• 로봇학회논문지
• /
• 제3권4호
• /
• pp.287-299
• /
• 2008

People have expected a humanoid robot to move as naturally as a human being does. The natural movements of humanoid robot may provide people with safer physical services and communicate with persons through motions more correctly. This work presented a methodology to generate the natural motions for a humanoid robot, which are converted from human motion capture data. The methodology produces not only kinematically mapped motions but dynamically mapped ones. The kinematical mapping reflects the human-likeness in the converted motions, while the dynamical mapping could ensure the movement stability of whole body motions of a humanoid robot. The methodology consists of three processes: (a) Human modeling, (b) Kinematic mapping and (c) Dynamic mapping. The human modeling based on optimization gives the ZMP (Zero Moment Point) and COM (Center of Mass) time trajectories of an actor. Those trajectories are modified for a humanoid robot through the kinematic mapping. In addition to modifying the ZMP and COM trajectories, the lower body (pelvis and legs) motion of the actor is then scaled kinematically and converted to the motion available to the humanoid robot considering dynamical aspects. The KIST humanoid robot, Mahru, imitated a dancing motion to evaluate the methodology, showing the good agreement in the motion.

• 한국멀티미디어학회논문지
• /
• 제20권5호
• /
• pp.769-781
• /
• 2017

Deep learning networks like Convolutional Neural Networks (CNNs) show successful performances in many computer vision applications such as image classification, object detection, and so on. For implementation of deep learning networks in embedded system with limited processing power and memory, deep learning network may need to be simplified. However, simplified deep learning network cannot learn every possible scene. One realistic strategy for embedded deep learning network is to construct a simplified deep learning network model optimized for the scene images of the installation place. Then, automatic training will be necessitated for commercialization. In this paper, as an intermediate step toward automatic training under fisheye camera environments, we study more precise human localization in fisheye images, and propose an accurate human localization method, Automatic Ground-Truth Labelling Method (AGTLM). AGTLM first localizes candidate human object bounding boxes by utilizing GoogLeNet-LSTM approach, and after reassurance process by GoogLeNet-based CNN network, finally refines them more correctly and precisely(tightly) by applying saliency object detection technique. The performance improvement of the proposed human localization method, AGTLM with respect to accuracy and tightness is shown through several experiments.

• 대한인간공학회지
• /
• 제18권3호
• /
• pp.141-152
• /
• 1999

In this paper, We performed the human body dynamic modelling for the realistic animation based on the dynamical behavior of human body, and designed controller for the effective control of complicate human dynamic model. The human body was simplified as a rigid body which consists of 18 actuated degrees of freedom for the real time computation. Complex human kinematic mechanism was regarded as a composition of 6 serial kinematic chains : left arm, right arm, support leg, free leg, body, and head. Based on the this kinematic analysis, dynamic model of human body was determined using Newton-Euler formulation recursively. The balance controller was designed in order to control the nonlinear dynamics model of human body. The effectiveness of designed controller was examined by the graphical simulation of human walking motion. The simulation results were compared with the model base control results. And it was demonstrated that, the balance controller showed better performance in mimicking the dynamic motion of human walking.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.764-767
• /
• 2009

Motion capture systems allow to measure the precise position of markers on the human body in real time. These captured motion data, the marker position data, have to be fitted by a human skeleton model to represent the motion of the human. Typical human skeleton models approximate the joints using a ball joint model. However, because this model cannot represent the human skeleton precisely, errors between the motion data and the movements of the simplified human skeleton model happen. We propose in this paper a method for measuring a translation component of wrist, and elbow joints on upper limb using optical motion capture system. Then we study the errors between the ball joint model and acquired motion data. In addition, we discuss the problem to estimate motion of human joint using optical motion capture system.

• 대한의용생체공학회:의공학회지
• /
• 제20권4호
• /
• pp.485-493
• /
• 1999

Human can generate various posture and motion with nearly 350 muscle pairs. From the viewpoint of mechanisms, the human skeleton mechanism represents great kinematic and dynamical complexity. Physical and behavioral fidelity of human motion requires dynamically accurate modeling and controling. This paper describes a mathematical modeling, and dynamic simulation of human body. The human dynamic model is simplified as a rigid body consisting of 18 actuated degrees of freedom for the real time computation. Complex kinematic chain of human body is partitioned as 6 serial kinematic chains that is, left arm, right arm, support leg, free leg, body, and head. Modeling is developed based on Newton-Euler formulation. The validity of proposed dynamic model, which represents mathematically high order differential equation, is verified through the dynamic simulation.

• Journal of Magnetics
• /
• 제21권4호
• /
• pp.516-523
• /
• 2016

In this paper, coupling factors are calculated based on numerical analysis in order to assess various non-uniform low-frequency magnetic field exposure situations. Two types of non-uniform magnetic field sources are considered; circular coil and parallel wires with balanced currents. For each magnetic field source, source current values are determined so that reference magnetic field magnitude can be measured at the specified point on the human model. Various exposure situations are investigated by changing parameters such as the distance between source and human model, radius of circular coil, and the gap between parallel wires. For equivalent human models, prolate spheroid model and simplified human model from IEC 62311 standard are used. The calculated coupling factor values are compared with those obtained by 2D uniform disk human model, and the dependence of coupling factor on the choice of equivalent human model is analyzed.

• 한국안전학회지
• /
• 제13권4호
• /
• pp.230-239
• /
• 1998

Though limited, Model Human Processor (MHP) has been used to explain the complex users' behaviors during human-computer interactions in a simplified manner. MHP consists of perceptual, cognitive and motor systems, each with processors and memories interacting with each other in serial or parallel mode. The important parameters of memory include the storage capacity, the decay time, and the code type of a memorized item. The important parameter of a processor is the cycle time. Using these features of the model, this study developed a computerized cognitive process simulator to predict the cognitive process time of a class match task process. An experimental validity test result showed that the mean prediction time for cognitive process of the class match task simulated 50 times by the simulator was consistent with the mean cognitive process time of the same task performed by 37 subjects. Animation of the data flow during the class match task simulation will help understand the invisible human cognitive process.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2013년도 춘계학술대회 논문집
• /
• pp.516-517
• /
• 2013

Whole body vibration is very important for operators in construction equipment (CE) industry. There is ISO 2631 regulation to protect operators of CE. Recently WBV is one of critical performance parameters of CE to give operators much better comfortable working environment. And there are many kinds of numerically simplified human body model for the motor industry. We applied one human body model in ISO 5982 for the CE development at early stage. And we've checked the validity of this model to consider WBV by the operator comfort point of view.

• 한국소음진동공학회논문집
• /
• 제22권6호
• /
• pp.502-508
• /
• 2012

An analytical model for a human body is important to predict muscle and joint forces. Because it is difficult to estimate muscle or joint forces from a human body, the objective of this study is the development of a reliable analytical model for a human body to evaluate the lower extremity muscle and joint forces. The musculoskeletal system of the human lower extremity is modeled as a multibody system employing the Hill-type muscle model. Muscle forces are determined to minimize energy consumption, and we assume that motion is constrained in the sagittal plane. Muscle forces are calculated through an equilibrium analysis while rising from a seated position. The musculoskeletal model consists of four segments. Each segment is a rigid body and connected by frictionless revolute joints. Muscles of the lower extremity are simplified to seven muscles with those that are not related to the sagittal plane motion are ignored. Muscles that play a similar role are combined together. The results of the present study are compared with experimental results to validate the lower extremity model and the assumptions of the present study.

• 조명전기설비학회논문지
• /
• 제28권3호
• /
• pp.63-71
• /
• 2014

A single-inverted-pendulum model is presented to simulate and predict the passive response of human balance control. This simplified biomechanical model was comprised of a torsional spring and damper, and a lump mass. An estimation of frequency response function was conducted to parameterize the complexity. The frequency domain identification method is used to identify the parameters of the model. The equivalent viscoelastic parameters of standing body were obtained and there was good conformity between the simulation and experimental result.