• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.4
• /
• pp.457-466
• /
• 2005

In this paper, we propose a wavelet neural network(WNN) based predictive control method for path tracking of mobile robots with multi-input and multi-output. In our control method, we use a WNN as a state predictor which combines the capability of artificial neural networks in learning processes and the capability of wavelet decomposition. A WNN predictor is tuned to minimize errors between the WNN outputs and the states of mobile robot using the gradient descent rule. And control signals, linear velocity and angular velocity, are calculated to minimize the predefined cost function using errors between the reference states and the predicted states. Through a computer simulation for the tracking performance according to varied track, we demonstrate the efficiency and the feasibility of our predictive control system.

• The KIPS Transactions:PartB
• /
• v.8B no.3
• /
• pp.237-242
• /
• 2001

고전적인 GLA 알고리즘과 마찬가지로 Kohonen 학습법은 경도 강하법으로 오차함수의 해에 접근해 나간다. 따라서 KLA의 이러한 문제를 극복하기 위해 모의 담금질법의 일종인 Cauchy 학습법을 응용을 제안한다. 그러나 이 방법은 학습시간이 느리다고 하는 단점이 있다. 본 논문 이 점을 개선시키기 위해 Cauchy 학습법과 Kohonen 학습법을 순차 결합시킨 또 다른 학습법을 제안한다. 그 결과 코시 학습법과 마찬가지로 국부최적 문제를 극복하면서도 삭습시간을 단축할 수 있었다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.04b
• /
• pp.675-677
• /
• 2000

지적 능력은 정상인데 해당 연령층에서 기대하는 학습 결과를 나타내지 못하여 교육성취도가 정규학년 수준에 크게 미달되거나 학습 속도가 부진한 특수학습장애 아동들은 일반 아동들과 같은 수업을 받는다. 이런 현실은 학부모나 교사들의 특별한 배려가 없다면 당연히 장애 아동들의 학습에 대한 흥미를 떨어뜨릴 수 있다. 우리 나라에서 특수교육을 받는 장애 아동 중 가장 많은 수를 차지하는 특수학습장애 아동들을 위한 특별한 프로그램의 개발이 절실하다. 이에 본 논문은 동기 이론들 기반으로 특수학습장애 아동들의 흥미를 길러 줄 수 있는 프로그램을 개발하였다. 이의 개발로 특수학습장애 아동들의 주의 집중 능력과 성취동기를 길러주고, 흥미를 키워주므로 학습에 대한 의욕을 가질 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.512-515
• /
• 2020

학습 속도가 느린 강화학습을 범용적으로 활용할 수 있도록 연구가 활발하게 이루어지고 있다. 사전 지식을 제공해서 학습 속도를 높일 수 있지만, 잘못된 사전 지식을 제공했을 위험이 존재한다. 본 연구는 불확실하거나 잘못된 사전 지식이 학습에 어떤 영향을 미치는지 살펴본다. OpenAI Gym 라이브러리를 이용해서 만든 Gamble 환경, Cliff 환경, 그리고 Maze 환경에서 실험을 진행했다. 그 결과 사전 지식을 통해 에이전트의 행동에 경향성을 부여할 수 있다는 것을 확인했다. 또한, 경로탐색에 있어서 잘못된 사전 지식이 얼마나 학습을 방해하는지 알아보았다.

• Journal of Korean Society of Transportation
• /
• v.25 no.3
• /
• pp.87-97
• /
• 2007

This paper proposes drivers' rational belief formation under a bounded traffic environment. This is to escape the criticism that excessive rationality (e.g., a driver's calculating ability and memory capacity) is required of drivers. Under bounded traffic environments. drivers do not have structural knowledge of traffic conditions and others' decisions. Simulations are carried out using a program coded in C. Consequently, the author found the learning process of drivers and the value of information can be differentiated by route conditions and the characteristics of driver groups. Also, it was found that rational drivers form different beliefs about traffic conditions even though they have the same traffic environment in a bounded traffic environment.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.24 no.1
• /
• pp.61-69
• /
• 2021

Recently, autonomous navigation technology is actively being developed due to the increasing demand of an unmanned surface vehicle(USV). Local planning is essential for the USV to safely reach its destination along paths. the dynamic window approach(DWA) algorithm is a well-known navigation scheme as a local path planning. However, the existing DWA algorithm does not consider path line tracking, and the fixed weight coefficient of the evaluation function, which is a core part, cannot provide flexible path planning for all situations. Therefore, in this paper, we propose a new DWA algorithm that can follow path lines in all situations. Fixed weight coefficients were trained using reinforcement learning(RL) which has been actively studied recently. We implemented the simulation and compared the existing DWA algorithm with the DWA algorithm proposed in this paper. As a result, we confirmed the effectiveness of the proposed algorithm.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.458-459
• /
• 2021

Recently, researches using deep learning technology based on Wi-Fi fingerprints have been conducted for accurate services in indoor location-based services. Among the deep learning models, an RNN model that can store information from the past can store continuous movements in indoor positioning, thereby reducing positioning errors. At this time, continuous sequential data is required as training data. However, since Wi-Fi fingerprint data is generally managed only with signals for a specific location, it is inappropriate to use it as training data for an RNN model. This paper proposes a path generation method through prediction of a moving path based on Wi-Fi fingerprint data extended to region data through clustering to generate sequential input data of the RNN model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.6
• /
• pp.1081-1086
• /
• 2016

While smart learning have been introduced for more learning effect, this study is to understand instructor's smart learning acceptance using technology acceptance model(TAM). This study developed the extended TAM model, including external pressure for smart learning and smart self efficacy for smart devices as study variables and attempted to examine the research model through the empirical analysis. The research model has the 7 variables including smart self-efficacy and external pressure. For the empirical study, the survey was conducted for the one month, March, 2016, and the total 143 data among the collected 167 responses were used for the empirical analysis. As the result of the analysis through the structural equation model, the 9 paths among the total 10 paths show the significant relationships between the variables. Through using the result of this study, it is to provide suggestions for the improvement of smart learning environments.

• Journal of Internet Computing and Services
• /
• v.16 no.6
• /
• pp.39-46
• /
• 2015

Knowledge space theory is a theory that provides a guidelines for human learners' possible education decisions and has been used in various educational environment. However, traditional methodologies using the knowledge space theory have always depended on handwork system and it is necessary to learn programming language such as Visual Basic and R, causing time consuming situations. In order to overcome those issues on the environment of education we propose a new Knowledge Structure Analysis System that not just analyzes learners' knowledge structures automatically but to provide critical learning path for the learners based on knowledge space theory. Proposed system is implemented by using rApache generating critical learning path computing Chi-square value. This provides an automatic way of analyzing knowledge structure in learners' knowledge space and shows systematic reviews for the knowledge space.

• Korea Trade Review
• /
• v.42 no.5
• /
• pp.93-112
• /
• 2017

The purpose of this study is to analyze the effects of FTA information learning in export SMEs. Therefore, this study has constructed an integrated model including the moderating effects of FTA information learning on the of export performance in SME. The relationship between SMEs' localization strategy, product innovation capacity, and FTA information learning was linked to export performance, and an empirical analysis was conducted on 195 export SMEs. The path analysis was performed using the structural equation model(SEM), and six hypotheses including the control effect were tested. As a result, the localization strategy of SMEs positively influenced product innovation capacity. On the other hand, FTA information learning did not show significant results. Product innovation capacity and FTA information learning as an antecedents showed significant results in terms of export performance. In the moderated effects analysis, the moderated effect between the localization strategy and FTA information learning did not show significant effect on the product innovation capacity. Whereas the moderated effect between the product innovation capacity and the FTA information learning significant influence on the export performance of SMEs.