• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.1539-1543
• /
• 2004

This paper describes platform overview, system integration and dynamic walking control of the humanoid robot, KHR-2 (KAIST Humanoid Robot - 2). We have developed KHR-2 since 2003. KHR-2 has totally 41 DOF (Degree Of Freedom). Each arm including a hand has 11 DOF and each leg has 6 DOF. Head and trunk also has 6 DOF and 1 DOF respectively. In head, two CCD cameras are used for eye. In order to control all joints, distributed control architecture is adopted to reduce the computation burden of the main controller and to expand the devices easily. The main controller attached its back communicates with sub-controllers in real-time by using CAN (Controller Area Network) protocol. We used Windows XP as its OS (Operating System) for fast development of main control program and easy extension of peripheral devices. And RTX, HAL(Hardware Abstraction Layer) extension program, is used to realize the real-time control in Windows XP environment. We present about real-time control of KHR-2 in Windows XP with RTX and basic walking control algorithm. Details of the KHR-2 are described in this paper.

• 한국지능시스템학회논문지
• /
• 제24권4호
• /
• pp.450-456
• /
• 2014

휴머노이드 로봇은 인간-로봇 상호작용에 가장 효과적인 로봇 플랫폼이지만 20개 이상의 관절로 구성되어 있을 만큼 복잡한 구조여서 전통적인 역기구학적 방법으로 안정되면서도 정교한 자세를 생성하기가 어렵다. 본 논문에서는 고속 연산최적화 기법인 Particle Swarm Optimization 기법을 사용해서 앞쪽 지면에 놓인 물체를 단측지지 상태로 상체를 굽혀서 왼팔이나 오른팔로 집는 고난도의 자세를 생성하고, 이를 상용 휴머노이드 로봇 플랫폼에 적용하여 검증함으로써 제안 된 방법의 적용 가능성을 확인한다.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2004년도 추계학술대회
• /
• pp.579-584
• /
• 2004

In this paper, we present the mechanical, electrical system design and system integration of controllers including sensory devices of the humanoid, KHR-2 (KAIST Humanoid Robot - 2). We have developed KHR-2 since 2003. Total number of DOF of KHR-2 is 41. Each arm including a hand has 11 DOF and each leg has 6 DOF. Head and trunk also has 6 DOF and 1 DOF respectively. In head, two CCD cameras are used for eye. To control all axes efficiently, distributed control architecture is used to reduce computation burden of main controller and to expand devices easily. So we developed the sub-controller as a servo motor controller and a sensor interfacing devices using microprocessor. The main controller attached its back communicates with sub-controllers in real-time by CAN (Controller Area Network) protocol. We used Windows XP as its OS (Operation System) for fast development of main control program and easy extension of peripheral devices. And RTX HAL extension commercial software is used to realize the real-time control in Windows XP environment.

• 한국컴퓨터그래픽스학회논문지
• /
• 제23권3호
• /
• pp.1-8
• /
• 2017

본 논문에서는 인체 모션 데이터를 공간적 제약이 많은 환경 속에 있는 휴머노이드의 몸체에 맞추어 리타겟팅하는 기술을 소개한다. 주어진 모션 데이터는 물체를 손으로 잡거나 장애물 사이를 피해 다니는 등의 세밀한 인터랙션을 포함하고 있다고 가정한다. 또한 휴머노이드의 관절 구조는 인체 관절 구조와 다르며 주변 환경의 모양도 원본 모션이 촬영 될 당시와 서로 다르다고 가정한다. 이러한 조건 하에서 단순히 몸체의 변화만 고려한 리타겟팅 기법을 적용한다면 원본 모션 데이터에서 나타난 인터랙션의 내용을 그대로 보존하기 어렵다. 본 논문에서는 모션 데이터를 휴머노이드 몸체에 맞게 리타겟팅하는 문제와 인터랙션의 내용을 보존하는 문제를 나누어서 독립적으로 해결하는 방법을 제안한다. 먼저 환경 모델과의 인터랙션은 무시하고 모션 데이터를 휴머노이드 몸체에 맞게 리타겟팅 한다. 다음, 환경 모델의 모양을 휴머노이드 모션에 부합하도록 변형하여 원본 데이터에서 나타난 인터랙션이 재현되도록 한다. 마지막으로 휴머노이드 몸체와 환경 모델 사이의 공간적 상관 관계에 대한 제약 조건을 설정하고 환경 모델은 다시 원래 모양으로 되돌린다. 보스턴 다이나믹 사의 아틀라스 로봇 모델을 사용한 실험을 통해 제시된 방법의 유용성을 검증하였다. 향후 모션 데이터 트레킹을 통한 휴머노이드 동작 제어 문제에 사용될 수 있을 것으로 기대된다.

• 제어로봇시스템학회논문지
• /
• 제15권3호
• /
• pp.300-305
• /
• 2009

This paper describes the intelligent robot's hand with three-axis finger force sensors for an intelligent robot. In order to grasp an unknown object safely, it should measure the mass of the object, and determine the grasping force using the mass, then control the robot's fingers with the grasping force. In this paper, the intelligent robot's hand for an intelligent robot was developed. First, the three-axis finger force sensors were designed and manufactured, second, the intelligent robot's hand with three-axis finger force sensors were designed and fabricated, third, the high-speed control system was designed and manufactured using DSP( digital signal processor), finally, the characteristic test to grasp an unknown object safely was carried out. It was confirmed that the developed intelligent robot's hand could grasp an unknown object safely.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 추계학술대회논문집
• /
• pp.1203-1210
• /
• 2007

Robots have begun to perform various tasks on replacing the human in the daily life such as cleaning, entertainments etc. In order to accomplish the effective performance of intricate and precise tasks, robot hand must have special capabilities, such as decision making in given condition, autonomy in unknown situation and stable manipulation of object. In this study, we addresses the development of a 3-fingered humanoid robot hand system. We execute static analysis, vibration analysis and flexible dynamics to reserve stability at the design. Grasp motion of the finger uses a linear actuator and gears. Motion can be distinguished into four parts depending on the grasping thin paper, sphere, and column. In each motion, we compare the displacement of the case to be rigid with the case to be flexible. As a result, manufactured and feasibility of the robot hand is validated through preliminary experiments.

• 제어로봇시스템학회논문지
• /
• 제17권4호
• /
• pp.368-374
• /
• 2011

This paper proposes a robot control human interface using Markov model (HMM) based hand signal recognizer. The command receiving humanoid robot sends webcam images to a client computer. The client computer then extracts the intended commanding hum n's hand motion descriptors. Upon the feature acquisition, the hand signal recognizer carries out the recognition procedure. The recognition result is then sent back to the robot for responsive actions. The system performance is evaluated by measuring the recognition of '48 hand signal set' which is created randomly using fundamental hand motion set. For isolated motion recognition, '48 hand signal set' shows 97.07% recognition rate while the 'baseline hand signal set' shows 92.4%. This result validates the proposed hand signal recognizer is indeed highly discernable. For the '48 hand signal set' connected motions, it shows 97.37% recognition rate. The relevant experiments demonstrate that the proposed system is promising for real world human-robot interface application.

• 대한기계학회논문집A
• /
• 제34권4호
• /
• pp.399-405
• /
• 2010

강력한 파지와 정밀 조작이 가능한 인간형 로봇 손에 대한 연구는 인간의 손동작을 파악하여 메커니즘을 분석하고, 로봇 손에 적절하게 응용하여야 한다. 본 논문에서는 최소의 자유도를 적용한 실용적인 인간형 로봇 손의 파지 동작을 모사하고자, 고속카메라를 사용하여 인간의 손동작에 대한 움직임을 추출하고, 이를 분석하였다. 이러한 촬영 데이터를 필터링하여 실험 데이터를 얻었으며, 이를 활용한 수학적 모델을 구하였다. 또한, 말절골(DIP)은 중절골(PIP) 및 기절골(MCP)의 움직임과 비선형 관계 모델을 고찰하여, 신경회로망 및 커널 회귀 모델을 사용한 인간형 로봇 손의 파지 모델을 얻었다. 신경회로망과 커널 회귀를 이용한 과정을 비교함으로써, 커널 회귀 알고리즘을 통한 모델링이 실험 데이터와의 근접성이 우수한 유효 모델임을 입증하였다.

• International Journal of Control, Automation, and Systems
• /
• 제5권4호
• /
• pp.419-428
• /
• 2007

This paper describes the development of a six-axis force/moment sensor with rectangular taper beams for an intelligent robot's wrist and ankle. In order to accurately push and pull an object with an intelligent robot's hand, and in order to safely walk with an intelligent robot's foot, the robot's wrist and ankle should measure three forces Fx, Fy, and Fz, and three moments Mx, My, and Mz simultaneously from the mounted six-axis force/moment sensor to the intelligent robot's wrist and ankle. Unfortunately, the developed six-axis force/moment sensor utilized in other industrial fields is not proper for an intelligent robot's wrist and ankle in the size and the rated output of the six-axis force/moment sensor. In this paper, the structure of a six-axis force/moment sensor with rectangular taper beams was newly modeled for an intelligent robot's wrist and ankle, and the sensing elements were designed by using the derived equations, following which the six-axis force/moment sensor was fabricated by attaching strain-gages on the sensing elements. Moreover, the characteristic test of the developed sensor was carried out by using the six-component force/moment sensor testing machine. The rated outputs from the derived equations agree well with those from the experiments. The interference error of the sensor is less than 2.87%.

• 한국산업융합학회 논문집
• /
• 제20권4호
• /
• pp.285-297
• /
• 2017

This study proposed a new approach to develop an Intelligent control algorithm for cooperative working of human and robot based on voice recognition. In general case of speaker verification, Gaussian Mixture Model is used to model the feature vectors of reference speech signals. On the other hand, Dynamic Time Warping based template matching techniques were presented for the voice recognition about several years ago. We converge these two different concepts in a single method and then implement in a real time voice recognition enough to make reference model to satisfy 95% of recognition performance. In this paper it was illustrated the reliability of voice recognition by simulation and experiments for humanoid robot with 18 joints.