• 전기학회논문지P
• /
• 제59권1호
• /
• pp.29-34
• /
• 2010

This paper presents an intelligent learning controller for repetitive walking motion of biped walking robot. The proposed learning controller consists of an iterative learning controller and a direct learning controller. In the iterative learning controller, the PID feedback controller takes part in stabilizing the learning control system while the feedforward learning controller plays a role in compensating for the nonlinearity of uncertain biped walking robot. In the direct learning controller, the desired learning input for new joint trajectories with different time scales from the learned ones is generated directly based on the previous learned input profiles obtained from the iterative learning process. The effectiveness and tracking performance of the proposed learning controller to biped robotic motion is shown by mathematical analysis and computer simulation with 12 DOF biped walking robot.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1995년도 추계학술대회 논문집
• /
• pp.1039-1042
• /
• 1995

In real time application such as motion control, it is hard to find the application of non-causal filtering due to its need for future position data, even though it shows wide usage in off-line digital signal processing. Recently, some of motion control areas such as learning and repetitive control use non-causal filtering technique in their application. these kinds of zero-lag non-causal filter application are very usful not only to reduce the machine vibration, but also to increase control accuracy with comparatively less work. In this paper, genuine method to implement zero-lag non-causal filter in a CNC is introduced. Also the variation of this implementation for the learning operation is suggested to give the NC better control performance for a specific job. By adopting the new NC architecture call Soft-NC, all these implementions are made possible here, and especially large memory requirement which hinders their usage for many years is no longer barrier in their real world application.

• Structural Engineering and Mechanics
• /
• 제16권4호
• /
• pp.423-440
• /
• 2003

Arbitrary Lagrangian Eulerian finite element methods gain interest for the capability to control mesh geometry independently from material geometry, the ALE methods are used to create a new undistorted mesh for the fluid domain. In this paper we use the ALE technique to solve fuel slosh problem. Fuel slosh is an important design consideration not only for the fuel tank, but also for the structure supporting the fuel tank. "Fuel slosh" can be generated by many ways: abrupt changes in acceleration (braking), as well as abrupt changes in direction (highway exit-ramp). Repetitive motion can also be involved if a "sloshing resonance" is generated. These sloshing events can in turn affect the overall performance of the parent structure. A finite element analysis method has been developed to analyze this complex event. A new ALE formulation for the fluid mesh has been developed to keep the fluid mesh integrity during the motion of the tank. This paper explains the analysis capabilities on a technical level. Following the explanation, the analysis capabilities are validated against theoretical using potential flow for calculating fuel slosh frequency.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2019년도 춘계 학술논문 발표대회
• /
• pp.123-124
• /
• 2019

In the labor-intensive construction site, potential threats of the musculoskeletal diseases mainly caused by various repetitive physical tasks, vulnerable environment, and the aging of the labor worker exist. However, quantitative measuring method of construction labor worker's work posture has not been improved yet. This study proposed musculoskeletal measuring method by using wearable motion recognition sensor for quantitative evaluation and analysis of working posture of construction workers. This method is expected to be used as a basic data for posture analysis and prevention construction safety accidents, as well as physical workload and labor productivity analysis by labor work type.

• 대한인간공학회:학술대회논문집
• /
• 대한인간공학회 1996년도 춘계학술대회논문집
• /
• pp.301-306
• /
• 1996

Different patterns of dynamic trunk performance between normal subjects and low-back pain (LBP) patients have been investigated. Ten healthy subjects and ten LBP patients were recruited for this study. An experiment was designed to quantify the dynamic motion of the back and hip during repetitive flexion/extension of the trunk. The angular velocity, angular acceleration and the phase angle difference between the hip and back were recorded as dependent measures via electro- mechanical devices. Results showed the significant differences in the velocity and acceleration of the hip and back and the phase angle between normal subjects and LBP patients. The consistency of kinematic variables during motion cycles was also examined in terms of variance ratio (Hershler and Milner, 1978). Based on the results of the study, these quantifiable variables such as trunk kinematics and hip-spine coordination can be developed as a medical tool to identify LBP patients in addition to current imaging techniques.

• 산업기술연구
• /
• 제31권A호
• /
• pp.11-19
• /
• 2011

A new PCD (Pitch Circle Diameter) adjustment system has been designed in this study. Traditional system requires repetitive same works in changing automobile tires. The system proposed in this study requires just one simultaneous work to finish the job. We designed the system using Solidworks CAD software. We further investigate the jobs for both cases through work factor method. Based on careful clarification of the job, work factor units were assigned to each work factor. Even if further analysis should be processed for actual field application, the mechanism proposed in this study may provide good reference for practical usage.

• 한국운동역학회지
• /
• 제26권4호
• /
• pp.369-375
• /
• 2016

Objective: The purpose of this study was to investigate the possibility of injuries and the types of movement related to damage by body parts, and to prepare for prevention of injuries and development of a training program. Method: For this study, the experiment was conducted according to levels of 60 percentages (ST) and 85 percentages (MA) and 10 subjects from the Korean elite national weightlifting team were included. Furthermore, we analyzed joint moment and muscle activation pattern with three-dimensional video analysis. Ground reaction force and EMG analyses were performed to measure the factors related to injuries and motion. Results: Knee reinjuries such as anterior cruciate ligament damage caused by deterioration of the control ability for the forward movement function of the tibia based on the movement of the biceps femoris when the rectus femoris is activated with the powerful last-pull movement. In particular, athletes with previous or current injuries should perceive a careful contiguity of the ratio of the biceps femoris to the rectus femoris. This shows that athletes can exert five times greater force than the injury threshold in contrast to the inversion moment of the ankle, which is actively performed for a powerful last pull motion and is positively considered in terms of intentional motion. It is activated by excessive adduction and internal rotation moment to avoid excessive abduction and external rotation of the knee at lockout motion. It is an injury risk to muscles and ligaments, causing large adduction moment and internal rotation moment at the knee. Adduction moment in the elbow joint increased to higher than the injury threshold at ST (60% level) in the lockout phase. Hence, all athletes are indicated to be at a high risk of injury of the elbow adductor muscle. Lockout motion is similar to the "high five" posture, and repetitive training in this motion increases the likelihood of injuries because of occurrence of strong internal rotation and adduction of the shoulder. Training volume of lockout motion has to be considered when developing a training program. Conclusion: The important factors related to injury at snatch include B/R rate, muscles to activate the adduction moment and internal rotation moment at the elbow joint in the lockout phase, and muscles to activate the internal rotation moment at the shoulder joint in the lockout phase.

• 대한의용생체공학회:의공학회지
• /
• 제28권2호
• /
• pp.218-228
• /
• 2007

Modem concepts of gait rehabilitation after stroke favor a task-specific repetitive approach. In practice, the required physical effort of the therapists limits the realization of this approach. Therefore, a mechanized gait trainer enabling nonambulatory patients to have the repetitive practice of a gait-like movement without overstraining therapists was constructed. In this study, we developed an active gait training system for patients with gait disorder. This system provides joint movements to patients who cannot carry out an independent gait. It provides a normal stance-swing ratio of 60:40 using an eccentric configuration of two gears. Joint motions of the knee and the ankle were evaluated with using the 3D motion analysis system and compared with the results from the multi-body dynamics simulation. In addition, clinical investigations were also performed for low stroke patients during the 6-week gait training. Results from the dynamics simulation showed that joint movements of the knee and the ankle were affected by the gear size, the step length and the length of the foot plate, except the radius of curvature of the foot guide plate. Also, the 6-week gait training revealed relevant improvements of the gait ability in all low subjects. Functional ambulation category levels of subjects after training were 2 in three patients and 1 in a patient. The developed active gait trainer seems feasible as an adjunctive tool in gait rehabilitation after stroke.

• 대한전자공학회논문지SD
• /
• 제45권6호
• /
• pp.37-42
• /
• 2008

H.264/AVC 복호기에서는 인트라 예측기 뿐만 아니라 움직임 보상기, 디블럭킹 필터 등 각 IP들이 복호화를 위한 참조 영상 값들을 필요로 한다. 이들 IP들은 참조 영상을 읽어들이기 위하여 외부 메모리에 빈번하게 접근하는데, 이때문에 시스템 동작 속도도 낮아지고 전력 소모도 증가한다. 본 논문에서는 공통적이고 반복적인 블록의 재사용을 통하여 연산량을 줄이고 전력 소모 및 메모리 대역폭을 최소화하도록 외부 메모리를 사용하지 않는 움직임 보상기와 연계한 인트라 예측기를 제안하였다. 제안된 인트라 예측기는 기존에 비해 $45%\;{\sim}\;75%$ 가량 사이클 수를 감소시켰다.

• 한국유체기계학회 논문집
• /
• 제20권1호
• /
• pp.59-64
• /
• 2017

Diverter is an essential element in gravimetric calibration method of flowmeter. Error of diverter are influenced by flow velocity profile of nozzle outlet, motion velocity of diverter and detecting location. That's why, time detection position of diverter is tuned through repetitive test for minimizing error of diverter. Further the diverter must be compared with the other institutions test since the influence on the accuracy of the flow meter used in the test. In this paper, errors (flow velocity profile of nozzle outlet, motion velocity of diverter and detecting location) of diverter are decreased by produced uni-direction diverter and error of gravimetric calibration system is decreased. Uni-direction diverter is calibrated by gravimetric calibration system with precision flowmeter, the flowmeter is calibrated by pipe prover and other institutions and uni-direction diverter is evaluated. Uni-direction diverter is not influenced by flow velocity profile of nozzle outlet, motion velocity of diverter and detecting location. As a result, Uni-direction diverter can calibrate in wider scope since increasing ratio of maximum and minimum flow rate of uni-direction diverter.