• Proceedings of the KIEE Conference
• /
• 1998.07g
• /
• pp.2193-2196
• /
• 1998

Cellular automata are dynamical systems in which space and time are discrete, where each cell has a finite number of states and updates its states by interactive rules among the cell-neighborhood. From the characteristics of self-reproduction and self- organization, it is possible to create a neural network which has the specific patterns or structures dynamically. CAM-Brain is a kind of such neural network system which evolves its structure by adopting evolutionary computations like genetic algorithms (GA). In this paper, we suggest the evolution strategies for the structure of neural networks based on cellular automata.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.69.2-69
• /
• 2001

This paper presents a new approach to the design of neural control system using digital signal processors in order to improve the precision and robustness. Robotic manipulators have become increasingly important in the field of flexible automation. High speed and high-precision trajectory tracking are indispensable capabilities for their versatile application. The need to meet demanding control requirement in increasingly complex dynamical control systems under significant uncertainties, leads toward design of intelligent manipulation robots. The TMS320C31 is used in implementing real time neural control to provide an enhanced motion control for robotic manipulators. In this control scheme, the networks introduced are neural nets with dynamic neurons, whose dynamics are distributed over all the network nodes.

• Proceedings of the KIEE Conference
• /
• 2000.07d
• /
• pp.2693-2695
• /
• 2000

Generally, neural networks can be used efficiently for the identification and control of nonlinear dynamical system, then it always needs to learn in order that the output values is closed to desired values. But, if plant input to control is limited to certain bounded values, former learning rules has the another problem. This paper demonstrates algorithm to control the bounded-input plant using neural network controller.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.200-200
• /
• 2000

Highly nonlinear dynamical systems are easily identified using neural networks. When disturbances are included in the learning data set Int system modeling, modeling process will be poorly performed. Since the radial basis functions in the radial basis function network(RBFN) are centered at the points specified by the weights, RBF networks are robust for approximating the process including the narrow-band disturbances deviating significantly from the regular signals. To exclude(filter) these disturbances, a robust algorithm for system identification, based on the RBFN, is proposed. The performance of system identification excluding disturbances is investigated and compared with the one including disturbances.

• Korean Chemical Engineering Research
• /
• v.62 no.1
• /
• pp.118-124
• /
• 2024

A neural network is utilized for preprocessing of de-noizing in electrocardiogram signals, retinal images, seismic waves, etc. However, the de-noizing process could provoke increase of computational time and distortion of the original signals. In this study, we investigated a neural network architecture to analyze measurement data without additional de-noizing process. From the dynamical behaviors of DNA in aqueous solution, our neural network model aimed to predict the mole fraction of each DNA in the solution. By adding white noise to the dynamics data of DNA artificially, we investigated the effect of the noise to neural network's predictions. As a result, our model was able to predict the DNA mole fraction with an error of O(0.01) when signal-to-noise ratio was O(1). This work can be applied as a efficient artificial intelligence methodology for analyzing DNA related to genetic disease or cancer cells which would be sensitive to background measuring noise.

• International Journal of Control, Automation, and Systems
• /
• v.2 no.1
• /
• pp.83-91
• /
• 2004

The purpose of this paper is to obtain an accurate nonlinear system model to test various control schemes for a motion control system that requires high speed, robustness and accuracy. An industrial sewing machine equipped with a Brushless DC motor is considered. It is modeled by a neural network that is configured as an output-error dynamical system. The identified model is essentially a one step ahead prediction structure in which past inputs and outputs are used to calculate the current output. Using the model, a 2 degree-of-freedom PID controller to compensate the effects of disturbance without degrading tracking performance has been de-signed. In this experiment, it is not preferable for safety reasons to tune the controller online on the actual machinery. Experimental results confirm that the model is a good approximation of sewing machine dynamics and that the proposed control methodology is effective.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.5 no.1
• /
• pp.33-44
• /
• 2012

A class of recurrent neural networks is proposed and proven to be capable of identifying any discrete-time dynamical system. The applications of the proposed network are addressed in the encoding, identification, and extraction of finite state automata. Simulation results show that the identification of finite state automata using the proposed network, trained by the hybrid greedy simulated annealing with a modified error function in the learning stage, exhibits generally better performance than other conventional identification schemes.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.5
• /
• pp.1130-1137
• /
• 1996

Every neural network has some kind of feedback. For the sake of analyzing fundamental aspects of information processing in neural nets, a net without feedbacks is an important theoretical model. But here we focus on a recurrent neural net which delay synapses as a realistic dynamical model of nervous systems. Synaptic connections are determined by a version of the Hebb rule (correlation type rule). We use a statistical neurodynamic method to explain the retrieval dynamics of the network. The result of the analysis for the sequential associativ e memory with delay synapses is compared with computer simulation. We have succeeded in explaining the dynamics of this network by theoretical analyses.

• Proceedings of the KIEE Conference
• /
• 1995.11a
• /
• pp.161-164
• /
• 1995

This paper introduces an identification model called the Dynamic Neural Network(DNN) with a multilayer neural network in the forward path and a linear dynamical system in the feedback path, and defines Dynamic BackPropagation(DBP) as a learning algorithm for it. This identification model uses the feedback of its own output as a learning signal, which is not affected by a noise added to the output terminal of the plant so, it can be considered as a parallel identification model, and when compared with a series-parallel model which does not use the concept of the feedback, the proposed identification scheme exhibits more robust performance.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1997.11a
• /
• pp.120-123
• /
• 1997

In this paper, we propose the Fuzzy Inference-based Reinforcement Learning Algorithm. We offer more similar learning scheme to the psychological learning of the higher animal's including human, by using Fuzzy Inference in Reinforcement Learning. The proposed method follows the way linguistic and conceptional expression have an effect on human's behavior by reasoning reinforcement based on fuzzy rule. The intervals of fuzzy membership functions are found optimally by genetic algorithms. And using Recurrent state is considered to make an action in dynamical environment. We show the validity of the proposed learning algorithm by applying to the inverted pendulum control problem.