• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권12호
• /
• pp.3416-3435
• /
• 2023

Task offloading in vehicular networks is hot topic in the development of autonomous driving. In these scenarios, due to the role of vehicles and pedestrians, task characteristics are changing constantly. The classical deep learning algorithm always uses a pre-trained neural network to optimize task offloading, which leads to system performance degradation. Therefore, this paper proposes a neural network fine-tuning task offloading algorithm, combining with location prediction for pedestrians and vehicles by the Payne model of fluid dynamics and the car-following model, respectively. After the locations are predicted, characteristics of tasks can be obtained and the neural network will be fine-tuned. Finally, the proposed algorithm continuously predicts task characteristics and fine-tunes a neural network to maintain high system performance and meet low delay requirements. From the simulation results, compared with other algorithms, the proposed algorithm still guarantees a lower task offloading delay, especially when congestion occurs.

• 대한산업공학회지
• /
• 제40권3호
• /
• pp.267-274
• /
• 2014

The purpose of this study is to identify a relationship between the time taken and the characteristics of touch key for touch-screen-based in-vehicle information system (IVIS) and to suggest a new Fitts' law formula that is added a driving speed parameter. Many studies already have shown that Fitts' law is well fitted in various devices for primary tasks, but there is no study of Fitts' law for secondary task in dual-task situation. Fitts' law may not be applied to the secondary task as it is, because the secondary task performance can be affected by the amount of attention for the primary task. To verify this, we carried out an experiment that showed whether pointing task to touch-screen-based IVIS during driving is affected by driving speeds or not. In the experiment, 30 people were volunteered for participants and the participants carried out driving task and pointing task on the screen of IVIS simultaneously. We measured the time to point a touch key on IVIS for every condition (3 driving speeds${\times}5$ touch key sizes${\times}7$ distances between steering wheel and touch key). As a result, there was an effect of driving speed on the pointing time. As we extended the index of difficulty of the conventional Fitts' law formula by incorporating driving speed, we established an extended Fitts' law formula for pointing on IVIS, which showed better accordance with dual task situation. This study can be evidence that secondary task performance is affected by degree of concentration on primary task, and the extended Fitts' law formula can be useful to design interfaces of IVIS.

• 대한인간공학회지
• /
• 제29권4호
• /
• pp.605-610
• /
• 2010

In order to design a touch interface for the elderly, human performance of input tasks on the touch screen was investigated by the laboratory experiment. Input times and input errors were measured in the experimental conditions that were changed according to age, key size, interkey space and input tool(finger or stylus pen). In the most of all experimental conditions, the task performance of the elderly was lower than that of the young. However, there were significantly different performance patterns between both groups. As the difficulty of task was getting higher, the task performance of the elderly was sharply decreased; pressing small key button by finger sharply increased input time and error rate, compared to that of the young. Therefore, the square key size suitable to the elderly may be over $8.0{\times}8.0mm$. While the interkey space did not influence to the input task performance of the young, the task performance of the elderly was influenced. The elderly showed big difference of task performance according to input tool. However, the young were less influenced by input tool.

• ETRI Journal
• /
• 제42권5호
• /
• pp.748-760
• /
• 2020

MapReduce is a framework that can process huge datasets in parallel and distributed computing environments. However, a single machine failure during the runtime of MapReduce tasks can increase completion time by 50%. MapReduce handles task failures by restarting the failed task and re-computing all input data from scratch, regardless of how much data had already been processed. To solve this issue, we need the computed key-value pairs to persist in a storage system to avoid re-computing them during the restarting process. In this paper, the task failure resilience (TFR) technique is proposed, which allows the execution of a failed task to continue from the point it was interrupted without having to redo all the work. Amazon ElastiCache for Redis is used as a non-volatile cache for the key-value pairs. We measured the performance of TFR by running different Hadoop benchmarking suites. TFR was implemented using the Hadoop software framework, and the experimental results showed significant performance improvements when compared with the performance of the default Hadoop implementation.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.149.3-149
• /
• 2001

Task allocation is a key problem in multiagent systems. The importance of automated negotiation protocols for solving the task allocation problem is increasing as a consequence of increased multi-agent applications. In this paper, we introduce the multiagent Traveling Salesman Problem(TSP) as an example of task reallocation problem, and suggest Vickery auction as an inter-agent coordination mechanism for solving this problem. In order to apply this market-based coordination mechanism into multiagent TSPs, we define the profit of each agent, the ultimate goal of negotiation, cities to be traded out through auctions, the bidding strategy, and the order of auctions. The primary advantage of such approach is that it can find an optimal task allocation ...

• 한국IT서비스학회지
• /
• 제18권3호
• /
• pp.75-93
• /
• 2019

Digital twin technology is one of the key technologies to strengthen the competitiveness of manufacturing industry from the viewpoint of digital transformation in the era of $4^{th}$ industrial revolution. In this research, the important role in using digital twin technology in product design, This paper summarizes and empirically verifies the technical characteristics of digital twins, a key concept in the digital transition of the manufacturing industry. In this study, the technology characteristics of digital twin which is key concept in the digital transformation of manufacturing industry are summarized and empirically validated which factors militate a critical role in the use of digital twin technology in product design which is key area of product development. As a result of analysis, datafication, intellectualization which are characteristics of digital twin technology and task characteristics of product design influence Task Technology Fit (TTF) and Task Technology Fit (TTF) influences Technology (UTAUT) And finally, performance expectancy, effort expectancy, social influence and facilitating conditions affect usage intention.

• 대한수학교육학회지:수학교육학연구
• /
• 제8권1호
• /
• pp.163-182
• /
• 1998

Performance assessment is one of the authentic assessment method that are consistent with new curriculum goal, concentrated on the process rather than the results of problem solving. But the key to good assessment is matching the assessment task to intended objectives. Based on the review of literatures, the current performance assessment task was critically analysed. As a result, this study developed appropriate model and criteria of performance assessment task to elementary mathematics.

• 한국환경교육학회지:환경교육
• /
• 제19권3호
• /
• pp.57-66
• /
• 2006

This study is purposed to compose systematic and efficient curricula of the courses centered on the duties of forest interpreter. In the first step, the purpose of task analysis is to identify the forest interpreter's duties and tasks. 13 professional were designated as panel, and the task chart was completed via DACUM analysis. The tasks performed by forest interpreter are categorized in the development of specialty, program planning, comprehension on the engaged forest, program development, program execution and program evaluation, which are classified into 59 sub-tasks. In the second step, need analysis is focused on the evaluation of the degree of job importance, the necessity of education based on the results from the task analysis. In consequence, 23 key tasks are determined. In the third step, knowledge, skill, tool, and attitude required for key tasks were analyzed and reorganized into 23 subjects. This study has significance in 3 respects. Firstly, the tasks of forest interpreter are analyzed to define their roles. Secondly, the curricula composed according to the results of task analysis and need analysis allow the realizable and prerequisite subjects within the restricted resources. Finally, this study suggests the curriculum, which shall be the bases for the program planning and operation of a lot of educational organization and institutions.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권2호
• /
• pp.483-503
• /
• 2014

In this paper, human action recognition using pyramid histograms of oriented gradients and collaborative multi-task learning is proposed. First, we accumulate global activities and construct motion history image (MHI) for both RGB and depth channels respectively to encode the dynamics of one action in different modalities, and then different action descriptors are extracted from depth and RGB MHI to represent global textual and structural characteristics of these actions. Specially, average value in hierarchical block, GIST and pyramid histograms of oriented gradients descriptors are employed to represent human motion. To demonstrate the superiority of the proposed method, we evaluate them by KNN, SVM with linear and RBF kernels, SRC and CRC models on DHA dataset, the well-known dataset for human action recognition. Large scale experimental results show our descriptors are robust, stable and efficient, and outperform the state-of-the-art methods. In addition, we investigate the performance of our descriptors further by combining these descriptors on DHA dataset, and observe that the performances of combined descriptors are much better than just using only sole descriptor. With multimodal features, we also propose a collaborative multi-task learning method for model learning and inference based on transfer learning theory. The main contributions lie in four aspects: 1) the proposed encoding the scheme can filter the stationary part of human body and reduce noise interference; 2) different kind of features and models are assessed, and the neighbor gradients information and pyramid layers are very helpful for representing these actions; 3) The proposed model can fuse the features from different modalities regardless of the sensor types, the ranges of the value, and the dimensions of different features; 4) The latent common knowledge among different modalities can be discovered by transfer learning to boost the performance.

• Journal of Information Processing Systems
• /
• 제14권1호
• /
• pp.129-139
• /
• 2018

The crowdsourcing software development (CSD) is growing rapidly in the open call format in a competitive environment. In CSD, tasks are posted on a web-based CSD platform for CSD workers to compete for the task and win rewards. Task searching and assigning are very important aspects of the CSD environment because tasks posted on different platforms are in hundreds. To search and evaluate a thousand submissions on the platform are very difficult and time-consuming process for both the developer and platform. However, there are many other problems that are affecting CSD quality and reliability of CSD workers to assign the task which include the required knowledge, large participation, time complexity and incentive motivations. In order to attract the right person for the right task, the execution of action plans will help the CSD platform as well the CSD worker for the best matching with their tasks. This study formalized the task assignment method by utilizing different situations in a CSD competition-based environment in artificial intelligence (AI) planning. The results from this study suggested that assigning the task has many challenges whenever there are undefined conditions, especially in a competitive environment. Our main focus is to evaluate the AI automated planning to provide the best possible solution to matching the CSD worker with their personality type.