Scheduled System Maintenance:
Some services will be unavailable Sunday, March 29th through Monday, March 30th. We apologize for the inconvenience.
By Topic

Systems, Man, and Cybernetics, Part B: Cybernetics, IEEE Transactions on

Issue 2 • Date Apr 2000

Filter Results

Displaying Results 1 - 12 of 12
  • GA-based fuzzy reinforcement learning for control of a magnetic bearing system

    Publication Year: 2000 , Page(s): 276 - 289
    Cited by:  Papers (22)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (324 KB)  

    This paper proposes a TD (temporal difference) and GA (genetic algorithm)-based reinforcement (TDGAR) learning method and applies it to the control of a real magnetic bearing system. The TDGAR learning scheme is a new hybrid GA, which integrates the TD prediction method and the GA to perform the reinforcement learning task. The TDGAR learning system is composed of two integrated feedforward networks. One neural network acts as a critic network to guide the learning of the other network (the action network) which determines the outputs (actions) of the TDGAR learning system. The action network can be a normal neural network or a neural fuzzy network. Using the TD prediction method, the critic network can predict the external reinforcement signal and provide a more informative internal reinforcement signal to the action network. The action network uses the GA to adapt itself according to the internal reinforcement signal. The key concept of the TDGAR learning scheme is to formulate the internal reinforcement signal as the fitness function for the GA such that the GA can evaluate the candidate solutions (chromosomes) regularly, even during periods without external feedback from the environment. This enables the GA to proceed to new generations regularly without waiting for the arrival of the external reinforcement signal. This can usually accelerate the GA learning since a reinforcement signal may only be available at a time long after a sequence of actions has occurred in the reinforcement learning problem. The proposed TDGAR learning system has been used to control an active magnetic bearing (AMB) system in practice. A systematic design procedure is developed to achieve successful integration of all the subsystems including magnetic suspension, mechanical structure, and controller training. The results show that the TDGAR learning scheme can successfully find a neural controller or a neural fuzzy controller for a self-designed magnetic bearing system View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A two-level hybrid evolutionary algorithm for modeling one-dimensional dynamic systems by higher-order ODE models

    Publication Year: 2000 , Page(s): 351 - 357
    Cited by:  Papers (4)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (200 KB)  

    This paper presents a new algorithm for modeling one-dimensional (1-D) dynamic systems by higher-order ordinary differential equation (HODE) models instead of the ARMA models as used in traditional time series analysis. A two-level hybrid evolutionary modeling algorithm (THEMA) is used to approach the modeling problem of HODE's for dynamic systems. The main idea of this modeling algorithm is to embed a genetic algorithm (GA) into genetic programming (GP), where GP is employed to optimize the structure of a model (the upper level), while a GA is employed to optimize the parameters of the model (the lower level). In the GA, we use a novel crossover operator based on a nonconvex linear combination of multiple parents which works efficiently and quickly in parameter optimization tasks. Two practical examples of time series are used to demonstrate the THEMA's effectiveness and advantages View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Semantic coherency: the basis of an image interpretation device-application to the cadastral map interpretation

    Publication Year: 2000 , Page(s): 322 - 338
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (964 KB)  

    This paper describes the formal architecture of a device capable of interpreting technical and cartographic documents. Our approach is two-fold, based on a model of the document and on the implementation of a set of “builders”, the aim of which is to progressively construct information of as high a semantic level as that provided by the document drawer. The implementation support is the French cadaster. The interpretation process is performed in two stages: the first one consists in constructing the information, through a bottom-up approach. Then, this information is analyzed and the set of objects considered as nonvalid (“inconsistent”) with regard to the document model are reexamined through a cycling stage. The whole approach is presented, and the first implementation has enabled us to quantify the interpretation results and to verify the relevance of the cycling stage View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A robust position/force learning controller of manipulators via nonlinear H∞ control and neural networks

    Publication Year: 2000 , Page(s): 310 - 321
    Cited by:  Papers (17)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (312 KB)  

    A new robust learning controller for simultaneous position and force control of uncertain constrained manipulators is presented. Using models of the manipulator dynamics and environmental constraint, a task-space reduced-order position dynamics and an algebraic description for the interacting force between the manipulator and its environment are constructed. Based on this treatment, the robust nonlinear H∞ control approach and direct adaptive neural network (NN) technique are then integrated together. The role of NN devices is to adaptively learn those manipulators' structured/unstructured uncertain dynamics as well as the uncertainties with environmental modelling. Then, the effects on tracking performance attributable to the approximation errors of NN devices are attenuated to a prescribed level by the embedded nonlinear H∞ control. Whenever the adopted NN devices have the potential to effectively approximate those nonlinear mappings which are to be learned, then this new control scheme can be ultimately less conservative than its counterpart H∞ only position/force tracking control scheme. This is shown analytically in the form of theorem. Finally, a simulation study for a constrained two-link planar manipulator is given. Simulation results indicate that the proposed adaptive H∞ NN position/force tracking controller performs better in both force and position tracking tasks than its counterpart H∞ only position/force tracking control scheme View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Dynamic fuzzy neural networks-a novel approach to function approximation

    Publication Year: 2000 , Page(s): 358 - 364
    Cited by:  Papers (118)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (220 KB)  

    In this paper, an architecture of dynamic fuzzy neural networks (D-FNN) implementing Takagi-Sugeno-Kang (TSK) fuzzy systems based on extended radial basis function (RBF) neural networks is proposed. A novel learning algorithm based on D-FNN is also presented. The salient characteristics of the algorithm are: 1) hierarchical on-line self-organizing learning is used; 2) neurons can be recruited or deleted dynamically according to their significance to the system's performance; and 3) fast learning speed can be achieved. Simulation studies and comprehensive comparisons with some other learning algorithms demonstrate that a more compact structure with higher performance can be achieved by the proposed approach View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Temperature prediction using fuzzy time series

    Publication Year: 2000 , Page(s): 263 - 275
    Cited by:  Papers (88)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (340 KB)  

    A drawback of traditional forecasting methods is that they can not deal with forecasting problems in which the historical data are represented by linguistic values. Using fuzzy time series to deal with forecasting problems can overcome this drawback. In this paper, we propose a new fuzzy time series model called the two-factors time-variant fuzzy time series model to deal with forecasting problems. Based on the proposed model, we develop two algorithms for temperature prediction. Both algorithms have the advantage of obtaining good forecasting results View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Nonstationary time series analysis by temporal clustering

    Publication Year: 2000 , Page(s): 339 - 343
    Cited by:  Papers (9)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (160 KB)  

    The object of this paper is to present a model and a set of algorithms for estimating the parameters of a nonstationary time series generated by a continuous change in regime. We apply fuzzy clustering methods to the task of estimating the continuous drift in the time series distribution and interpret the resulting temporal membership matrix as weights in a time varying, mixture probability distribution function (PDF). We analyze the stopping conditions of the algorithm to infer a novel cluster validity criterion for fuzzy clustering algorithms of temporal patterns. The algorithm performance is demonstrated with three different types of signals View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Model-based recurrent neural network for modeling nonlinear dynamic systems

    Publication Year: 2000 , Page(s): 344 - 351
    Cited by:  Papers (7)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (232 KB)  

    A model-based recurrent neural network (MBRNN) is introduced for modeling dynamic systems. This network has a fixed structure that is defined according to the linearized state-space model of the plant. Therefore, the MBRNN has the ability to incorporate the analytical knowledge of the plant in its formulation. With its original topology intact, the MBRNN can then be trained to represent the plant nonlinearities through modifying its nodes' activation functions, which consist of contours of Gaussian radial basis functions (RBFs). Training in MBRNN involves adjusting the weights of the RBF's so as to modify the contours representing the activation functions. The performance of the MBRNN is demonstrated via several examples. The results indicate that it requires much shorter training than needed by ordinary recurrent networks. This efficiency in training is attributed to the MBRNN's fixed topology which is independent of training View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Investigating a relevance of fuzzy mappings

    Publication Year: 2000 , Page(s): 249 - 262
    Cited by:  Papers (3)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (536 KB)  

    The study introduces a concept of relevance of fuzzy mappings regarded as fundamental constructs of granular computing and rule-based systems, in particular. The notion of relevance of the fuzzy mappings is instrumental in the quantification of the quality of such mappings prior to their detailed construction. For the purposes of such quantification, we introduce shadowed sets and discuss as an algorithmic framework to be instrumental in expressing and quantifying the property of relevance of the fuzzy mappings. It is revealed that shadowed sets provide an interesting three-valued quantification of this property (such as acceptable mapping, marginal mapping, and a lack of mapping). The paper includes a number of detailed calculations concerning two commonly exploited classes of triangular and Gaussian fuzzy sets. Numerical studies are discussed as well View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Design and stability analysis of single-input fuzzy logic controller

    Publication Year: 2000 , Page(s): 303 - 309
    Cited by:  Papers (39)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (168 KB)  

    In existing fuzzy logic controllers (FLCs), input variables are mostly the error and the change-of-error regardless of complexity of controlled plants. Either control input u or the change of control input Δu is commonly used as its output variable. A rule table is then constructed on a two-dimensional (2-D) space. This scheme naturally inherits from conventional proportional-derivative (PD) or proportional-integral (PI) controller. Observing that 1) rule tables of most FLCs have skew-symmetric property and 2) the absolute magnitude of the control input |u| or |Δu| is proportional to the distance from its main diagonal line in the normalized input space, we derive a new variable called the signed distance, which is used as a sole fuzzy input variable in our simple FLC called single-input FLC (SFLC). The SFLC has many advantages: The total number of rules is greatly reduced compared to existing FLCs, and hence, generation and tuning of control rules are much easier. The proposed SFLC is proven to be absolutely stable using Popov criterion. Furthermore, the control performance is nearly the same as that of existing FLCs, which is revealed via computer simulations using two nonlinear plants View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Tracking control of a rolling disk

    Publication Year: 2000 , Page(s): 364 - 372
    Cited by:  Papers (1)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (392 KB)  

    The tracking control of a disk rolling without slipping on the horizontal (X, Y)-plane is considered. The motion of the disk can be controlled via a tilting torque and a pedaling torque. The concept of path controllability of the disk is introduced and then used to calculate control laws such that the disk tracks a given path in the (X, Y)-plane View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Genetic reinforcement learning through symbiotic evolution for fuzzy controller design

    Publication Year: 2000 , Page(s): 290 - 302
    Cited by:  Papers (82)
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (244 KB)  

    An efficient genetic reinforcement learning algorithm for designing fuzzy controllers is proposed in this paper. The genetic algorithm (GA) adopted in this paper is based upon symbiotic evolution which, when applied to fuzzy controller design, complements the local mapping property of a fuzzy rule. Using this Symbiotic-Evolution-based Fuzzy Controller (SEFC) design method, the number of control trials, as well as consumed CPU time, are considerably reduced when compared to traditional GA-based fuzzy controller design methods and other types of genetic reinforcement learning schemes. Moreover, unlike traditional fuzzy controllers, which partition the input space into a grid, SEFC partitions the input space in a flexible way, thus creating fewer fuzzy rules. In SEFC, different types of fuzzy rules whose consequent parts are singletons, fuzzy sets, or linear equations (TSK-type fuzzy rules) are allowed. Further, the free parameters (e.g., centers and widths of membership functions) and fuzzy rules are all tuned automatically. For the TSK-type fuzzy rule especially, which put the proposed learning algorithm in use, only the significant input variables are selected to participate in the consequent of a rule. The proposed SEFC design method has been applied to different simulated control problems, including the cart-pole balancing system, a magnetic levitation system, and a water bath temperature control system. The proposed SEFC has been verified to be efficient and superior from these control problems, and from comparisons with some traditional GA-based fuzzy systems View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.

Aims & Scope

IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics focuses on cybernetics, including communication and control across humans, machines and organizations at the structural or neural level

 

This Transaction ceased production in 2012. The current retitled publication is IEEE Transactions on Cybernetics.

Full Aims & Scope

Meet Our Editors

Editor-in-Chief
Dr. Eugene Santos, Jr.
Thayer School of Engineering
Dartmouth College