• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1996.11a
• /
• pp.759-763
• /
• 1996

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. The nets are trained by the distributed dynamic back propagation algorithm. The proposed neural network control scheme is simple in structure, fast in computation, and suitable for implementation of real-time control. Performance of the neural controller is illustrated by simulation and experimental results for a SCARA robot.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.215-218
• /
• 1997

A dynamic back-propagation neural network is addressed for adaptive neural control system to approximate non-linear control system rather than static networks. It has the capability to represent the approximation of nonlinear system without mathematical analysis and to carry out the on-line learning algorithm for real time application. The simulated results show fast tracking capability and adaptive response by using dynamic back-propagation neurons.

• The Korean Journal of Physiology
• /
• v.24 no.2
• /
• pp.389-402
• /
• 1990

The physiological characteristics of the neurons receiving the ventral root afferent inputs were investigated in the cat. A total of 70 cells were identified in the lumbosacral spinal cord. All these cells responded only to the C-strength stimulation of the distal stump of cut ventral root and the estimated conduction velocities of the VRA fibers were not faster than 4 m/sec. The majority of them were silent in resting state. For 49 cells, their peripheral receptive fields were characterized. Among them, 25 cells were exclusively excited by VRA inputs, 8 were inhibited and the remaining cells recevied both excitatory and inhibitory VRA inputs. According to the response pattern to the mechanical stimuli applied to their receptive fields, only a fourth of them were typical high threshold cell, a sixth, wide dynamic range cells, while remainings were a rather complex cells. Most of the cells receiving VRA inputs, received only the A ${\delta}-peripheral$ nerve inputs. Intravenous injection of morphine decreased the response of spinal cells to the VRA activation. The responses were abolished completely by counter irritation to the common peroneal nerve with C-strength-low frequency stimuli. These physiological properties of the spinal neurons receiving the VRA inputs are differ in some aspect from the spinal neurons receiving nociceptive inputs from the periphery, but still were consistent with the contention that VRA system might carry nociceptive informations arising from the spinal cord and/or neraby surrounding tissues.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1997.10a
• /
• pp.413-416
• /
• 1997

Recently, dynamic recurrent neural networks(DRNN) for identification of nonlinear dynamic systems have been researched extensively. In general, dynamic backpropagation was used to adjust the weights of neural networks. But, this method requires many complex calculations and has the possibility of falling into a local minimum. So, we propose a new approach to identify nonlinear dynamic systems using DRNN. In order to adjust the weights of neurons, we use evolution strategies, which is a method used to solve an optimal problem having many local minimums. DRNN trained by evolution strategies with mutation as the main operator can act as a plant emulator. And the fitness function of evolution strategies is based on the difference of the plant's outputs and DRNN's outputs. Thus, this new approach at identifying nonlinear dynamic system, when applied to the simulation of a two-link robot manipulator, demonstrates the performance and efficiency of this proposed approach.

• 제어로봇시스템학회:학술대회논문집
• /
• 1995.10a
• /
• pp.60-64
• /
• 1995

The conventional neural network models are a parody of biological neural structures, and have very slow learning. In order to emulate some dynamic functions, such as learning and adaption, and to better reflect the dynamics of biological neurons, M.M. Gupta and D.H. Rao have developed a 'dynamic neural model'(DNU). Proposed neural unit model is to introduce some dynamics to the neuron transfer function, such that the neuron activity depends on internal states. Integrating an dynamic elementry processor within the neuron allows the neuron to act dynamic response Numerical examples are presented for a model system. Those case studies showed that the proposed DNU is so useful in practical sense.

• The Korean Journal of Physiology and Pharmacology
• /
• v.1 no.3
• /
• pp.241-249
• /
• 1997

The aim of the present study is to examine the brainstem sites where the electrical stimulation produces a suppression of dorsal horn neuron responses of neuropathic rats. An experimental neuropathy was induced by a unilateral ligation of L5-L6 spinal nerves of rats. Ten to 15 days after surgery, the spinal cord was exposed and single-unit recording was made on wide dynamic range (WDR) neurons in the dorsal horn. Neuronal responses to mechanical stimuli applied to somatic receptive fields were examined to see if they were modulated by electrical stimulation of various brainstem sites. Electrical stimulation of periaqueductal gray (PAG), n. raphe magnus (RMg) or n. reticularis gigantocellularis (Gi) significantly suppressed responses of WDR neurons -to both noxious and non-noxious stimuli. Electrical stimulation of other brainstem areas, such as locus coeruleus. (LC) and n. reticularis paragigantocellularis lateralis (LPGi), produced little or no suppression. Microinjection of morphine into PAG, RMg, or Gi also produced a suppression as similar pattern to the case of electrical stimulation, whereas morphine injection into LC or LPGi exerted no effects. The results suggest that PAG, NRM and Gi are the principle brainstem nuclei involved in the descending inhibitory systems responsible for the control of neuropathic pain. These systems are likely activated by endogenous opioids and exert their inhibitory effect by acting on WDR neurons in the spinal cord.

• The Korean Journal of Physiology and Pharmacology
• /
• v.4 no.1
• /
• pp.15-24
• /
• 2000

Although nociceptive informations are thought to be processed via different neural mechanisms depending on the types of stimuli, sufficient data have not been accumulated yet. We performed a series of experiments to elucidate the possible neural mechanisms as to chemical stimuli such as formalin, capsaicin and ATP. Single unit activity of wide dynamic range (WDR) neurons and high threshold cells were recorded extracellularly from the lumbosacral enlargement of cat spinal cord before and after chemical stimulation to its receptive field (RF). Each chemical substance - formalin $(20{\mu}l,\;4%),$ capsaicin (33 mM) or Mg-ATP (5 mM)- was injected intradermally into the RFs and then the changes in the spontaneous activity, mechanical threshold and responses to the peripheral mechanical stimuli were observed. In many cases, intradermal injection of formalin (5/11) and capsaicin (8/11) resulted in increase of the spontaneous activity with a biphasic pattern, whereas ATP (8/8) only showed initial responses. Time courses of the biphasic pattern, especially the late response, differed between formalin and capsaicin experiments. One hour after injection of each chemical (formalin, capsaicin, or ATP), the responses of the dorsal horn neurons to mechanical stimuli increased at large and the RFs were expended, suggesting development of hypersensitization (formalin 6/10, capsaicin 8/11, and ATP 15/19, respectively). These results are suggested that formalin stimulates peripheral nociceptor, local inflammation and involvement of central sensitization, capsaicin induces central sensitization as well as affects the peripheral C-polymodal nociceptors and neurogenic inflammation, and ATP directly stimulates peripheral nociceptors.

• The Korean Journal of Physiology and Pharmacology
• /
• v.2 no.2
• /
• pp.155-163
• /
• 1998

The purpose of present study was to compare the effects of somatostatin (SOM) and morphine (Mor) on the responses of wide dynamic range (WDR) cells to peripheral noxious stimulation. Single neuronal activity was recorded with a carbon-filament electrode at the lumbosacral enlargement of cat spinal cord. After identifying WDR cells, their responses to peripheral noxious mechanical or thermal stimuli were characterized and the effects of SOM and Mor, applied either iontophoretically or intrathecally, were studied. In most cells SOM and Mor suppressed noxious stimulus-evoked WDR neuronal activity, though a few WDR neurons showed no change or were excited by SOM and Mor. Systemically applied naloxone, a non-specific opioid antagonist, always reversed the Mor induced suppression of neuronal activity evoked by noxious mechanical stimuli, but did not always reverse the suppression of neuronal activity elicited by SOM. The suppressive effect of Mor on thermal stimulus-evoked neuronal activity was partially reversed by naloxone, while that of SOM were not reversed at all. The above results suggest that both Mor and SOM exert an inhibitory effect on thermal and mechanical stimulus-evoked WDR neuronal activity in cat spinal dorsal horn, but the mechanisms are dependent upon the functional populations of dorsal horn nociceptive neurons.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.8 no.5
• /
• pp.29-38
• /
• 1998

This paper proposes the neural network self-tuning PID controller for the load frequency control of 2- areas power system, namely, the prompt convergence of frequency and tie-line power flow deviation. The neural network applied to computer simulation consists of neurons of two inputs, ten hiddens and tliree outputs layer. Neurons of two inputs layer receive the error and its change rate of the system and cutputs layer consists of three neurons for the parameters of the PID controller. The simulation results shows that the proposed neural network self-tuning PID controller is superior to conventional control t~:chniques(Optimal, PID) in dynamic response and control performance.

• Journal of Magnetics
• /
• v.19 no.4
• /
• pp.315-322
• /
• 2014

In this work, Artificial Neural Network (ANN) was used to model the dynamic behavior of ferromagnetic hysteresis derived from performing the mean-field analysis on the Ising model. The effect of field parameters and system structure (via coordination number) on dynamic critical points was elucidated. The Ising magnetization equation was drawn from mean-field picture where the steady hysteresis loops were extracted, and series of the dynamic critical points for constructing dynamic phase-diagram were depicted. From the dynamic critical points, the field parameters and the coordination number were treated as inputs whereas the dynamic critical temperature was considered as the output of the ANN. The input-output datasets were divided into training, validating and testing datasets. The number of neurons in hidden layer was varied in structuring ANN network with highest accuracy. The network was then used to predict dynamic critical points of the untrained input. The predicted and the targeted outputs were found to match well over an extensive range even for systems with different structures and field parameters. This therefore confirms the ANN capabilities and indicates the ANN ability in modeling the ferromagnetic dynamic hysteresis behavior for establishing the dynamic-phase-diagram.