• Journal of Internet Computing and Services
• /
• v.22 no.3
• /
• pp.9-16
• /
• 2021

This study aims to develop a human activity recognition (HAR) system as a Deep-Learning (DL) classification model, distinguishing various human activities. We solely rely on the signals from a wristband accelerometer worn by a person for the user's convenience. 3-axis sequential acceleration signal data are gathered within a predefined time-window-slice, and they are used as input to the classification system. We are particularly interested in developing a Deep-Learning model that can outperform conventional machine learning classification performance. A total of 13 activities based on the laboratory experiments' data are used for the initial performance comparison. We have improved classification performance using the Convolutional Neural Network (CNN) combined with an auto-encoder feature reduction and parameter tuning. With various publically available HAR datasets, we could also achieve significant improvement in HAR classification. Our CNN model is also compared against Recurrent-Neural-Network(RNN) with Long Short-Term Memory(LSTM) to demonstrate its superiority. Noticeably, our model could distinguish both general activities and near-identical activities such as sitting down on the chair and floor, with almost perfect classification accuracy.

• Steel and Composite Structures
• /
• v.37 no.6
• /
• pp.679-693
• /
• 2020

This paper investigates the seismic response of buildings equipped with Self-Centering Energy Dissipating (SCED) braces. Two-dimensional models of 3, 6, 12 and 16-story SCED buildings considering both material and geometric nonlinearities are investigated by carrying out pushover and nonlinear time-history analyses. The response indicators of the buildings are studied for weight-scaled ground motions to represent the Design Basis Earthquake (DBE) level and the Maximum Considered Earthquake (MCE) event. The fragility curves of the buildings for two Immediate Occupancy (IO) and Life Safety (LS) performance levels are developed using Incremental Dynamic Analysis (IDA). Results of the nonlinear response history analyses indicate that the maximum inter-story drift occurs at the taller buildings. The mean peak inter-story drift is less than 2% in both hazard levels. High floor acceleration peaks are observed in all the SCED frames regardless of the building height. The overall ductility and ductility demand increase when the number of stories reduces. The results also showed the residual displacement is negligible for all of case study buildings. The 3 and 6-story buildings exhibit desirable performance in IO and LS performance levels according to fragility curves results, while 12 and 16-story frames show poor performance especially in IO level. The results indicated the SCED braces performance is generally better in lower-rise buildings.

• Structural Engineering and Mechanics
• /
• v.83 no.6
• /
• pp.811-824
• /
• 2022

Despite its disadvantages, soft story can reduce the damage to the upper floors by concentrating drift in that specific story provided that large drifts are avoided. Gapped-Inclined Brace (GIB) with reduced P-delta effects and the control of soft story stiffness makes it possible to take advantage of the soft story in buildings and increase their capacity for energy dissipation. OpenSees software is used in this study to validate and modify the GIB model's shortcomings. Also, the analysis method for this element is changed for design. The modified element is evaluated in 3D analysis. Finally, to retrofit an existing building, this element is used. Based on the Iranian seismic code, a six-story reinforced concrete building is modelled and studied with 3D analysis. In this building, the construction shortcomings and elimination of infills on the ground floor cause the formation of a soft story. Results of nonlinear static analysis, nonlinear dynamic, and incremental dynamic analysis using both components of seismic acceleration applied to the structure at different angles and the fragility curves indicate the improvement of the retrofitted structure's performance using the modified element to reach the required performance level following the retrofit code.

• Journal of the Semiconductor & Display Technology
• /
• v.21 no.4
• /
• pp.121-125
• /
• 2022

This paper deals with the development of a deep-learning-based robot that recognizes various types of stairs and performs a mission to go up to the target floor. The overall motion sequence of the robot is performed based on the ROS robot operating system, and it is possible to detect the shape of the stairs required to implement the motion sequence through rapid object recognition through YOLOv4 and Cuda acceleration calculations. Using the ROS operating system installed in Jetson Nano, a system was built to support communication between Arduino DUE and OpenCM 9.04 with heterogeneous hardware and to control the movement of the robot by aligning the received sensors and data. In addition, the web server for robot control was manufactured as ROS web server, and flow chart and basic ROS communication were designed to enable control through computer and smartphone through message passing.

• Journal of Korean Association for Spatial Structures
• /
• v.22 no.1
• /
• pp.33-40
• /
• 2022

When the center of stiffness and the center of mass of the structure differ under the seismic load, torsion is caused by eccentricity. In this study, an analysis model was modeled in which the positions of the core and the plane rotation axis of a 60-story torsional atypical structure with a plane rotation angle of 1 degree per floor were different. The structural behavior of the analysis model was analyzed, and the earthquake response behavior of the structure was analyzed based on the time history analysis results. As a result, as the eccentricity of the structure increased, the eccentricity response was amplified in the high-rise part, and the bending and torsional behavior responses were complex in the low-order vibration mode. As a result of the analysis, the maximum displacement and story drift ratio increased due to the torsional behavior. The maximum story shear force and the story absolute maximum acceleration showed similarities for each analysis model according to the shape of the vibration mode of the analysis model.

• Journal of Korean Port Research
• /
• v.6 no.2
• /
• pp.43-64
• /
• 1992

This study discusses the interacting with deep water waves approaching from deep water based on the linear wave theory and steep sloping sea bottom floor by the numerical procedure. The results of particular interest are particle velocity and acceleration in x, y, z direction wave height amplification factor reflection coefficient and dimensionless pressure distribution on the steep sloping bottom with respect to the various incident wave angle. The wave loads relative to various bottom slopes, incident wave angles and wave periods on submerged breakwater and pipe are represented in comparison with mild sloping bottom the wave load parameters on the steep sloping bottom seemed to be influenced by variation of incident wave angle. In general the particle velocities and accelerations in x, y, z directions on the steep sloping bottom represented larger value or about two than those on the mild sloping bottom according to incident wave angle. However, the wave height amplification factors did not show distinct difference, but the slight variation with respect to the various incident angle showed on mild sloping bottom. The reflection coefficient increased with respect to increase of the incident angle on the steep sloping bottom the results also indicate that the very steep sloping beach produces a rather substantial amount of reflection as we expected. No significant variation of wave pressure was shown on the steep sloping bottom but it represented a certain amount of variation on the mild sloping bottom according to the various incident wave angle. The analysis at the OTEC site also showed similar results.

• The Journal of Korea Robotics Society
• /
• v.18 no.4
• /
• pp.427-435
• /
• 2023

This paper proposes a method of estimating the pose of a mobile robot by using a learning model. When estimating the pose of a mobile robot, wheel encoder and inertial measurement unit (IMU) data are generally utilized. However, depending on the condition of the ground surface, slip occurs due to interaction between the wheel and the floor. In this case, it is hard to predict pose accurately by using only encoder and IMU. Thus, in order to reduce pose error even in such conditions, this paper introduces a pose estimation method based on a learning model using data of the wheel encoder and IMU. As the learning model, long short-term memory (LSTM) network is adopted. The inputs to LSTM are velocity and acceleration data from the wheel encoder and IMU. Outputs from network are corrected linear and angular velocity. Estimated pose is calculated through numerically integrating output velocities. Dataset used as ground truth of learning model is collected in various ground conditions. Experimental results demonstrate that proposed learning model has higher accuracy of pose estimation than extended Kalman filter (EKF) and other learning models using the same data under various ground conditions.

• International Journal of High-Rise Buildings
• /
• v.12 no.3
• /
• pp.251-262
• /
• 2023

In response to China's "Regulations on the Management of Earthquake Resistance of Building Constructions" on the provision of eight types of important buildings to maintain functional after fortified earthquakes occur, "Guidelines for Seismic Design of post-quake functional buildings (Draft for Review)" distinguishes Class I and Class II buildings, and gives the performance objectives and seismic verification requirements for design earthquakes and severe earthquakes respectively. In this paper, a hospital and a school building are selected as examples to design according to the requirements of fortification of Intensity 8 and 7 respectively. Two design strategies, the seismic isolation scheme and energy dissipation scheme, are considered which are evaluated through elastic-plastic dynamic time-history analysis to meet the requirement of post-quake functional buildings. The results show that the seismic isolation design can meet the requirements in the above cases, and the energy dissipation scheme is difficult to meet the requirements of the "Guidelines" on floor acceleration in some cases, for which the scheme shall be made valid through the seismic resilience assessment. The research in this paper can provide a reference for designers to choose schemes for post-quake functional buildings.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.23 no.7
• /
• pp.87-94
• /
• 2019

Recently, the advance of accurate dynamic displacement measurement devices, such as GPS, computer vision, and optic laser sensor, has enhanced the structural monitoring technology. In this study, the dynamic displacement data was used to verify the applicability of the structural physical parameter estimation method through subspace system identification. The subspace system identification theory for estimating state-space model from measured data and physics-based interpretation for deriving the physical parameter of the estimated system are presented. Three-degree-freedom steel structures were fabricated for the experimental verification of the theory in this study. Laser displacement sensor and accelerometer were used to measure the displacement data of each floor and the acceleration data of the shaking table. Discrete state-space model generated from measured data was verified for precision. The discrete state-space model generated from the measured data extracted the floor stiffness of the building after accuracy verification. In addition, based on the story stiffness extracted from the state space model, five column stiffening and damage samples were set up to extract the change rate of story stiffness for each sample. As a result, in case of reinforcement and damage under the same condition, the stiffness change showed a high matching rate.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.5 no.5
• /
• pp.73-83
• /
• 2001

The mass and stiffness of upper wall-lower frame system(mixed building structures) change sharply at transfer floor due to different structural system in upper and lower part. These mixed building structures generally show the stiffness, weight or geometric vertical irregularities. The purpose of this study is to investigate the response characteristics of these structures by push-over analysis and nonlinear time history analysis. For four types of analysed models, only the variation of upper wall stories was considered. The conclusions of this study are following; (1) In the push-over analysis, yielding hinges in beams and columns of lower frame occurred at the base shear of similar magnitude in all models. But as the number of stories of upper wall increases, yielding hinges at ends of coupling beams were observed in the small magnitude of base shear. (2) In the nonlinear time history analysis, yielding of lower frame occurred at beams with as small ground acceleration as 55gal, and in upper walls yielding was concentrated on coupling beams and shear walls near the transfer floor. (3) As the number of stories of upper walls decreases, the story stiffness of the lower frames decreased relatively and the occurrence of soft stories in the lower frame was observed.