• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10b
• /
• pp.990-995
• /
• 1990

A minimum-time trajectory planning for two robot arms with designated paths and coordination is proposed. The problem considered in this paper is a subproblem of hierarchically decomposed trajectory planning approach for multiple robots : i) path planning, ii) coordination planning, iii) velocity planning. In coordination planning stage, coordination space, a specific form of configuration space, is constructed to determine collision region and collision-free region, and a collision-free coordination curve (CFCC) passing collision-free region is selected. In velocity planning stage, normal dynamic equations of the robots, described by joint angles, velocities and accelerations, are converted into simpler forms which are described by traveling distance along collision-free coordination curve. By utilizing maximum allowable torques and joint velocity limits, admissible range of velocity and acceleration along CFCC is derived, and a minimum-time velocity planning is calculated in phase plane. Also the planning algorithm itself is converted to simple numerical iterative calculation form based on the concept of neural optimization network, which gives a feasible approximate solution to this planning problem. To show the usefulness of proposed method, an example of trajectory planning for 2 SCARA type robots in common workspace is illustrated.

• Journal of Korea Water Resources Association
• /
• v.55 no.spc1
• /
• pp.1251-1260
• /
• 2022

Recently, due to the deterioration of watermains and the detachment of scale which is accumulated on the watermain surface, water quality accidents in a water supply network occur frequently. As scale accumulated on watermains is stabilized, it may not cause water quality accidents under the normal operating condition. However, due to water hammer or transient flow caused by the abrupt velocity and/or direction of flow change, it can be detached from the watermain surface resulting in water quality accidents. To prevent these kinds of water quality accidents, it is required to remove scale by watermain cleaning regularly. Many researches about flushing which is the most popular water cleaning method are focused on the desirable velocity criteria and the cleaning condition to accomplish the effect of flushing whereas less amount of research effort is given to develop a method to consider whether the desirable velocity for flushing can be obtained before flushing is performed. During flushing, the major and minor headloss is occurred when flushing water flows through a hydrant or drain valve. These headloss may slow down the velocity of flushing water so that it can reduce the flushing effect. Thus, in this study, we suggest a method to simulate the flow velocity of flushing water using "MinorLoss Coefficient" and "Emitter Coefficient" in EPANET. The suggested method is applied to a sample network and the water supply network of "A" city in Korea to compare the flushing effect between "flushing through a hydrant" and "flushing through a drain valve". In case of "flushing through a hydrant", if the hydraulic condition ocurring from a watermain pipe connecting to the inlet pipe of a hydrant to the outlet of a hydrant is not considered, the actual flowrate and velocity of a flow is less than the simulated flowrate and velocity of a flow. In case of "flushing through a drain valve", the flushing velocity and flowrate can be easily simulated and the difference between the simulated and the actual velocity and flowrate is not significant. Also, "flushing through a drain valve" is very effective to flushing a long-length pipe section because of its efficiency to obtain the flushing velocity. However, the number and location of a drain valve is limited compared to a hydrant so that "flushing through a drain valve" has a limited application in the field. For this reason, the engineer should consider various field conditions to come up with a proper flushing plan.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.6
• /
• pp.1380-1386
• /
• 2015

Recently, the importance of users' geographic location information has been highlighted with a rapid increase of online social network services. In this paper, by utilizing geo-tagged tweets that provides high-precision location information of users, we first identify both Twitter users' exact location and the corresponding timestamp when the tweet was sent. Then, we analyze a relationship between the tweet frequency and the average user velocity. Specifically, we introduce a tweet-frequency computing algorithm, and show analysis results by country and by city. As a main result, it is shown that the tweet frequency according to user velocity follows a power-law distribution (i.e., Zipf' distribution or a Pareto distribution). In addition, by performing a comparison between the United States and Japan, one can see that the exponent of the distribution in Japan is smaller than that in the United States.

• Geophysics and Geophysical Exploration
• /
• v.26 no.2
• /
• pp.84-93
• /
• 2023

The supervised learning-based deep-learning seismic inversion techniques have demonstrated successful performance in synthetic data examples targeting small-scale areas. The supervised learning-based deep-learning seismic inversion uses time-domain wavefields as input and subsurface velocity models as output. Because the time-domain wavefields contain various types of wave information, the data size is considerably large. Therefore, research applying supervised learning-based deep-learning seismic inversion trained with a significant amount of field-scale data has not yet been conducted. In this study, we predict subsurface velocity models using Laplace-domain wavefields as input instead of time-domain wavefields to apply a supervised learning-based deep-learning seismic inversion technique to field-scale data. Using Laplace-domain wavefields instead of time-domain wavefields significantly reduces the size of the input data, thereby accelerating the neural network training, although the resolution of the results is reduced. Additionally, a large grid interval can be used to efficiently predict the velocity model of the field data size, and the results obtained can be used as the initial model for subsequent inversions. The neural network is trained using only synthetic data by generating a massive synthetic velocity model and Laplace-domain wavefields of the same size as the field-scale data. In addition, we adopt a towed-streamer acquisition geometry to simulate a marine seismic survey. Testing the trained network on numerical examples using the test data and a benchmark model yielded appropriate background velocity models.

• Journal of the Korean Geotechnical Society
• /
• v.36 no.8
• /
• pp.7-15
• /
• 2020

Constructing the maximum displacement and shear force database for the seismic performance of building with soil-structure interaction under varied earthquake scenarios and geotechnical conditions is critical in developing the neural network-based prediction models. However, using the available 3D FEM-based computer simulation techniques causes high computation costs in developing the database. This study introduces the framework of developing the artificial neural network (ANN) model to predict the seismic performance of building at given Poisson's ratio and shear wave velocity of soil. The simple Single-Degree-Of-Freedom system was used to develop the database and the performance of the developed neural network model is discussed through the evaluated coefficient of determination (R²). In addition, ANN models were developed for 90~100% percentile of the database to assess the accuracy of the developed ANN models in each percentile.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.10
• /
• pp.2529-2547
• /
• 2012

In this paper, we use a quadrotor-based mobile sink to gather sensor data from the terrestrial deployed wireless sensor network. By analyzing the flight features of the mobile sink node, we theoretically study the flight constraints of height, velocity, and trajectory of the mobile sink node so as to communicate with the terrestrial wireless sensor network. Moreover, we analyze the data amount which the mobile sink can send when it satisfies these flight constraints. Based on these analysis results, we propose a data acquisition approach for the mobile sink node, which is discussed detailed in terms of network performance such as the transmission delay, packet loss rate, sojourning time and mobile trajectory when given the flying speed and height of the mobile sink node. Extensive simulation results validate the efficiency of the proposed scheme.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10a
• /
• pp.7-12
• /
• 1990

In this paper, we propose a new manipulator control scheme based on the CMAG neural network. The proposed control consists of two components. The feedforward component is an output of trained CMAC neural network and the feedback component is a modified sliding mode control. The CMAC accepts the position, velocity and acceleration of manipulator as input and outputs two values for the controller : One is the nominal torque used for feedforward compensation(M1 network) and the other is the inertia matrix related information used for the feedback component(M2 network). Since the used control algorithm guarantees the robust trajectory tracking in spite of modeling errors, the CMAC mapping errors due to the memory limitation are little worth consideration.

• Journal of information and communication convergence engineering
• /
• v.5 no.3
• /
• pp.209-214
• /
• 2007

In this paper, we proposed a new MANET (Mobile Ad hoc Networks) technology of routing protocol. The MANET has a mobility formation of mobile nodes in the wireless networks. Wireless network have two types architecture: the Tree based multicast and shared tree based. The two kind's architecture of general wireless networks have difficult to solve the problems existing in the network, such as connectivity, safety, and reliability. For this purpose, as using that ADDMRP (Ad hoc network Doppler effect-based for Dynamic Multicast Routing Protocol), this study gives the following suggestion for new topology through network durability and Omni-directional information. The proposed architectures have considered the mobility location, mobility time, density, velocity and simultaneous using node by Doppler effects and improved the performance.

• Wind and Structures
• /
• v.15 no.1
• /
• pp.43-64
• /
• 2012

Many potential small wind turbine locations are near obstacles such as buildings and shelterbelts, which can have a significant, detrimental effect on the local wind climate. A neural network-based model has been developed which predicts mean wind speed and turbulence intensity at points in an obstacle's region of influence, relative to unsheltered conditions. The neural network was trained using measurements collected in the wakes of 18 scale building models exposed to a simulated rural atmospheric boundary layer in a wind tunnel. The model obstacles covered a range of heights, widths, depths, and roof pitches typical of rural buildings. A field experiment was conducted using three unique full scale obstacles to validate model predictions and wind tunnel measurements. The accuracy of the neural network model varies with the quantity predicted and position in the obstacle wake. In general, predictions of mean velocity deficit in the far wake region are most accurate. The overall estimated mean uncertainties associated with model predictions of normalized mean wind speed and turbulence intensity are 4.9% and 12.8%, respectively.

• 제어로봇시스템학회:학술대회논문집
• /
• 1995.10a
• /
• pp.166-169
• /
• 1995

An intelligent system which is an integration of multi-functional instrumentation (MFI) and a neural network is discussed. According to some experiments of temperature and wind velocity it is clear that this system can learn the data structure of two parameters above. So it makes good performances for estimations of non-sample data.