• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1992년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 19-21 Oct. 1992
• /
• pp.874-879
• /
• 1992

It is shown that there exists a nonlinear mappping which transforms features and their changes to the desired camera motion without measurement of the relative distance between the camera and the part, and the nonlinear mapping can eliminate several difficulties encountered when using the inverse of the feature Jacobian as in the usual feature-based visual feedback controls. And instead of analytically deriving the closed form of such a nonlinear mapping, a fuzzy membership function (FMF) based neural network is then proposed to approximate the nonlinear mapping, where the structure of proposed networks is similar to that of radial basis function neural network which is known to be very useful in function approximations. The proposed FMF network is trained to be capable of tracking moving parts in the whole work space along the line of sight. For the effective implementation of proposed IMF networks, an image feature selection processing is investigated, and required fuzzy membership functions are designed. Finally, several numerical examples are illustrated to show the validities of our proposed visual servoing method.

• Journal of Electrical Engineering and Technology
• /
• 제8권2호
• /
• pp.385-392
• /
• 2013

An uncalibrated visual servo control method for tracking a target is presented. We define the robot-positioning problem as an unconstrained optimization problem to minimize the image error between the target feature and the robot end-effector feature. We propose a method to find the residual term for more precise modeling using the secant approximation method. The composite image Jacobian is estimated by the proper method for eye-to-hand configuration without knowledge of the kinematic structure, imaging geometry and intrinsic parameter of camera. This method is independent of the motion of a target feature. The algorithm for regulation of the joint velocity for safety and stability is presented using the cost function. Adaptive regulation for visibility constraints is proposed using the adaptive parameter.

• 한국지능시스템학회논문지
• /
• 제21권5호
• /
• pp.595-601
• /
• 2011

본 논문에서는 영상 정보를 이용하여 로보커 팔위의 역진자의 밸런싱 제어를 한다. 로봇 팔위에 놓인 역진자의 각도는 카메라로 검출하고 검출된 각도 값은 제어기로 귀환되어 오차를 생성한다. 따라서 전체 제어루프는 폐회로 루프를 형성한다. 제어 성능을 높이기 위해 기존 선형제어기에 신경망 제어기를 더하였다. RBF 네트워크의 학습 알고리즘은 FPGA에 설계된 부동소수점 연산이 가능한 디지털 제어기에 의해 수행된다. 실험을 통하여 전체 시스템 성능을 검증하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.100.2-100
• /
• 2002

In this paper we propose a new visual servoing method for door opening with mobile manipulator. We use an eye-to-hand system that stereo camera is mounted on mobile platform, and adopt the position-based method. The previous methods for door opening mostly used eye-in-hand system with mono camera and required predefined knowledge such as radius and position about door grip, which was mainly caused by using mono cam era. This is also a severe constraint for pursuing general-purpose algorithm for door opening. For overcoming such drawback, we use stereo camera and suggest a new method that detect the door grip and estimate its pose from stereo depth information without predefined knowledge. Al...

• 제어로봇시스템학회논문지
• /
• 제15권7호
• /
• pp.694-699
• /
• 2009

This paper presents the implementation of the real-time object tracking control of the ROBOKER head. The visual servoing technique is used to track the moving object, but suffers from ill-estimated Jacobian of the virtual link design. To improve the tracking performance, the RBF(Radial Basis Function) network is used to compensate for uncertainties in the kinematics of the robot head in on-line fashion. The reference compensation technique is employed as a neural network control scheme. Performances of three schemes, the kinematic based scheme, the Jacobian based scheme, and the neural network compensation scheme are verified by experimental studies. The neural compensation scheme performs best.

• 제어로봇시스템학회논문지
• /
• 제4권6호
• /
• pp.753-758
• /
• 1998

본 논문은 카메라 파라메터의 측정오차에 대하여 강인성을 보이는 새로운 로봇의 스테레오 시각 위치제어 알고리즘을 제시한다. 제시된 알고리즘은 카메라로부터 측정된 영상 데이터만을 이용함으로써, 특히 파라메터 측정오차에 대하여 매우 민감함을 보이는 영상 데이터로부터 작업 공간에서의 위치로의 변환, 즉 역변환 추정장치의 필요성을 제거하였다. 이러한 특징이 기존 개발된 시각 제어기와의 큰 차이를 두고 있다. 그럼에도 불구하고 제시된 제어기는 전 작업 영역 내에서 시스템 안정성을 갖는다. 또한 카메라의 위치 측정 오차에 대하여 전혀 영향을 받지 않음이 증명되어지고 방향 폭정 오류에 대해서도 기존 제어기보다 강인함을 시뮬레이션을 통하여 보여진다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2002년도 합동 추계학술대회 논문집 정보 및 제어부문
• /
• pp.134-137
• /
• 2002

This paper presents a robot manipulator kinematic motion control scheme based on velocity feedback loop. The desired joint velocity is obtained by the feature-based visual servoing and is used in the joint velocity control loop system for trajectory control of the robot manipulator. The asymptotic stability of the closed loop system is shown by the Lyapunov method. Effectiveness of the proposed method is shown by simulation and experimental results on a robot manipulator with two degree of freedom.

• Journal of Electrical Engineering and information Science
• /
• 제2권6호
• /
• pp.177-182
• /
• 1997

A visual servoing method is proposed for a robot with a camera in hand. Specifically, vanishing point features are suggested by employing a viewing model of perspective projection to calculate the relative rolling, pitching and yawing angles between the object and the camera. To compensate dynamic characteristics of the robot, desired feature trajectories for the learning of visually guided line-of-sight robot motion are obtained by measuring features by the camera in hand not in the entire workspace, but on a single linear path along which the robot moves under the control of a commercially provided function of linear motion. And then, control actions of the camera are approximately found by fuzzy-neural networks to follow such desired feature trajectories. To show the validity of proposed algorithm, some experimental results are illustrated, where a four axis SCARA robot with a B/W CCD camera is used.

• 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.

• 한국정밀공학회지
• /
• 제17권9호
• /
• pp.13-20
• /
• 2000