• Proceedings of the KIEE Conference
• /
• 2009.04a
• /
• pp.216-218
• /
• 2009

In this paper, a real time monitoring and management system being operated in the ubiquitous environment was developed to monitor leakage current, load current, and arc-fault, and an electrical safety network for reasonable management of electrical risk factor was proposed. For confirmation of usefulness and reliability of the proposed safety network and system, the developed intelligent panels were applied to 28 Korean traditional houses in Jeonjoo city, and the network including the panels was operated. If the National Electrical Safety Network is completely constructed in the houses of general electrical users, the network will have an effect on that a main manager transfers from general people to expert. As a result, the electrical fires caused by an over-load, an arc-fault, and an earth-fault will be prevented.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.2 no.4
• /
• pp.221-226
• /
• 2007

This paper propose a control network system based on IEEE1394 for a multi body robot control. The IEEE1394 has the characteristic of high speed(400Mbps), real-time, stability and plug&play. And IEEE1394 also supports freeform daisy chaining, branching and hot plugging, which reduce cabling complexity and make a system simple. Especially, multi host and broad casting support network data sharing method which is suitable for control network for multi body robot. Through experiment, we show that the proposed control network can interface 48 joints (BLDC motors, gears, and encoders) and four 6-axis force/torque sensors with 4Khz communication bandwidth, which is adequate for a multi body robot.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.31 no.6 s.261
• /
• pp.651-658
• /
• 2007

A new tree-structured neural network classifier is proposed for the automatic real-time inspection of cold-rolled steel strip surface defects. The defects are classified into 3 groups such as area type, disk type, area & line type in the first stage of the tree-structured neural network. The defects are classified in more detail into 11 major defect types which are considered as serious defects in the second stage of neural network. The tree-structured neural network classifier consists of 4 different neural networks and optimum features are selected for each neural network classifier by using SFFS algorithm and correlation test. The developed classifier demonstrates very plausible result which is compatible with commercial products having high world-wide market shares.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.52 no.8
• /
• pp.471-481
• /
• 2003

The acts of biological immune system are similar to the navigation for autonomous mobile robots under dynamically changing environments. In recent years, many researchers have studied navigation algorithms using artificial immune networks. Conventional artificial immune algorithms consist of an obstacle-avoidance behavior and a goal-reaching behavior. To select a proper action, the navigation algorithm should combine the obstacle-avoidance behavior with the goal-reaching behavior. In this paper, the neural network is employed to combine the behaviors. The neural network is trained with the surrounding information. the outputs of the neural network are proper combinational weights of the behaviors in real-time. Also, a velocity control algorithm is constructed with the artificial immune network. Through a simulation study and experimental results for a autonomous mobile robot, we have shown the validity of the proposed navigation algorithm.

• Asia pacific journal of information systems
• /
• v.1 no.1
• /
• pp.63-86
• /
• 1991

In this paper, our concern is the artificial neural network-based patten classification, when can resolve the difficulties in the Autoregressive Moving Average(ARMA) model identification problem To effectively classify a time series into an approriate ARMA model, we adopt the Multi-layered Backpropagation Network (MLBPN) as a pattern classifier, and Extended Sample Autocorrelation Function (ESACF) as a feature extractor. To improve the classification power of MLBPN's we suggest an integrated neural network system which consists of an AR Network and many small-sized MA Networks. The output of AR Network which will gives the MA order. A step-by-step training strategy is also suggested so that the learned MLBPN's can effectively ESACF patterns contaminated by the high level of noises. The experiment with the artificially generated test data and real world data showed the promising results. Our approach, combined with a statistical parameter estimation method, will provide a way to the automation of ARMA modeling.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.17 no.2
• /
• pp.61-72
• /
• 2007

With the increasing growth of Internet and wireless IP networks, Multimedia systems need to be envisaged as information resources where users can access anywhere and anytime. However, efficient services in these multimedia systems are open and challenging research problem due to user mobility, limited resources in wireless devices and expensive radio bandwidth. To implement multimedia services over heterogeneous network, the IP header compression scheme can be used for saving bandwidth. In this paper, we present an efficient solution for header compression, which is modified form of ECRTP. It shows an architectural framework adopting modified ECRTP when IP tunneling network using GRE over IPSec is implemented. We have conducted simulations in order to analyze the effects of different header compression techniques while delivering real-time services to the wireless access network through secured IP Network. The impacts on performance have been investigated through a series of experiments.

• International Journal of Internet, Broadcasting and Communication
• /
• v.13 no.1
• /
• pp.210-218
• /
• 2021

CNN (Convolutional Neural Network), a type of deep learning algorithm, is a type of artificial neural network used to analyze visual images. In deep learning, it is classified as a deep neural network and is most commonly used for visual image analysis. Accordingly, an AI autonomous driving model was constructed through real-time image processing, and a crosswalk image of a road was used as an obstacle. In this paper, we proposed a low-cost model that can actually implement autonomous driving based on the CNN model. The most well-known deep neural network technique for autonomous driving is investigated and an end-to-end model is applied. In particular, it was shown that training and self-driving on a simulated road is possible through a practical approach to realizing lane detection and keeping.

• Journal of Korea Multimedia Society
• /
• v.20 no.8
• /
• pp.1421-1429
• /
• 2017

Human activity recognition has been studied using various sensors and algorithms. Human activity recognition can be divided into sensor based and vision based on the method. In this paper, we proposed an activity recognition system using acceleration sensor and gyroscope sensor in smartphone among sensor based methods. We used Deep Belief Network (DBN), which is one of the most popular deep learning methods, to improve an accuracy of human activity recognition. DBN uses the entire input set as a common input. However, because of the characteristics of different time window depending on the type of human activity, the RBMs, which is a component of DBN, are configured hierarchically by combining them from different time windows. As a result of applying to real data, The proposed human activity recognition system showed stable precision.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.1022-1027
• /
• 1991

A new method to control a robot manipulator by neural networks is proposed. The controller is composed of both a PD controller and a neural network-based feedforward controller. MLP(multi-layer perceptron) neural network is used for the feedforward controller and trained by BP(back-propagation) learning rule. Error terms for BP learning rule are composed of the outputs of a PD controller and the acceleration errors of manipulator joints. We compare the proposed method with existing ones and contrast performances of them by simulation. Also, We discuss the real application of the proposed method in consideration of the learning time of the neural network and the time required for sensing the joint acceleration.

• 제어로봇시스템학회:학술대회논문집
• /
• 1992.10a
• /
• pp.55-60
• /
• 1992

The IEEE 802.4 priority mechanism can be used to handle multiple data access classes of traffic. Several timers are used to realize the priority mechanism. The performance and stability of a token bus network depend on the assignment of such timers. In this peper, we present a dynamic timer assignment algorithm for the token passing bus network. The presented algorithm has simple structure for real-time applications and adaptively controls the set of initial timer values according to the offered traffic load. The assignment of the set of timers becomes easy due to the presented algorithm. Based on the iterative algorithm, some solutions such as mean waiting time are derived.