• 대한의용생체공학회:의공학회지
• /
• 제35권5호
• /
• pp.169-176
• /
• 2014

As body postures on bed affects various sleep related diseases, it is considered as important information when monitoring sleeping in daily life. Though there have already been a few approaches to monitor body postures on bed conventionally, the development for simple and unconstrained methods is still needed to realize the long-term daily monitoring. Focusing on the fact that ballistocardiogram changes depending on the body postures on bed, we developed a novel method to estimate body posturesusing extremely simple, film-type ballistocardiogram sensor which is based on polyvinylidene fluoride(PVDF) film. With 10 subjects, we performed two experiments. One was for an estimation test to show that body postures on bed can be estimated by ballistocardiogram, and the other was for a reproducibility test to present the feasibility of ballistocardiogram based body postures monitoring. To estimate body postures on bed, we made an individual template set of body postures by designating one ballistocardiogram (BCG) sample as a template in each postures. Then, we calculated Pearson's correlation coefficients between a sample and each templates and estimated the body posture of the sample by choosing a posture which corresponds to the most significant correlation coefficients. As a result, we estimated body postures on bed with 99.2% accuracy in average and found that the estimation using ballistocardiogram is reproducible.

• Journal of Advanced Marine Engineering and Technology
• /
• 제17권5호
• /
• pp.44-53
• /
• 1993

In this paper, we discuss on the control algorithm to make the wheeled inverse pendulum type mobile robot move in two dimensional plane. The robot considered in this paper has two independently driven wheels in same axel which suport and move it-self, and is assumed to have the fyro type sensor to know the inclination algle of the body and rotary encoders to know wheel's rotation angular velocity. The control algorithm is divided into three parts. The first part is for the posture and velocity control for forward-backward direction, the second is the steering control, and the last part is for the control of total system to track the given trajectory. We handle the running velocity control of the robot as part of the posture control to keep the balance because the posture relates deeply with the velocity and can be controlled by the velocities of the wheels. The control problem is analyzed as the tracking control, and the controller is realized with the state feedback and feed-forward of the reference velocity. Constructing the control system which contained one intergrator in forward path, we also realized the control system without observer for the estimation of the accumulated errors in the inclination angle of the body. To prevent the robot from being unstable state by sudden variation of the reference velocity when it starts and stops, or changes velocity, the reference velocity of which acceleration is slowly changing, is ordered to the robot. To control its steering, we give the different reference velocities for both wheels which are calculated from the desired angular velocity of the body. Finally, we presents the experimental results of the experimental robot Yamabico Kurara in which the proposed control algorithm had been implemented.

• 대한인간공학회지
• /
• 제30권3호
• /
• pp.419-426
• /
• 2011

Objective: The objective of this study was to evaluate a work of guards, using an ergonomic method(work analysis and posture analysis). Background: Most studies about guards were conducted in the field of medical, problems of shift, and the physical problems of old workers and social problems. But, guards consist of vulnerability group so it needs an ergonomic research in musculoskeletal disorders. Method: A head of an ergonomic estimation was work analysis(determination of combined task, work tool, work time and frequency of combined task) and posture analysis(upper body and lower body) of workers based on the video. Results: The result showed that combined task of guards was classification of patrolling, security, cleaning and waiting. The security indicated the highest ratio in the work time of combined tasks. The results of posture analysis for guards indicated high value in neutral. But, lower arm indicated high value in bending(left: 59%, right: 50%). Conclusion: The results of ergonomic methods indicated that guards' physical work load was not high during work, but comfortable work environment would be required for old guards. Application: If an ergonomic rule can be integrated into existing work environments, the risk of occupational injuries and stress will be reduced.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제11권10호
• /
• pp.409-418
• /
• 2022

본 논문에서는 연주자 자세의 그라운드 트루스 획득을 위한 반자동 주석 방법인 SAAnnot-C3Pap를 제안한다. 기존 음악 도메인에서 2차원 관절 위치에 대한 그라운드 트루스를 획득하기 위하여 2차원 자세 추정 방법인 오픈포즈를 활용하거나 수작업으로 라벨링 하였다. 하지만 기존의 오픈포즈와 같은 자동 주석 방법은 빠르지만 부정확한 결과를 보인다는 단점이 있고, 사용자가 직접 주석을 생성하는 수작업 주석화의 경우 많은 노동력이 필요하다는 한계점이 있다. 따라서 본 논문에서는 그 둘의 절충 방안인 반자동 주석화 방법인 SAAnnot-C3Pap을 제안한다. 제안하는 SAAnnot-C3Pap은 크게 3가지 과정으로 오픈포즈를 사용하여 자세를 추출하고, 추출된 부분 중 오류가 있는 부분을 슈퍼바이즐리를 사용하여 수정한 뒤, 오픈포즈와 슈퍼바이즐리의 결과값을 동기화하는 과정을 수행한다. 제안하는 방법을 통하여 오픈포즈에서 발생하는 잘못된 2차원 관절 위치 검출 결과를 교정할 수 있었고, 2명 이상의 사람을 검출하는 문제를 해결하였으며, 연주 자세 그라운드 트루스 획득이 가능하였다. 실험에서는 반자동 주석 방법인 오픈포즈와 본 논문에서 제안하는 SAAnnot-C3Pap의 결과를 비교·분석한다. 비교 결과, 제안하는 SAAnnot-C3Pap는 오픈포즈로 잘못 수집된 자세 정보를 개선한 결과를 보였다.

• 한국재난정보학회:학술대회논문집
• /
• 한국재난정보학회 2023년 정기학술대회 논문집
• /
• pp.349-350
• /
• 2023

본 논문에서는 FMCW(Frequency Modulated Continuous Wave) 레이더를 사용하여 재난 상황에서 누워 있는 사람의 다양한 각도의 자세를 통해 사람의 상태를 파악하거나 위치를 추정하고자하였다. 사람의 세 가지 누운 자세 데이터를 전처리하고 이미지로 변환한 데이터를 CNN(Convolutional Neural Network) 1D 모델로 학습시켜 누운 자세를 다양한 각도에서 구별할 수 있는지 분석하여 확인하고자하였으며, 분석 결과 CNN 1D 모델은 99.27%의 정확도를 보였다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제5권4호
• /
• pp.316-321
• /
• 2005

We designed and implemented fish robots for various purposes such as autonomous navigation, maneuverability control, posture balancing and improvement of quick turns in a tank of 120 X 120 X 180cm size. Typically, fish robots have 30-50 X 15-25 X 10-20cm dimensions; length, width and height, respectively. It is essential to have the ability of quick and smooth turning to avoid collision with obstacles or walls of the water pool at a close distance. Infrared distance sensors are used to detect obstacles, magneto-resistive sensors are used to read direction information, and a two-axis accelerometer is mounted to compensate output of direction sensors. Because of the swing action of its head due to the tail fin movement, the outputs of an infrared distance sensor contain a huge amount of noise around true distances. With the information from accelerometers and e-compass, much improved distance data can be obtained by fuzzy logic based estimation. Successful swimming and smooth turns without collision demonstrated the effectiveness of the distance estimation.

• 대한기계학회논문집A
• /
• 제40권8호
• /
• pp.703-710
• /
• 2016

로보틱스 연구에서, 모바일 로봇의 모션 제어를 위해서는 로봇의 실제 위치를 정확히 추정하는 것이 중요하다. 이를 위해 본 연구에서는, 두 개의 서로 다른 센서 데이터를 칼만필터로 융합하여 로봇의 위치인식을 개선하는 연구를 진행한다. 칼만필터로 융합한 두 개의 센서 측정값은 카메라 영상으로부터 측정된 모바일 로봇의 전역(global) 위치 좌표(x, y)값과 모바일 로봇 바퀴에 부착된 엔코더로부터 측정된 로봇의 직선 및 각속도 값이다. 다음으로 칼만필터로부터 계산된 모바일 로봇의 위치값을 모바일 로봇의 자세 안정화에 피드백하여 모션 제어의 퍼포먼스를 향상시켰다. 최종적으로 논문에서 제안한 센서융합 위치인식 기술과 모션제어기를 실제 로봇에 적용하여 실험적으로 검증하였다. 또한 모션제어에 단일 센서를 피드백으로 사용한 경우와 칼만필터로 융합한 위치 값을 사용한 경우를 비교하므로 칼만필터 기반 센서 융합 기술을 사용한 경우의 퍼포먼스 향상을 확인하였다.

• 로봇학회논문지
• /
• 제14권1호
• /
• pp.1-7
• /
• 2019

Recently, smart factories have attracted much attention as a result of the 4th Industrial Revolution. Existing factory automation technologies are generally designed for simple repetition without using vision sensors. Even small object assemblies are still dependent on manual work. To satisfy the needs for replacing the existing system with new technology such as bin picking and visual servoing, precision and real-time application should be core. Therefore in our work we focused on the core elements by using deep learning algorithm to detect and classify the target object for real-time and analyzing the object features. We chose YOLO CNN which is capable of real-time working and combining the two tasks as mentioned above though there are lots of good deep learning algorithms such as Mask R-CNN and Fast R-CNN. Then through the line and inside features extracted from target object, we can obtain final outline and estimate object posture.

• Journal of information and communication convergence engineering
• /
• 제21권4호
• /
• pp.337-345
• /
• 2023

In this study, we formulated a method that evaluates Taekwondo Poomsae performance using a series of choreographed training movements. Despite recent achievements in 3D human pose estimation (HPE) performance, the analysis of human actions remains challenging. In particular, Taekwondo Poomsae action analysis is challenging owing to the absence of time synchronization data and necessity to compare postures, rather than directly relying on joint locations owing to differences in human shapes. To address these challenges, we first decomposed human joint representation into joint rotation (posture) and limb length (body shape), then synchronized a comparison between test and reference pose sequences using DTW (dynamic time warping), and finally compared pose angles for each joint. Experimental results demonstrate that our method successfully synchronizes test action sequences with the reference sequence and reflects a considerable gap in performance between practitioners and professionals. Thus, our method can detect incorrect poses and help practitioners improve accuracy, balance, and speed of movement.

• 대한인간공학회지
• /
• 제30권3호
• /
• pp.427-435
• /
• 2011

Objective: The objective of this study was to evaluate a work of building cleaners using the ergonomic methods. Background: Previous studies about cleaning worker describe typical physical demands of this work. They found that the most significant risk factors associated with the physical work of cleaners are static loads and repetitive movements and high output of force. Method: A head of ergonomics estimation was work analysis(define of combined task, work tool, work time and frequency of combined task) and posture analysis of worker. Results: The results showed that combined task of building cleaners was classification sweeping, mopping(wet), mopping(oil), moving barrels/carts, dumping trash bags, scrubbing, arrangement of cleaning tool, arrangement of circumferential, moving of cleaning tool, and waiting. The work time of combined task such as mopping(wet) and scrubbing indicated high ratio. The posture analysis of building cleaners indicated high value in bending of the head, lower arm, and hands. Conclusion: The findings appear to indicate that building cleaner were related to high risk of work-related musculoskeletal disorders. So, building cleaner would be required an interventional strategy, improvement of cleaning tools and working environment. Application: If ergonomics rule can be integrated into existing cleaning tools and work environments, the risk of occupational injuries will be reduced.