• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.6
• /
• pp.567-572
• /
• 2019

In this paper, we propose an algorithm to perform convolution, pooling, and ReLU operations in CNN using binary image and binary kernel. It decomposes 256 gray-scale images into 8 bit planes and uses a binary kernel consisting of -1 and 1. The convolution operation of binary image and binary kernel is performed by addition and subtraction. Logically, it is a binary operation algorithm using the XNOR and comparator. ReLU and pooling operations are performed by using XNOR and OR logic operations, respectively. Through the experiments to verify the usefulness of the proposed algorithm, We confirm that the CNN operation can be performed by converting it to binary logic operation. It is an algorithm that can implement deep running even in a system with weak computing power. It can be applied to a variety of embedded systems such as smart phones, intelligent CCTV, IoT system, and autonomous car.

• Journal of Advanced Navigation Technology
• /
• v.22 no.3
• /
• pp.213-219
• /
• 2018

In this paper, we propose a design method and process for hardware and software of hybrid V2X communication module that supports both C-ITS communication protocol designed for vehicle environment and Legacy LTE communication technology. C-ITS is suitable for safety service applications due to its low latency characteristics, and Legacy LTE is a technology suitable for non-safety applications such as traffic information and infotainment due to high latency and high capacity. The hybrid V2X communication module supports multiple communication technologies of WAVE and LTE, in which WAVE supports multiple channels, so that it is designed to transmit road information such as LDM and positioning correction information to an autonomous vehicle in real time. The main design results presented in this paper will be applied to the implementation of future hybrid V2X communication terminals for vehicles.

• Education of Primary School Mathematics
• /
• v.2 no.1
• /
• pp.3-13
• /
• 1998

In mathematics education we have focused on how to improve the problem-solving ability, which makes its way to the new direction with the introduction of meta-cognition. As meta-cognition is based on cognitive activity of learners and concerned about internal properties, we may find a more effective way to generate learners problem-solving power. Its means that learners can regulate cognitive process according to their gorls of learning by themselves. Moreover, they are expected to make active participation through this process. If specific meta problems designed to develop meta-cognition are offered, learners are able to work alone by means of their own cognition and regulation while solving problems. They can transfer meta-cognition to the other subjects as well as mathematics. The studies on meta-cognition conducted so far may be divided into these three types. First in Flavell(［3］) meta-cognition is defined as the matter of being conscious of one's own cognition, that is, recognizing cognition. He conducted an experiment with presschoolers and children who just entered primary school and concluded that their cognition may be described as general stage that can not link to specific situation in line with Piaget. Second, Brown(［1］, ［2］) and others argued that meta-cognition means control and regulation of one's own cognition and tried to apply such concept to classrooms. He tried to fined out the strategies used by intelligent students and teach such types of activity to other students. Third, Merleary-Ponty (1962) claimed that meta-cognition is children's way of understanding phenomena or objects. They worked on what would come out in children's cognition responding to their surrounding world. In this paper following the model of meta-cognition produced by Lester (［7］) based on such ideas, we develop types of meta-cognition. In the process of meta-cognition, the meta-cognition working for it is to be intentionally developed and to help unskilled students conduct meta-cognition. When meta-cognition is disciplined through meta problems, their problem-solving power will provide more refined methods for the given problems through autonomous meta-cognitive activity without any further meta problems.

• Journal of the Korean Society for Railway
• /
• v.11 no.4
• /
• pp.364-370
• /
• 2008

In this paper, we proposed new sensor network architecture with autonomous robots based on beacon mode and implemented real time monitoring system in real test-bed environment. The proposed scheme offers beacon based real-time scheduling for reliable association process with parent nodes and dynamically assigns network address by using NAA (Next Address Assignment) mechanism. For the large scale multi-sensor processing, our real-time monitoring system accomplished the intelligent database processing, which can generate not only the alert messages to the civilians but also process various sensing data such as fire, air, temperature and etc. Moreover, we also developed mobile robot which can support network mobility. Though the performance evaluation by using real test-bed system, we illustrate that our proposed system demonstrates promising performance for emergence monitoring systems.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39C no.11
• /
• pp.1177-1184
• /
• 2014

An wired ship area network has functionality of remote control and autonomous management of various sensors and instruments embedded or boarded in a ship. For such environment, the DiffServ (Differentiated Services) realizes that the high-speed real-time flow with the higher priority has the guaranteed minimum data rate and is delivered faster. As a result of this DiffServ effect, the intelligent Ship Area Networks can be implemented. In this paper, an packet drop technique is proposed to outperform the previous RIO (RED In and Out) drop mechanism for DiffServ in ship area networks. the proposed packet drop technique does not manage the individual flows and divides them into several flow groups according to a criterion. And it guarantees the fairness between individual flows in the same QoS class through the group-based control. In simulation results of the proposed packet drop technique, the link utilization decreases than RIO. But it guarantees more data rates to DiffServ flows passing multiple bottleneck links.

• The Journal of Korea Robotics Society
• /
• v.1 no.1
• /
• pp.17-24
• /
• 2006

Though the final goal of mobile robot navigation is to be autonomous, operators' intelligent and skillful decisions are necessary when there are many scattered obstacles. There are several limitations even in the camera-based tele-operation of a mobile robot, which is very popular for the mobile robot navigation. For examples, shadowed and curved areas cannot be viewed using a narrow view-angle camera, especially in bad weather such as on snowy or rainy days. Therefore, it is necessary to have other sensory information for reliable tele-operations. In this paper, sixteen ultrasonic sensors are attached around a mobile robot in a ring pattern to measure the distances to obstacles. A collision vector is introduced in this paper as a new tool for obstacle avoidance, which is defined as a normal vector from an obstacle to the mobile robot. Based on this collision vector, a virtual reflection force is generated to avoid the obstacles and then the reflection force is transferred to an operator who is holding a joystick to control the mobile robot. Relying on the reflection force, the operator can control the mobile robot more smoothly and safely. For this bi-directional tele-operation, a master joystick system using a hall sensor was designed to resolve the existence of nonlinear sections, which are usual for a general joystick with two motors and potentiometers. Finally, the efficiency of a force reflection joystick is verified through the comparison of two vision-based tele-operation experiments, with and without force reflection.

• Journal of KIISE
• /
• v.43 no.7
• /
• pp.781-785
• /
• 2016

Learning from human behaviors in the real world is essential for human-aware intelligent systems such as smart assistants and autonomous robots. Most of research focuses on correlations between sensory patterns and a label for each activity. However, human activity is a combination of several event contexts and is a narrative story in and of itself. We propose a novel approach of human activity prediction based on event cognition. Egocentric multi-sensor data are collected from an individual's daily life by using a wearable device and smartphone. Event contexts about location, scene and activities are then recognized, and finally the users" daily activities are predicted from a decision rule based on the event contexts. The proposed method has been evaluated on a wearable sensor data collected from the real world over 2 weeks by 2 people. Experimental results showed improved recognition accuracies when using the proposed method comparing to results directly using sensory features.

• Journal of the Korean Society of Industry Convergence
• /
• v.20 no.2
• /
• pp.177-186
• /
• 2017

With the latest developments in ICT(Information Communication Technology) technology, research on Intelligent Car, Connected Car that support autonomous driving or services is actively underway. It is true that the number of inputs linked to external connections is likely to be exposed to a malicious intrusion. I studied possible security issues that may occur within the Connected Car. A variety of security issues may arise in the use of CAN, the most typical internal network of vehicles. The data can be encrypted by encrypting the entire data within the CAN network system to resolve the security issues, but can be time-consuming and time-consuming, and can cause the authentication process to be carried out in the event of a certification procedure. To resolve this problem, CAN network system can be used to authenticate nodes in the network to perform a unique authentication of nodes using nodes in the network to authenticate nodes in the nodes and By encoding the ID, identifying the identity of the data, changing the identity of the ID and decryption algorithm, and identifying the cipher and certification techniques of the external invader, the encryption and authentication techniques could be detected by detecting and verifying the external intruder. Add a monitoring node to the CAN network to resolve this. Share a unique ID that can be authenticated using the server that performs the initial certification of nodes within the network and encrypt IDs to secure data. By detecting external invaders, designing encryption and authentication techniques was designed to detect external intrusion and certification techniques, enabling them to detect external intrusions.

• Journal of Internet Computing and Services
• /
• v.22 no.4
• /
• pp.13-28
• /
• 2021

Reinforcement Learning (RL) is a machine learning algorithm that repeat the closed-loop process that agents perform actions specified by the policy, the action is evaluated with a reward function, and the policy gets updated accordingly. The key benefit of RL is the ability to optimze the policy with action evaluation. Hence, it can effectively be applied to developing advanced intelligent systems and autonomous systems. Conventional RL incoporates a single policy, a reward function, and relatively simple policy update, and hence its utilization was limited. In this paper, we propose an extended RL model that considers multiple instances of RL elements. We define a formal model of the key elements and their computing model of the extended RL. Then, we propose design methods for applying to system development. As a case stud of applying the proposed formal model and the design methods, we present the design and implementation of an advanced car navigator system that guides multiple cars to reaching their destinations efficiently.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.5
• /
• pp.13-20
• /
• 2019

Recently, object detection technology based on CNN has been actively studied. Object detection technology is used as an important technology in autonomous vehicles, intelligent image analysis, and so on. In this paper, we propose a rotation change robust object detection system by applying MI-FL (Moment Invariant-Feature Layer) to SSD (Single Shot Multibox Detector) which is one of CNN-based object detectors. First, the features of the input image are extracted based on the VGG network. Then, a total of six feature layers are applied to generate bounding boxes by predicting the location and type of object. We then use the NMS algorithm to get the bounding box that is the most likely object. Once an object bounding box has been determined, the invariant moment feature of the corresponding region is extracted using MI-FL, and stored and learned in advance. In the detection process, it is possible to detect the rotated image more robust than the conventional method by using the previously stored moment invariant feature information. The performance improvement of about 4 ~ 5% was confirmed by comparing SSD with existing SSD and MI-FL.