• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.1173-1176
• /
• 1996

Multilayer feedforward networks may be applied to identify the deterministic relationship between input and output data. When the results from the network require a high level of assurance, considering of the stochastic relationship between the data may be very important. The variance is one of the useful parameters to represent the stochastic relationship. This paper presents a new algorithm for a multilayer feedforward network to learn the variance of dispersed data without preliminary calculation of variance. In this paper, the network with this learning algorithm is named as a variance learning neural network(VALEAN). Computer simulation examples are utilized for the demonstration and the evaluation of VALEAN.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.1
• /
• pp.251-268
• /
• 1996

In this paper, a higher-order feedforward neural network called ridge polynomial network (RPN) which shows good approximation capability for nonlnear continuous functions defined on compact subsets in multi-dimensional Euclidean spaces, is presented. This network provides more efficient and regular structure as compared to ordinary higher-order feedforward networks based on Gabor-Kolmogrov polynomial expansions, while maintating their fast learning property. the ridge polynomial network is a generalization of the pi-sigma network (PSN) and uses a specialform of ridge polynomials. It is shown that any multivariate polynomial can be exactly represented in this form, and thus realized by a RPN. The approximation capability of the RPNs for arbitrary continuous functions is shown by this representation theorem and the classical weierstrass polynomial approximation theorem. The RPN provides a natural mechanism for incremental function approximation based on learning algorithm of the PSN. Simulation results on several applications such as multivariate function approximation and pattern classification assert nonlinear approximation capability of the RPN.

• Journal of Institute of Control, Robotics and Systems
• /
• v.10 no.12
• /
• pp.1295-1304
• /
• 2004

In this paper, the Bayesian recurrent neural network is proposed to predict time series data. A neural network predictor requests proper learning strategy to adjust the network weights, and one needs to prepare for non-linear and non-stationary evolution of network weights. The Bayesian neural network in this paper estimates not the single set of weights but the probability distributions of weights. In other words, the weights vector is set as a state vector of state space method, and its probability distributions are estimated in accordance with the particle filtering process. This approach makes it possible to obtain more exact estimation of the weights. In the aspect of network architecture, it is known that the recurrent feedback structure is superior to the feedforward structure for the problem of time series prediction. Therefore, the recurrent neural network with Bayesian inference, what we call Bayesian recurrent neural network (BRNN), is expected to show higher performance than the normal neural network. To verify the proposed method, the time series data are numerically generated and various kinds of neural network predictor are applied on it in order to be compared. As a result, feedback structure and Bayesian learning are better than feedforward structure and backpropagation learning, respectively. Consequently, it is verified that the Bayesian reccurent neural network shows better a prediction result than the common Bayesian neural network.

• Journal of the Korean Society for Precision Engineering
• /
• v.29 no.1
• /
• pp.79-86
• /
• 2012

This paper proposes PID and RIC (Robust Internal-loop Compensator) based motion controller using dual learning algorithm for position control of linear synchronous motor respectively. Its gains are auto-tuned by using two learning algorithms, reinforcement learning and neural network. The feedback controller gains are tuned by reinforcement learning, and then the feedforward controller gains are tuned by neural network. Experiments prove the validity of dual learning algorithm. The RIC controller has better performance than does the PID-feedforward controller in reducing tracking error and disturbance rejection. Neural network shows its ability to decrease tracking error and to reject disturbance in the stop range of the target position and home.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.28 no.8 s.227
• /
• pp.984-988
• /
• 2004

This study presents the process of the solenoid parameter estimation of an common-rail injector fer HSDI(High Speed Direct Injection) diesel engines. The EMF(Electromotive Force) and solenoid inductance are the major parameters for presenting the injector dynamics, and also these parameters are estimated by using a multi-layer feedforward artificial neural networks(ANN). The performances of parameter estimators are verified by the simulation with injector model. The feasibility of this methodology is closely examined through the simulation in the various operating points of injector. The simulation results have revealed that estimated parameters show favorable agreements with the common-rail injector model.

• Journal of Advanced Marine Engineering and Technology
• /
• v.20 no.4
• /
• pp.11-19
• /
• 1996

Recently, fuzzy and neural network techniques have been successfully applied to control of complex and ill-defined system in a wide variety of areas, such as robot, water purification, automatic train operation system and automatic container crane operation system, etc. In this paper, we present a neuro-fuzzy controller which unifies both fuzzy logic and multi-layered feedforward neural networks. Fuzzy logic provides a means for converting linguistic control knowledge into control actions. On the other hand, feedforward neural networks provide salient features, such as learning and parallelism. In the proposed neuro-fuzzy controller, the parameters of membership functions in the antecedent part of fuzzy inference rules are identified by using the error backpropagation algorithm as a learning rule, while the coefficients of the linear combination of input variables in the consequent part are determined by using the least square estimation method. Finally, the effectiveness of the proposed controller is verified through computer simulation of an inverted pendulum system.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.230-235
• /
• 1992

We propose a method of generating data to train a neural network controller. The data can be prepared directly by an iterative learning technique which repeatedly adjusts the control input to improve the tracking quality of the desired trajectory. Instead of storing control input data in memory as in iterative learning control, the neural network stores the mapping between the control input and the desired output. We apply this concept to the trajectory control of a two link robot manipulator with a feedforward neural network controller and a feedback linear controller. Simulation results show good generalization of the neural network controller.

• The Transactions of the Korean Institute of Electrical Engineers B
• /
• v.49 no.8
• /
• pp.557-564
• /
• 2000

A very simple control approach using neural network for the robust position control of a Permanent Magnet Synchronous Motor(PMSM) is presented. The linear quadratic controller plus feedforward neural network is employed to obtain the robust PMSM system approximately linearized using field-orientation method for an AC servo. The neural network is trained in on-line phases and this neural network is composed by a feedforward recall and error back-propagation training. Since the total number of nodes are only eight, this system can be easily realized by the general microprocessor. During the normal operation, the input-output response is sampled and the weighting value is trained multi-times by error back-propagation method at each sample period to accommodate the possible variations in the parameters or load torque. In addition, the robustness is also obtained without affecting overall system response. This method is realized by a floating-point Digital Signal Processor DS1102 Board (TMS320C31).

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10a
• /
• pp.7-12
• /
• 1990

In this paper, we propose a new manipulator control scheme based on the CMAG neural network. The proposed control consists of two components. The feedforward component is an output of trained CMAC neural network and the feedback component is a modified sliding mode control. The CMAC accepts the position, velocity and acceleration of manipulator as input and outputs two values for the controller : One is the nominal torque used for feedforward compensation(M1 network) and the other is the inertia matrix related information used for the feedback component(M2 network). Since the used control algorithm guarantees the robust trajectory tracking in spite of modeling errors, the CMAC mapping errors due to the memory limitation are little worth consideration.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.223-226
• /
• 1997

Usually, the Evolutionary Algorithms(EAs) are considered more efficient for optimal, system design because EAs can provide higher opportunity for obtaining the global optimal solution. This paper presents a mechanism of co-evolution consists of the two genetic algorithms(GAs). This mechanism includes host populations and parasite populations. These two populations are closely related to each other, and the parasite populations plays an important role of searching for useful schema in host populations. Host population represented by feedforward neural network and the result of co-evolution we will find the optimal structure of the neural network. We used the genetic algorithm that search the structure of the feedforward neural network, and evolution strategies which train the weight of neuron, and optimize the net structure. The validity and effectiveness of the proposed method is exemplified on the stabilization and position control of the inverted-pendulum system.