• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.38 no.3
• /
• pp.247-255
• /
• 1989

This paper proposes an error model with integral action and a pole-place-ment self-tuning controller for robot manipulators in task coordinates. The controller can reject the offset due to any load disturbance without a detailed description of the robot dynamics. The error model parameters are estimated by the recursive least square identification algorithms, and controller parameters are determined by the pole-placement method. A computer simulation study has been conducted to demonstrate the performance of the proposed control system in task coordinates for a 3-joint and 2-link spatial robot manipulator with payload.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.5
• /
• pp.64-72
• /
• 2004

An extension of interaction matrix formulation to the problem of system and disturbance identification for a plant that is corrupted by both process and output disturbances is presented. The teaming control develops controllers that learn to improve their performance at executing a given task, based on experience performing this task. The simplest forms of loaming control are based on the same concept as integral control, but operating in the domain of the repetitions of the task. This paper studies the use of such controllers in a decentralized system, such as a robot moving on the vertical plane with the controller for each link acting independently. The basic result of the paper is to show that stability and iterative precision of the learning controllers for all subsystems when the coupling between subsystems is turned off, assures stability of the decentralized teaming in the coupled system, provided that the sample time in the digital teaming controller is sufficiently short. The methods of teaming system are shown up for the iterative precision of each link.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.2
• /
• pp.775-792
• /
• 2015

Person re-identification is an important and challenging task in computer vision with numerous real world applications. Despite significant progress has been made in the past few years, person re-identification remains an unsolved problem. This paper presents a novel appearance-based approach to person re-identification. The approach exploits region covariance matrix and color histograms to capture the statistical properties and chromatic information of each object. Robustness against low resolution, viewpoint changes and pose variations is achieved by a novel signature, that is, the combination of Log Covariance Matrix feature and HSV histogram (LCMH). In order to further improve re-identification performance, third-party image sets are utilized as a common reference to sufficiently represent any image set with the same type. Distinctive and reliable features for a given image set are extracted through decision boundary between the specific set and a third-party image set supervised by max-margin criteria. This method enables the usage of an existing dataset to represent new image data without time-consuming data collection and annotation. Comparisons with state-of-the-art methods carried out on benchmark datasets demonstrate promising performance of our method.

• Smart Structures and Systems
• /
• v.7 no.2
• /
• pp.133-152
• /
• 2011

System identification is a fundamental step towards the application of structural health monitoring and damage detection techniques. On this respect, the development of evolved identification strategies is a priority for obtaining reliable and repeatable baseline modal parameters of an undamaged structure to be adopted as references for future structural health assessments. The paper presents the identification of the modal parameters of the Guangzhou New Television Tower, China, using a data-driven stochastic subspace identification (SSI-data) approach complemented with an appropriate automatic mode selection strategy which proved to be successful in previous literature studies. This well-known approach is based on a clustering technique which is adopted to discriminate structural modes from spurious noise ones. The method is applied to the acceleration measurements made available within the task I of the ANCRiSST benchmark problem, which cover 24 hours of continuous monitoring of the structural response under ambient excitation. These records are then subdivided into a convenient number of data sets and the variability of modal parameter estimates with ambient temperature and mean wind velocity are pointed out. Both 10 minutes and 1 hour long records are considered for this purpose. A comparison with finite element model predictions is finally carried out, using the structural matrices provided within the benchmark, in order to check that all the structural modes contained in the considered frequency interval are effectively identified via SSI-data.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.728-735
• /
• 2022

Roofers get exposed to increased risk of knee musculoskeletal disorders (MSDs) at different phases of a sloped shingle installation task. As different phases are associated with different risk levels, this study explored the application of machine learning for automated classification of seven phases in a shingle installation task using knee kinematics and roof slope information. An optical motion capture system was used to collect knee kinematics data from nine subjects who mimicked shingle installation on a slope-adjustable wooden platform. Four features were used in building a phase classification model. They were three knee joint rotation angles (i.e., flexion, abduction-adduction, and internal-external rotation) of the subjects, and the roof slope at which they operated. Three ensemble machine learning algorithms (i.e., random forests, decision trees, and k-nearest neighbors) were used for training and prediction. The simulations indicate that the k-nearest neighbor classifier provided the best performance, with an overall accuracy of 92.62%, demonstrating the considerable potential of machine learning methods in detecting shingle installation phases from workers knee joint rotation and roof slope information. This knowledge, with further investigation, may facilitate knee MSD risk identification among roofers and intervention development.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.10
• /
• pp.5003-5011
• /
• 2013

This paper measured whether neural and behavior responses to attention-emotion task were reflected to emotion regulation capacities. For this purpose, Nineteen healthy right-handed graduates participated in the emotion-attention task three times for three days. Before and after the negative and positive video clips were shown, the participants performed emotion-attention task. EEG and response time were recorded during emotion-attention task. There was positive correlation between ERP P100 and P300 component. The larger the P100 amplitudes at the specific positions, the longer the P300 latencies at these same positions during attention-emotion task. The longer the P300 latencies at the specific positions, the longer the delay in response time. Also, there is and individual differences in ERP components and response time during attention-emotion integration task. Individuals who had lower amplitude and shorter latency of ERP showed faster response time during attention-emotion task, regardless of the type of video clips. This characteristic was interpreted to the lower emotional controls due to premature response for target identification.

• Phonetics and Speech Sciences
• /
• v.5 no.1
• /
• pp.91-98
• /
• 2013

This paper investigated Korean prosody as it relates to graded internal structure in an identification function. Within Korean prosody, variants regarded as dialectal variations can appear as different prosodic scales, which contain the range of within-category variations. The current experiment was intended to show how the prosodic scale corresponding to the range of within-category differences relates to f0 contours for speakers of two Korean dialects, North Kyungsang and South Cholla. In an identification task, participants responded by selecting an item from two answer choices. The probability of choosing the correct response from the two choices was computed by a logistic regression analysis using intercepts and slopes. That is, the correct response between two choices was used to show a linear line with an s-shape presentation. In this paper, to investigate the graded internal structure of labeling, 25%, 50%, and 75% of predicted probability were assessed. Listeners from North Kyungsang showed progressive variations, whereas listeners from South Cholla revealed random patterns in the internal structure of the identification function. In this paper, the results were plotted using scatterplot graphs, applying the range of within-category variation and predicted probability obtained from the logistic regression analyses. The scatterplot graphs showed the different degree of the responses for f0 scales (i.e., variations within categories). The results demonstrate that the gradient structures of native pitch accent users become more progressive in response to f0 scales.

• Industrial Engineering and Management Systems
• /
• v.16 no.1
• /
• pp.64-79
• /
• 2017

Identifying disease genes from human genome is a critical task in biomedical research. Important biological features to distinguish the disease genes from the non-disease genes have been mainly selected based on traditional feature selection approaches. However, the traditional feature selection approaches unnecessarily consider many unimportant biological features. As a result, although some of the existing classification techniques have been applied to disease gene identification, the prediction performance was not satisfactory. A small set of the most important biological features can enhance the accuracy of disease gene identification, as well as provide potentially useful knowledge for biologists or clinicians, who can further investigate the selected biological features as well as the potential disease genes. In this paper, we propose a new stepwise random forests (SRF) approach for biological feature selection and disease gene identification. The SRF approach consists of two stages. In the first stage, only important biological features are iteratively selected in a forward selection manner based on one-dimensional random forest regression, where the updated residual vector is considered as the current response vector. We can then determine a small set of important biological features. In the second stage, random forests classification with regard to the selected biological features is applied to identify disease genes. Our extensive experiments show that the proposed SRF approach outperforms the existing feature selection and classification techniques in terms of biological feature selection and disease gene identification.

• Smart Structures and Systems
• /
• v.10 no.4_5
• /
• pp.331-351
• /
• 2012

In this study an output-only system identification technique for civil structures under ambient vibrations is carried out, mainly focused on using the Stochastic Subspace Identification (SSI) based algorithms. A newly developed signal processing technique, called Singular Spectrum Analysis (SSA), capable to smooth a noisy signal, is adopted for preprocessing the measurement data. An SSA-based SSI algorithm with the aim of finding accurate and true modal parameters is developed through stabilization diagram which is constructed by plotting the identified system poles with increasing the size of data matrix. First, comparative study between different approaches, with and without using SSA to pre-process the data, on determining the model order and selecting the true system poles is examined in this study through numerical simulation. Finally, application of the proposed system identification task to the real large scale structure: Canton Tower, a benchmark problem for structural health monitoring of high-rise slender structures, using SSA-based SSI algorithm is carried out to extract the dynamic characteristics of the tower from output-only measurements.

• 제어로봇시스템학회:학술대회논문집
• /
• 1998.10a
• /
• pp.280-284
• /
• 1998

Many classes of nonlinear systems can be represented by Volterra kernel expansion. Therefore, identification of Volterra kernels of nonlinear system is an important task for obtaining the nonlinear characteristics of the nonlinear system. Although one of the authors has recently proposed a new method for obtaining the Volterra kernels of a nonlinear system by use of M-sequence and correlation technique, our mettled of nonlinear system identification is limited to Wiener-type nonlinear system and we can not apply this method to the identification of Hammerstein-type nonlinear system. This paper describes a new mettled for obtaining Volterra kernels of Hammerstein nonlinear system by adding a linear element in front of tile Hammerstein system. First we calculate the linear element of Hammerstein system by use of conventional correlation method. Secondly, we put a linear element in front of Hammerstein system. Then the total system becomes Wiener-type nonlinear system. Therefore we can use our method on Volterra kernel identification by use of M-sequence. Thus we get the coefficients of the approximation polynomial of nonlinear element of Hammerstein system. From the results of simulation, a good agreement with theoretical considerations is obtained, showing a wide applicability of our method.