• 제어로봇시스템학회논문지
• /
• 제21권8호
• /
• pp.753-757
• /
• 2015

This paper introduces an experimental rescue robot, HUBO T-100 and presents the optimal motion control method. The objective of the rescue robot is to extract patients or wounded soldiers in the battlefield and hazardous environments. Another mission is to dispose and transport an explosive ordnance to safe places. To execute these missions, the upper body of the rescue robot is humanoid in form to execute various kinds of tasks. The lower body features a hybrid tracked/legged design, which allows for a variety of mode of locomotion, depending on terrain conditions in order to increase traversability. The weight lifting motion is one of the most important task for performing rescue related missions because the robot must lift an object or impaired person lying on the ground for transferring. Here, dynamics based motion optimization is employed to minimize joint torques while maintaining stability simultaneously. Physical experiments with a real humanoid robot, HUBO T-100, are presented to verify the proposed method.

• Nuclear Engineering and Technology
• /
• 제56권7호
• /
• pp.2718-2731
• /
• 2024

To ensure the operational safety of nuclear power plants, we present a Quadruped Inspection Robot that can be used for many types of steam generators. Since the Inspection Robot relies on the Holding Modules to grip the tube-sheet, it can be regarded as a hybrid robot with variable configurations, switching between 4-RRR-RR, 3-RRR-RR, and two types of 2-RRR-RR, and the variable configurations bring a great challenge to dimensional synthesis. In this paper, the kinematic model of the Inspection Robot in multiple configurations is established, and the analytical solution is given. The workspace mapping is analyzed by the solution-space, and the workspace of multiple configurations is decomposed into the workspace of 2-RRR to reduce the analysis complexity, and the workspace calculation is simplified by using the envelope rings. The optimization problem of the manipulator is transformed into the calculation of the shortest contraction length of the swing leg. The switching performance of the Inspection Robot is evaluated by stride-length, turning-angle, and workspace overlap-ratio. The performance indexes are classified and transformed based on the proportions and variation trends of dimensional parameters to reduce the number of optimization objective functions, and Pareto optimal solutions are obtained using an intelligent optimization algorithm.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2007년도 추계학술대회 논문집
• /
• pp.285-286
• /
• 2007

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2013년도 춘계학술대회 논문집
• /
• pp.629-630
• /
• 2013

• 한국항행학회논문지
• /
• 제15권6호
• /
• pp.1143-1151
• /
• 2011

본 논문은 각 기관별로 운영되고 있는 게시판 정보를 하나로 통합하여 클라우드 기반의 스마트폰 환경에서 사용자가 원하는 정보를 제공해주는 하이브리드 웹 기반 통합 시스템을 구현하였다. 본 시스템은 대학의 각 기관에서 운영되고 있는 웹 서버의 게시물 정보를 적절한 시간 간격으로 수집하여 저장 및 관리하는 SaaS 서비스의 웹로봇과 스마트 폰의 UI를 통한 사용자에게 정보를 제공하는 게시물관리 PaaS 서버로 구성된다. 또한, 스마트폰 사용자의 GPS 정보를 활용하여 새로운 게시물의 정보를 알려주는 알리미 어플을 구현하였다.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2005년도 추계학술대회 논문집
• /
• pp.984-988
• /
• 2005

To accomplish various and complex tasks by intelligent robots, improvement is needed not only in mechanical system architecture but also in control system architecture. Hybrid control architecture has been suggested as a mutually complementing architecture of the weak points of a deliberative and a reactive control. This paper addresses a control architecture of robots, and a behavior representation methodology. The suggested control architecture consists of three layers of deliberative, sequencing, and reactive as hybrid control architecture. Multi-layer behavior model is employed to represent desired tasks. 3D simulation will be conducted to verify the applicability of suggested control architecture and behavior representation method.

• International Journal of Control, Automation, and Systems
• /
• 제5권6호
• /
• pp.674-680
• /
• 2007

The probabilistic roadmap (PRM) method, which is a popular path planning scheme, for a manipulator, can find a collision-free path by connecting the start and goal poses through a roadmap constructed by drawing random nodes in the free configuration space. PRM exhibits robust performance for static environments, but its performance is poor for dynamic environments. On the other hand, reinforcement learning, a behavior-based control technique, can deal with uncertainties in the environment. The reinforcement learning agent can establish a policy that maximizes the sum of rewards by selecting the optimal actions in any state through iterative interactions with the environment. In this paper, we propose efficient real-time path planning by combining PRM and reinforcement learning to deal with uncertain dynamic environments and similar environments. A series of experiments demonstrate that the proposed hybrid path planner can generate a collision-free path even for dynamic environments in which objects block the pre-planned global path. It is also shown that the hybrid path planner can adapt to the similar, previously learned environments without significant additional learning.

• 한국지능시스템학회논문지
• /
• 제17권7호
• /
• pp.913-918
• /
• 2007

본 논문에서는 인간과 로봇의 인터액션을 위해 연속된 이미지 정보로부터 얻을 수 있는 색상(color)과 기울기(gradient) 정보를 이용하여 인간 몸의 실루엣 추출 기법을 제안한다. 연속된 이미지 정보로부터 얻어진 RGB 영상 정보에서 색상 배경 제거 기법은 각각의 신체 비율 정보로부터 추출된 모션 영역 정보에서 색상 공판별 평균 이미지 정보를 얻고 옷 색상 정보를 볼록 합하여 계산된다. 기울기 배경 제거 기법은 공간상 정보와 시간상 정보의 볼록 합으로 계산된다. 최종적으로 색상 정보와 기울기 정보의 볼록 합을 하여 인간 몸의 실루엣을 추출한다. 마지막으로, 실험을 통하여 제안된 기법의 성능을 확인하였다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2002년도 추계학술발표논문집 (상)
• /
• pp.27-30
• /
• 2002

Recently, there are lots of concerning on the control of multi-agent control system and mixed reality integrating real and virtual environments. In this paper, an integrated agent control system based on mixed reality and mobile environment is proposed. The system consists of a manager, agents, a mapping module, a geographic information module, a WAP server, and a mobile manager. The manager has Hybrid Device Manager(HDM) working for revising images and position data of the agents to display that at the 3D virtual and mobile environments respectively. Especially, the images of the agent is translated to LBM format and displayed on the mobile phone, and the current position of the agent is displayed on the display screen, so that a user are able to control the agent when he is moving. Experimental results shows the proposed system lets us control the agents conveniently at a computer and a mobile phone.

• 한국HCI학회:학술대회논문집
• /
• 한국HCI학회 2008년도 학술대회 1부
• /
• pp.190-194
• /
• 2008

본 논문은 지능형 서비스 로봇의 여러 기술들 중에서 기본적인 기술인 화자식별 기술에 관한 내용이다. 화자식별 기술은 화자의 음성신호를 이용하여 등록된 화자들 중에서 가장 유사한 화자를 찾아내는 것이다. 기존의 mel-frequency cepstral coefficient 를 이용한 화자식별 시스템은 무잡음 환경에서는 높은 성능을 보장하지만 잡음환경에서는 성능이 급격하게 떨어진다. 이렇게 잡음환경에서 성능이 떨어지는 요인은 등록환경과 식별환경이 다른 불일치문제 때문이다. 본 논문에서는 불일치문제를 해결하기 위해 relative autocorrelation sequence mel-frequency cepstral coefficient 를 사용하였다. 또한, 기존의 relative autocorrelation sequence mel-frequency cepstral coefficient 의 제한된 정보문제와 잔여잡음문제를 해결하기 위해 멀티스트리밍 방법과 멀티스트리밍 방법에 특정벡터 재결합 방법을 결합한 하이브리드 방법을 제한 하였다. 실험결과 제한된 방법들이 기존의 특정벡터보다 잡음환경에서 높은 화자식별 성능을 보여주었다.