• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.174-174
• /
• 2000

Q-learning, based on discrete state and action space, is a most widely used reinforcement Learning. However, this requires a lot of memory and much time for learning all actions of each state when it is applied to a real mobile robot navigation using continuous state and action space Region-based Q-learning is a reinforcement learning method that estimates action values of real state by using triangular-type action distribution model and relationship with its neighboring state which was defined and learned before. This paper proposes a new Region-based Q-learning which uses a reward assigned only when the agent reached the target, and get out of the Local optimal path with adjustment of random action rate. If this is applied to mobile robot navigation, less memory can be used and robot can move smoothly, and optimal solution can be learned fast. To show the validity of our method, computer simulations are illusrated.

• 한국경영정보학회:학술대회논문집
• /
• 2008.06a
• /
• pp.428-433
• /
• 2008

The online advertising industry's business model undertakes the change from CPM (cost-per-mille)-based to CPC (cost-per-click)-based. However, due to the problem of 'Click Fraud', CPA (cost-per-action) has been regarded as a new step. For CPA, publishers need to get information after a user clicks an advertisement. Therefore, in CPA, the key is to get Conversion Action Data (CAD). This paper introduces two existing mechanisms for getting CAD, compare their characteristics, and analyze their limitations. Then the two new mechanisms are introduced and their requirements and feasibility are analyzed. Lastly, we compare the existing two and the new two mechanisms, and point out each mechanism's business possibility, value and Application Area. This paper will help publishers choose the most appropriate mechanism on the basis of their situation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.5
• /
• pp.2327-2347
• /
• 2018

Human actions can be recognized from depth sequences. In the proposed algorithm, we initially construct depth, motion maps (DMM) by projecting each depth frame onto three orthogonal Cartesian planes and add the motion energy for each view. The body part of the action (BPoA) is calculated by using bounding box with an optimal window size based on maximum spatial and temporal changes for each DMM. Furthermore, feature vector is constructed by using BPoA for each human action view. In this paper, we employed an ensemble based learning approach called Rotation Forest to recognize different actions Experimental results show that proposed method has significantly outperforms the state-of-the-art methods on Microsoft Research (MSR) Action 3D and MSR DailyActivity3D dataset.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.2148-2161
• /
• 2019

In this paper, we present a real-time cattle action recognition algorithm to detect the estrus phase of cattle from a live video stream. In order to classify cattle movement, specifically, to detect the mounting action, the most observable sign of the estrus phase, a simple yet effective feature description exploiting motion history images (MHI) is designed. By learning the proposed features using the support vector machine framework, various representative cattle actions, such as mounting, walking, tail wagging, and foot stamping, can be recognized robustly in complex scenes. Thanks to low complexity of the proposed action recognition algorithm, multiple cattle in three enclosures can be monitored simultaneously using a single fisheye camera. Through extensive experiments with real video streams, we confirmed that the proposed algorithm outperforms a conventional human action recognition algorithm by 18% in terms of recognition accuracy even with much smaller dimensional feature description.

• International Journal of Control, Automation, and Systems
• /
• v.6 no.2
• /
• pp.282-287
• /
• 2008

In distributed mobile robot systems, autonomous robots accomplish complicated tasks through intelligent cooperation with each other. This paper presents behavior learning and online distributed evolution for cooperative behavior of a group of autonomous robots. Learning and evolution capabilities are essential for a group of autonomous robots to adapt to unstructured environments. Behavior learning finds an optimal state-action mapping of a robot for a given operating condition. In behavior learning, a Q-learning algorithm is modified to handle delayed rewards in the distributed robot systems. A group of robots implements cooperative behaviors through communication with other robots. Individual robots improve the state-action mapping through online evolution with the crossover operator based on the Q-values and their update frequencies. A cooperative material search problem demonstrated the effectiveness of the proposed behavior learning and online distributed evolution method for implementing cooperative behavior of a group of autonomous mobile robots.

• Korean Journal of Human Ecology
• /
• v.6 no.1
• /
• pp.111-127
• /
• 1997

This study was to examine home economics (HE) teachers' and the 1st-grade students' needs perception toward the purposes of HE education in middle school which has been practced since 1995 for both male and female students. This study, attempted (1) to analyze needs priority among the educational purposes of HE subject in relation to three systems of actions; (2) to compare differences between HE teachers' and students' perception concerning the degree of importance and achievement of the educational purposes of HE subject: and (3) to examine what they conceive as the problems In the current HE education. The survey was conducted with the samples of 600 1st-grade middle school students and 101 middle school HE teachers during the period of February-March 1996. The questionnaire used in this study was a modified version which had already been developed along with the 6th HE curriculum. For data analyses, SAS program was utilized to get Means and to perform both discrepancy test and t-test. The findings of this study were summarized as follows: first, with respect to each group's perception of the importance of the purposes related to three systems of action, HE teachers emphasized the importance of the purposes related to emancipatory action, while students placed more emphasis on the purposes related to technical action. Second, in terms of the degree of achievement, students had more positive perception on the degree of achievement of the purposes related to technical action than HE teachers did. Both groups marked low level of recognition on the degree of achievement of the purposes related to emancipatory action. Third, with respect to needs priority, HE teachers placed the first priority on emancipatory action, the second on technical action, and the last on communicative action: in the case of students, the first priority was on technical action, the second on communicative action, and the last on emancipatory action. In addition, the analysis of the opinions on the 6th curriculum revealed that most respondents found it necessary to secure adequate amount of classes for HE education. Also they shared the recongnition that HE curriculum should be renovated into the one which would fully appreciate the purposes of HE education from the perspective of the practical concerns of action which are distinct from the functional and technical concerns of passive learning. The findings of this study can serve as basic data for establishing the new purposes of HE education which put more emphasis on the purposes related to emancipatory action: as well as for developing an enhanced curriculum and reinforcing the identity of HE education.

• Journal of Korean Home Economics Education Association
• /
• v.17 no.4 s.38
• /
• pp.117-131
• /
• 2005

The purpose of this research was to develop objectives of Housing contents in Technology$\cdot$Home Economics by three systems of action and to find out the importance of the objectives of the teachers have taught the class. The 303 teachers from 183 middle schools replied the mail questionnaire during September, 2003. The data were analyzed by SPSS/win. The 21 objectives for each system of action were developed based on the textbooks, teachers guides, and other related references. The importance of the objectives related to communicative system of action was the highest, respectively followed by the ones related to technical and emancipatory systems of action. Indoor Environment & Equipment was more likely to be important than Maintenance & Repairs as the objectives related to communicative system of action, while Usage of Living Space was less likely to be important than the other two sub-units. The importance of the objectives was somewhat differed by the general characteristics of the teachers. The older are the more important the objectives related to technical system of action. Female, Home Economics teachers, who learned philosophy of Home Economics were more likely to think than others objectives related to communicative and emancipatory systems of action to be important. This research showed the teachers' perspectives of the objectives of Housing contents were not the same among respondents and generally supported the previous results from other contents of Home Economics.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.532-532
• /
• 2000

Conventional artificial intelligence systems are not properly responding under dynamically changing environments. To overcome this problem, reactive planning systems implementing new Al principles, called behavior-based Al or emergent computation, have been proposed and confirmed their usefulness. As another alternative, biological information processing systems may provide many feasible ideas to these problems. Immune system, among these systems, plays important roles to maintain its own system against dynamically changing environments. Therefore, immune system would provide a new paradigm suitable for dynamic problem dealing with unknown environments. In this paper, a new approach to behavior-based Al by paying attention to biological immune system is investigated. The feasibility of this method is confirmed by applying to behavior control of an autonomous mobile robot in cluttered environment.

• Bulletin of the Korean Mathematical Society
• /
• v.55 no.4
• /
• pp.1125-1135
• /
• 2018

We show that for an M-action on a compact Hausdorff uniform space, if it has at least two disjoint compact invariant subsets, then it is thickly syndetically sensitive. Additionally, we point out that for a P-M-action of a discrete abelian group on a compact Hausdorff uniform space, the multi-sensitivity is equivalent to both thick sensitivity and thickly syndetic sensitivity.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.10 no.2
• /
• pp.48-55
• /
• 2001

An accumulator in hydraulic systems stores kinetic energy during braking action, and then that control hasty surge pres-sure. This study suggests a method to select the capacity of accumulator to control surge pressure to a desired degree. The selection method is based upon a trial and error approach and computer simulation. A mathematical dynamic model of the system was derived and the parameters in the model were identified from experimental data. A series of computer simulation were done for the brake action. The results of the simulation work were compared with those of experiments. These results of the computer simular-tion and experiments show that the proposed method can be applied effectively to control the surge pressure of the hydraulic regenerative brake systems.