• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.65-68
• /
• 1996

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. 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. The TMS32OC31 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
• /
• 2005.06a
• /
• pp.281-286
• /
• 2005

The main objective of the research work presented in this article is to present a systematic approach for designing a multilayer feed-forward artificial neural network based self-tuning power system stabilizer (ST-ANNPSS). In order to suggest an approach for selecting the number of neurons in the hidden layer, the dynamic performance of the system with ST-ANNPSS is studied and hence compared with that of conventional PSS. Finally the effect of variation of loading condition and equivalent reactance, Xe is investigated on dynamic performance of the system with ST-ANNPSS. Investigations reveal that ANN with one hidden layer comprising nine neurons is adequate and sufficient for ST-ANNPSS. Studies show that the dynamic performance of STANNPSS is quite superior to that of conventional PSS for the loading condition different from the nominal. Also it is revealed that the performance of ST-ANNPSS is quite robust to a wide variation in loading condition.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 2001.04a
• /
• pp.255-260
• /
• 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. 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.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 1999.10a
• /
• pp.113-118
• /
• 1999

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 n 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.

• Molecules and Cells
• /
• v.39 no.12
• /
• pp.841-846
• /
• 2016

Local protein synthesis mediates precise spatio-temporal regulation of gene expression for neuronal functions such as long-term plasticity, axon guidance and regeneration. To reveal the underlying mechanisms of local translation, it is crucial to understand mRNA transport, localization and translation in live neurons. Among various techniques for mRNA analysis, fluorescence microscopy has been widely used as the most direct method to study localization of mRNA. Live-cell imaging of single RNA molecules is particularly advantageous to dissect the highly heterogeneous and dynamic nature of messenger ribonucleoprotein (mRNP) complexes in neurons. Here, we review recent advances in the study of mRNA localization and translation in live neurons using novel techniques for single-RNA imaging.

• The Korean Journal of Physiology and Pharmacology
• /
• v.2 no.1
• /
• pp.9-19
• /
• 1998

The present study was primarily carried out to characterize the properties of the spinomesencephalic tract (SMT) neurons that project from the upper cervical spinal segments to the midbrain. It was also investigated whether these neurons received convergent afferent inputs from other sources in addition to cervical inputs. Extracellular single unit recordings were made from neurons antidromically activated by stimulation of midbrain. Recording sites were located in lamina $I{\sim}VIII\;of\;C1{\sim}C3$ segments of spinal cord. Receptive field (RF) and response properties to mechanical stimulation were studied in 71 SMT neurons. Response profiles were classified into six groups: complex (Comp, n=9), wide dynamic range (WDR, n=16), low threshold (LT, n=5), high threshold (HT, n=6), deep/tap (Deep, n=10), and non- responsive (NR, n=25). Distributions of stimulation and recording sites were not significantly different between SMT groups classified upon their locations and/or response profiles. Mean conduction velocity of SMT neurons was $16.7{\pm}1.28\;m/sec$. Conduction velocities of SMTs recorded in superficial dorsal horn (SDH, n=15) were significantly slower than those of SMTs recorded in deep dorsal horn (DDH, n=18), lateral reticulated area (LRA, n=21), and intermediate zone and ventral horn (IZ/VH, n=15). Somatic RFs for SMTs in LRA and IZ/VH were significantly larger than those in SDH and DDH. Five SMT units (4 Comps and 1 HT) had inhibitory somatic RFs. About half (25/46) of SMT units have their RFs over trigeminal dermatome. Excitabilities of 5/12 cells and 9/13 cells were modulated by stimulation of ipsilateral phrenic nerve and vagus nerve, respectively. These results suggest that upper cervical SMT neurons are heterogenous in their function by showing a wide range of variety in location within the spinal gray matter, in response profile, and in convergent afferent input.

• 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.

• The Korean Journal of Physiology and Pharmacology
• /
• v.3 no.5
• /
• pp.481-490
• /
• 1999

To investigate the changes in the responses of vestibular neurons with time during vestibular compensation, the resting activity and dynamic responses of type I and II neurons in the medial vestibular nuclei to sinusoidal angular acceleration were recorded following unilateral labyrinthectomy (ULX) in Sprague-Dawley rats. The unitary extracellular neuronal activity was recorded from the bilateral medial vestibular nuclei with stainless steel microelectrodes of $3{\sim}5\;M{\Omega}$ before ULX, and 6, 24, 48, 72 hours, and 1 week after ULX under pentobarbital sodium anesthesia (30 mg/kg, i.p.). Gain (spikes/s/deg/s) and phase (in degrees) were determined from the neuronal activity induced by sinusoidal head rotation with 0.05, 0.1, 0.2, and 0.4 Hz. The mean resting activity before ULX was $16.7{\pm}8.6$ spikes/s in type I neurons $(n=67,\;M{\pm}SD)$ and $14.5{\pm}8.4$ spikes/s in type II neurons (n=43). The activities of ipsilateral type I and contralateral type II neurons to the lesion side decreased markedly till 24 hr post-op, and a significant difference between ipsilateral and contralateral type I neurons sustained till 24 hr post-op. The gain at 4 different frequencies of sinusoidal rotation was depressed in all neurons till 6 or 24 hr post-op and then increased with time. The rate of decrease in gain was more prominent in ipsilateral type I and contralateral type II neurons immediately after ULX. Although the gain of those neurons increased gradually after 24 hours, it remained below normal levels. The phase was significantly advanced in all neurons following ULX. These results suggest that a depression of activities in ipsilateral type I and contralateral type II neurons is closely related with the occurrence of vestibular symptoms and restoration of activities in those neurons ameliorates the vestibular symptoms.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.3
• /
• pp.583-588
• /
• 2024

With the emergence of Brain-Computer Interface (BCI) technology, analyzing mirror neurons has become more feasible. However, evaluating the accuracy of BCI systems that rely on human thoughts poses challenges due to their qualitative nature. To harness the potential of BCI, we propose a new approach to measure accuracy based on the characteristics of mirror neurons in the human brain that are influenced by speech speed, depending on the ultimate goal of movement. In Chapter 2 of this paper, we introduce mirror neurons and provide an explanation of human posture estimation for mirror neurons. In Chapter 3, we present a powerful pose estimation method suitable for real-time dynamic environments using the technique of human posture estimation. Furthermore, we propose a method to analyze the accuracy of BCI using this robotic environment.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.7
• /
• pp.675-679
• /
• 2001

In this paper, we present a new competitive learning algorithm called Dynamic Competitive Learning (DCL). DCL is a supervised learning method that dynamically generates output neurons and initializes automatically the weight vectors from training patterns. It introduces a new parameter called LOG (Limit of Grade) to decide whether an output neuron is created or not. If the class of at least one among the LOG number of nearest output neurons is the same as the class of the present training pattern, then DCL adjusts the weight vector associated with the output neuron to learn the pattern. If the classes of all the nearest output neurons are different from the class of the training pattern, a new output neuron is created and the given training pattern is used to initialize the weight vector of the created neuron. The proposed method is significantly different from the previous competitive learning algorithms in the point that the selected neuron for learning is not limited only to the winner and the output neurons are dynamically generated during the learning process. In addition, the proposed algorithm has a small number of parameters, which are easy to be determined and applied to real-world problems. Experimental results for pattern recognition of remote sensing data and handwritten numeral data indicate the superiority of DCL in comparison to the conventional competitive learning methods.