• 로봇학회논문지
• /
• 제12권3호
• /
• pp.313-321
• /
• 2017

One of the most frequently performed tasks in human-robot interaction (HRI), intelligent vehicles, and security systems is face related applications such as face recognition, facial expression recognition, driver state monitoring, and gaze estimation. In these applications, accurate head pose estimation is an important issue. However, conventional methods have been lacking in accuracy, robustness or processing speed in practical use. In this paper, we propose a novel method for estimating head pose with a monocular camera. The proposed algorithm is based on a deep neural network for multi-task learning using a small grayscale image. This network jointly detects multi-view faces and estimates head pose in hard environmental conditions such as illumination change and large pose change. The proposed framework quantitatively and qualitatively outperforms the state-of-the-art method with an average head pose mean error of less than $4.5^{\circ}$ in real-time.

• 대한기계학회논문집A
• /
• 제24권4호
• /
• pp.848-857
• /
• 2000

A pick-and-place operation in 3-dimensional environment is basic operation for human and multi-purpose manipulators. However, there may be a difficult problem for such manipulators. Especially, if the object cannot be moved with a single grasp, regrasping, which can be a time-consuming process, should be carried out. Regrasping, given initial and final pose of the target object, is a construction of sequential transition of object poses that are compatible with two poses in the point of grasp configuration. This paper presents a novel approach for solving regrasp problem. The approach consists of a preprocessing and a planning stage. Preprocessing, which is done only once for a given robot, generates a look-up table which has information of kinematically feasible task space of end-effector through all the workspace. Then, using the table planning automatically determines possible intermediate location, pose and regrasp sequence leading from the pick-up to put-down grasp. Experiments show that the presented is complete in the total workspace. The regrasp planner was combined with existing path.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.2148-2153
• /
• 2005

For a legged robot such as a humanoid, balancing its body during a given motion is natural but the most important problem. Recently, a motion given to a humanoid is more and more complicated, and thus the balancing problem becomes much more critical. This paper suggests a real-time motion generation algorithm that guarantees a humanoid to be balanced during the motion. A desired motion of each arm and/or leg is planned by the conventional motion planning method without considering the balancing problem. In order to balance a humanoid, all the given motions are embedded into the COG Jacobian. The COG Jacobian is modified to include the desired motions and, in consequence, dimension of the COG Jacobian is drastically reduced. With the motion-embedded COG Jacobian, balancing and performing a task is completed simultaneously, without changing any other parameters related to the control or planning. Validity and efficiency of the proposed motion-embedded COG Jacobian is simulated in the paper.

• 제어로봇시스템학회논문지
• /
• 제16권12호
• /
• pp.1233-1240
• /
• 2010

In the flat panel display and semiconductor industries, the visual alignment system is considered as a core technology which determines the productivity of a manufacturing line. It consists of the vision system to extract the centroids of alignment marks and the stage control system to compensate the alignment error. In this paper, we develop a Kalman filter algorithm to estimate the alignment mark postures and propose a coarse-fine alignment control method which utilizes both original fine images and reduced coarse ones in the visual feedback. The error compensation trajectory for the distributed joint servos of the alignment stage is generated in terms of the inverse kinematic solution for the misalignment in task space. In constructing the estimation algorithm, the equation of motion for the alignment marks is given by using the forward kinematics of alignment stage. Secondly, the measurements for the alignment mark centroids are obtained from the reduced images by applying the geometric template matching. As a result, the proposed Kalman filter based coarse-fine alignment control method enables a considerable reduction of alignment time.

• 한국정밀공학회지
• /
• 제19권12호
• /
• pp.70-77
• /
• 2002

Necessity of the tole-operation has been increased in many fields. Since the Internet is inexpensive and available all over the world, it is a strong candidate for the transmission media of the tole-operation. However, the Internet has random time delays that may cause instability in the system especially if the tole -operation is bilateral. In the past few years many attempts have been made to overcome the random time delay, So far, they are still insufficient in terms of performance. The ‘Variable holding time’ is introduced to improve the performance of the ‘Event based tole-operation’ which controls a system with a non-time action reference. By holding each event for proper time, the system can quickly respond and be stabilized. The proper holding time should be selected based on the characteristics of the task that the system performs. The factors that reflect those characteristics are investigated. The fuzzy logic is employed to obtain the proper holding time for each event while the tole-operation system is in operation. The experimental results presented in this paper verify effectiveness of the proposed method.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2017년도 제55차 동계학술대회논문집 25권1호
• /
• pp.147-150
• /
• 2017

지능형 로봇의 사용자 품질을 결정하는 주요 요소들 중 하나는 짧은 부트 시간이다. 로봇 시스템에서는 부팅 과정 중에 침입자인지, 자택 순찰, 개인 비서, 엔터테인먼트와 같은 다수의 응용들이 동시에 초기화되는데, 고품질의 사용자 경험을 제공하기 위해서는 사용자 응답성이 중요한 응용들이 우선적으로 초기화되어야한다. 이를 위해 리눅스 기반 로봇 시스템에서 부트 시간을 단축하기 위한 다양한 연구들이 진행되어 왔다. 하지만 이들은 단일 응용 각각에 대한 초기화 시간을 단축하는 연구들이며, 응용들 간에 CPU, 메모리, I/O와 같은 자원 경쟁에 의한 지연 요소를 고려하지 않고 있다. 본 논문에서는 응용들 간의 각종 자원경쟁들을 고려하여 사용자 응답성이 중요한 응용을 우선적으로 초기화하기 위한 외부 컨텍스트 기반 선별적 자원 사용률 조정기법을 제안한다. 이를 리눅스 기반 시스템 상에 구현하여 검증한 결과 응용의 부트 시간이 기존 대비 33.02% 단축됨을 확인했다.

• 제어로봇시스템학회논문지
• /
• 제19권11호
• /
• pp.1036-1042
• /
• 2013

In this paper, an elderly assistance system is developed based on Xenomai, a real-time development framework cooperating with the Linux kernel. A Kinect sensor is used to recognize the behavior of the elderly and A-star search algorithm is implemented to find the shortest path to the person. The mobile robot also generates a trajectory using a digital convolution operator which is based on a Bezier curve for smooth driving. In order to follow the generated trajectory within the control period, we developed real-time tasks and compared the performance of the tracking trajectory with that of non real-time tasks. The real-time task has a better result on following the trajectory within the physical constraints which means that it is more appropriate to apply to an elderly assistant system.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제13권2호
• /
• pp.140-146
• /
• 2013

In this paper, we describe a vision sensor-based driving algorithm for indoor automatic guided vehicles (AGVs) that facilitates a path tracking task using two mono cameras for navigation. One camera is mounted on vehicle to observe the environment and to detect markers in front of the vehicle. The other camera is attached so the view is perpendicular to the floor, which compensates for the distance between the wheels and markers. The angle and distance from the center of the two wheels to the center of marker are also obtained using these two cameras. We propose five movement patterns for AGVs to guarantee smooth performance during path tracking: starting, moving straight, pre-turning, left/right turning, and stopping. This driving algorithm based on two vision sensors gives greater flexibility to AGVs, including easy layout change, autonomy, and even economy. The algorithm was validated in an experiment using a two-wheeled mobile robot.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제13권2호
• /
• pp.91-99
• /
• 2013

An autonomous unmanned underwater vehicle is a type of marine self-propelled robot that executes some specific mission and returns to base on completion of the task. In order to successfully execute the requested operations, the vehicle must be guided by an effective navigation algorithm that enables it to avoid obstacles and follow the best path. Architectures and principles for intelligent dynamic systems are being developed, not only in the underwater arena but also in related areas where the work does not fully justify the name. The problem of increasing the capacity of systems management is highly relevant based on the development of new methods for dynamic analysis, pattern recognition, artificial intelligence, and adaptation. Among the large variety of navigation methods that presently exist, the dynamic window approach is worth noting. It was originally presented by Fox et al. and has been implemented in indoor office robots. In this paper, the dynamic window approach is applied to the marine world by developing and extending it to manipulate vehicles in 3D marine environments. This algorithm is provided to enable efficient avoidance of obstacles and attainment of targets. Experiments conducted using the algorithm in MATLAB indicate that it is an effective obstacle avoidance approach for marine vehicles.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2006년도 가을 학술발표논문집 Vol.33 No.2 (B)
• /
• pp.319-323
• /
• 2006

본 논문에서는 동작 모델과 작업 목표에 따라 지능 로봇 시스템의 작업 순서를 결정하는 작업 계획기의 설계와 데모시스템에 대해 설명한다. 블랙보드 중심의 분산 제어 구조에서 하나의 독립적인 지식원으로 동작하는 작업 계획기는 작업 관리기의 요청이 있을 때마다 지식베이스로부터 동작 모델과 월드 상태 정보를 제공받아 작업 목표 달성을 위한 작업 계획을 생성한다. 그리고 이렇게 생성된 작업 계획은 오류로 인해 재 계획이 필요할 때까지 작업 관리기를 통해 실행된다. 본 연구의 작업 계획기는 효율적인 작업 계획 생성을 위해 지역 상태공간 탐색법의 하나인 EHC+ 탐색법과 계획그래프-기반의 휴우리스틱 계산법을 적용한다. 본 논문에서는 작업 계획기의 효율성과 블랙보드와의 연계성을 시험하기 위한 데모 시스템에 대해 자세히 설명한다. 이를 통해 지식베이스, 작업 관리기, 컴포넌트 서비스기 등 로봇 제어 구조내의 다른 지식원들과의 인터페이스를 위한 메시지 설계도 소개한다.