stability greatly depends upon characteristics of the elements in the system such as: a master arm which is combined with the human operator's hand and the environments with which the slave arm contacts. In normal practice, it is, therefore, very difficult to determine such maximum boundary of the gain. To overcome this difficulty, this paper proposes a force-reflection gain selecting algorithm based on artificial neural network and fuzzy logic. The method estimates characteristics of the master arm and the environments by using neural networks and, then, determines the force-reflection gain from the estimated characteristics by using fuzzy logic. In order to show the effectiveness of the proposed algorithm, a series of experiments are conducted under various conditions of teleoperation using a laboratory-made telerobot system.     

A neural network controller for congestion control in ATM multiplexers

This paper presents and adaptive approach to the problem of congestion control arising at the User-to Network Interface (UNI) of an ATM multiplexer. We view the ATM multiplexer as a non-linear stochastic system whose dynamics are ill-defined. Real-time measurements of the arrival rate process and the queueing process, are used to identify, and minimize congestion episodes. The performance of the system is evaluated using a performance-index function which is a quantative measure of “how well” the system is performing. A three-layers backpropagation neural network controller generates a signal that attempts to minimize congestion without degrading the quality of the traffic. During periods of buffer over-load the control signal, adaptively, modulates the arrival process such that its peak-rate is throttled-down. As soon as congestion is terminated, the control signal is adjusted such that the coding rates are restored back to their original values. Adaptability is achieved by continuously adjusting the weights of the neural network controller such that the performance of the system, measured by its performance index function, is maximized over a certain optimization period. The performance index function is defined in terms of two main objectives: (1) to minimize the cell loss rate (CLR), i.e., minimize congestion episodes, and (2) to maintain the quality of the video/audio traffic by maintaining its original source coding rate. The neural network learning process can be viewed as a specialized form of reinforcement learning in the sense that the control signal is reinforced if it tends to maximize the performance index function. Performance evaluation results prove that this approach is effective in controlling congestion while maintaining the quality of the traffic.     

A self-learning fuzzy logic controller using genetic algorithmswith reinforcements

This paper presents a new method for learning a fuzzy logic controller automatically. A reinforcement learning technique is applied to a multilayer neural network model of a fuzzy logic controller. The proposed self-learning fuzzy logic control that uses the genetic algorithm through reinforcement learning architecture, called a genetic reinforcement fuzzy logic controller, can also learn fuzzy logic control rules even when only weak information such as a binary target of “success” or “failure” signal is available. In this paper, the adaptive heuristic critic algorithm of Barto et al. (1987) is extended to include a priori control knowledge of human operators. It is shown that the system can solve more concretely a fairly difficult control learning problem. Also demonstrated is the feasibility of the method when applied to a cart-pole balancing problem via digital simulations     

基于模糊控制器的自适应广义通用模型控制

广义通用模型控制(GCMC)方法是一般模型控制(GMC)的改进,适用于相对阶大于1的复杂多输入多输出系统,该控制器参数具有明显的物理意义,但鲁棒性不够强。将模糊控制与广义通用模型控制相结合,构成模型参考自适应控制系统,从而加强了系统的鲁棒性,仿真实验证明了该策略的有效性。     

Design of an optimal fuzzy logic controller using response surfacemethodology

When the fuzzy logic controller (FLC)-designed based on the plant model-is applied to the real control system, satisfactory control performance may not be attained due to modeling errors from the plant model. Also, a controller that is designed under specific circumstances may not show satisfactory control performance when applied to other circumstances. In such cases, the control parameters of the controller must be adjusted to enhance control performance. Until now, the trial and error method has been used, consuming much time and effort. Also, the set of adjusted values is not guaranteed to be optimal. To resolve such problems, response surface methodology (RSM), a method of adjusting the control parameters of the controller, is suggested. This method is more systematic than the previous trial and error method, thus, optimal solutions can be provided with less tuning. First, the initial values of the control parameters are determined through the plant model and the optimization algorithm. Then designed experiments are performed in the region around the initial values, determining the optimal values of the control parameters that satisfy both the rise time and overshoot simultaneously     

Position control of electro-hydraulic actuator system using fuzzy logic controller optimized by particle swarm optimization

The position control system of an electro-hydraulic actuator system (EHAS) is investigated in this paper. The EHAS is developed by taking into consideration the nonlinearities of the system: the friction and the internal leakage. A variable load that simulates a realistic load in robotic excavator is taken as the trajectory reference. A method of control strategy that is implemented by employing a fuzzy logic controller (FLC) whose parameters are optimized using particle swarm optimization (PSO) is proposed. The scaling factors of the fuzzy inference system are tuned to obtain the optimal values which yield the best system performance. The simulation results show that the FLC is able to track the trajectory reference accurately for a range of values of orifice opening. Beyond that range, the orifice opening may introduce chattering, which the FLC alone is not sufficient to overcome. The PSO optimized FLC can reduce the chattering significantly. This result justifies the implementation of the proposed method in position control of EHAS.     

A self-organizing fuzzy logic controller for the active control of flexible structures using piezoelectric actuators

This paper proposes an on-line self-organizing fuzzy logic controller (FLC) design applied to the control of vibrations in flexible structures containing distributed piezoelectric actuator patches. In this methodology, the fuzzy rules are generated using the history of input/output (I/O) pairs without using any plant model. The generated rules are stored in the fuzzy rule space and updated on-line by a self-organizing procedure. The validity of the proposed fuzzy logic control has been demonstrated experimentally in a steel cantilever test beam and a set of experimental tests are made in the system to verify the efficiency of the on-line self-organizing fuzzy controller.     

Vague controller: A generalization of fuzzy logic controller

In this research, a vague controller (VC) is synthesized by using the notion of vague sets, which are a generalization of fuzzy sets and characterized by a truth-membership function and a falsity-membership function. The vague sets follow the basic set operations and logic operations defined for fuzzy sets, and are superior to fuzzy sets in that they could deal with the uncertainty encountered in real-world applications in a more natural way. Depending on the vague sets, the VC is developed as a generalization of fuzzy logic controller (FLC). The design procedures of the VC, which allow an arbitrary number of input variables, and each variable could have a distinct number of linguistic values, are outlined in this paper. In order to compensate the effort in constructing two series of membership functions for vague sets and to ease the difficulties in designing VCs, a new means of designating membership functions for VCs is also presented in this article. This method constructs a set of membership functions systematically by using only two parameters: number of linguistic values of a linguistic variable and shrinking factor. The membership functions generated by this method, shrinking-span membership functions (SSMFs), have different spans over the universe of discourse and, therefore, are more rational and more practical from the human expert's point of view.     

A fuzzy logic based supervisory hierarchical control scheme for real time pressure control

This paper describes a supervisory hierarchical fuzzy controller （SHFC） for regulating pressure in a real-time pilot pressure control system. The input scaling factor tuning of a direct expert controller is made using the error and process input parameters in a closed loop system in order to obtain better controller performance for set-point change and load disturbances. This on-line tuning method reduces operator involvement and enhances the controller performance to a wide operating range. The hierarchical control scheme consists of an intelligent upper level supervisory fuzzy controller and a lower level direct fuzzy controller. The upper level controller provides a mechanism to the main goal of the system and the lower level controller delivers the solutions to a particular situation. The control algorithm for the proposed scheme has been developed and tested using an ARM7 microcontroller-based embedded target board for a nonlinear pressure process having dead time. To demonstrate the effectiveness, the results of the proposed hierarchical controller, fuzzy controller and conventional proportional-integral （PI） controller are analyzed. The results prove that the SHFC performance is better in terms of stability and robustness than the conventional control methods.     

Freeway traffic control using fuzzy logic controllers

A major cause of freeway congestion before the traffic density becomes critical is the shock wave due to the speed differences between consecutive vehicles. Such disturbance can be reduced if we can impose homogeneous speed control on the vehicles. In this paper, a two-level model-free control scheme using neural-network-based fuzzy logic controllers is proposed which regulates the speed of the freeway through speed advisory boards. Using information from both measurement data and expert knowledge (e.g., environmental information and psychological factors), it is expected that this controller will outperform the conventional ones.     

A study of fuzzy logic controller robustness using the parameter plane

An attempt is made to obtain a quantitative analysis of fuzzy logic controller robustness using parameter plane techniques. The method requires the controller, which is modelled as a multilevel relay, to be represented by a describing function. Therefore, the analysis can only be applied to systems where the input to the non-linearity is sufficiently close to a sinusoid to justify this representation. It is found that, although general quantitative results cannot be obtained, the method does produce a good indication of controller robustness] for the plant investigated.     

基于Additive2multipl icative 模糊 神经网的ATM 网络拥塞控制

考虑了模糊神经网络的学习功能，提出利用Additive-multiplicative模糊神经网络(AMFNN)对ATM网络进行拥塞控制的方案．在拥塞控制过程中，利用AMFNN模糊神经网络预测下一个将要到达流的特征，结合当前缓冲区的队列信息预测网络是否发生拥塞．一旦预测出将有拥塞发生，控制器则向源端反馈拥塞控制信息，信源根据拥塞信息适当降低传输速率，从而避免了拥塞的发生．仿真结果表明，该方法可改善网络对拥塞的实时处理能力，提高网络资源的利用率．     

A combination of fuzzy logic and neural network controller for multiple-input multiple-output systems

Generally, the difficulty of multiple-input multiple-output (MIMO) systems control is how to overcome the coupling effects between the degrees of freedom. Owing to the computational burden and dynamic uncertainty of MIMO systems, the model-based decoupling approach is not practical for real-time control. A hybrid fuzzy logic and neural network controller (HFNC) is proposed here to overcome this problem and to improve the control performance. Firstly, a traditional fuzzy controller (TFC) is designed from a single-input single-output (SISO) systems viewpoint for controlling the degrees of freedom of a MIMO system. Secondly, an appropriate coupling neural network controller is introduced into the TFC for compensating the system coupling effects. This control strategy not only can simplify the implementation problem of fuzzy control but also can improve the control performance. The state-space approach for fuzzy control systems stability analysis is employed to evaluate the stability and robustness of this intelligent hybrid controller. In addition, a dynamic absorber with a twolevel mass-spring-damper structure was designed and constructed to verify the stability and robustness of a HFNC by numerical simulation and to investigate the control performance by comparing the experimental results of the HFNC with that of a TFC for this MIMO system.     

A hierarchical type-2 fuzzy logic control architecture for autonomous mobile robots

Autonomous mobile robots navigating in changing and dynamic unstructured environments like the outdoor environments need to cope with large amounts of uncertainties that are inherent of natural environments. The traditional type-1 fuzzy logic controller (FLC) using precise type-1 fuzzy sets cannot fully handle such uncertainties. A type-2 FLC using type-2 fuzzy sets can handle such uncertainties to produce a better performance. In this paper, we present a novel reactive control architecture for autonomous mobile robots that is based on type-2 FLC to implement the basic navigation behaviors and the coordination between these behaviors to produce a type-2 hierarchical FLC. In our experiments, we implemented this type-2 architecture in different types of mobile robots navigating in indoor and outdoor unstructured and challenging environments. The type-2-based control system dealt with the uncertainties facing mobile robots in unstructured environments and resulted in a very good performance that outperformed the type-1-based control system while achieving a significant rule reduction compared to the type-1 system.     

Routing with admission control in ATM networks

An optimal control parameter, , was developed by optimal control theory for Asynchronous Transfer Mode (ATM) networks. The parameter is related to link utilization, buffer occupancy, system throughput and cell rejection characteristics and is applied to routing and admission control in simulation of a network carrying multiclass traffic. Simulation results show that the parameter produces better performance with respect to cell rejection rate characteristics compared with the traditional control parameter for utilization, .This research has been supported by the ATI as part of the Navy Contract N00140-94-C-BC03. Some parts of the paper were presented at Globecom'94.