• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 1997.10a
• /
• pp.225-229
• /
• 1997

This paper describes the camera calibration based-neural network with a camera modeling that accounts for major sources of camera distortion, namely, radial, decentering, and thin prism distortion. Radial distortion causes and inward or outward displacement of a given image point from its ideal location. Actual optical systems are subject to various degrees of decentering, that is, the optical centers of lens elements are not strictly collinear. Thin prism distortion arises from imperfection in lens design and manufacturing as well as camera assembly. It is our purpose to develop the vision system for the pattern recognition and the automatic test of parts and to apply the line of manufacturing. The performance of proposed camera calibration is illustrated by simulation and experiment.

• Proceedings of the Korean Society of Machine Tool Engineers Conference
• /
• 1997.10a
• /
• pp.65-69
• /
• 1997

This paper presents a new approach to the calibration of a SCARA robot orientation with a camera modeling that accounts for major sources of camera distortion, namely, radial, decentering, and thin prism distortion. Radial distortion causes an inward or outward displacement of a given image point from its ideal location. Actual optical systems are subject to various degrees of decentering, that is, the optical centers of lens elements are not strictly collinear. Thin prism distortion arises from imperfection in lens design and manufacturing as well as camera assembly. It is our purpose to develop the vision system for the pattern recognition and the automatic test of parts and to apply the line of manufacturing.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1993.10a
• /
• pp.545-549
• /
• 1993

This paper presents the method of calibrating and compensating for the kinematic errors in robot manipulators. A calibration model is developed to represent any geometric errors in the manipulator's structure. A calibration jig is used to find the values of these kinematic errors in the end-effector's position and a calibration algormined for a SSR-6 robot manipulator developed by Samsung Heavy Industry, Daeduk R & D Center. Through this experiment the maximun kinematic error is reduced from 10mm to 0.4mm

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10b
• /
• pp.1173-1177
• /
• 1990

Cerebellar Model Linear Associator Net(CMLAN), a kind of neuro-net based adaptive control function generator, was applied to the problem of direct inverse calibration of three and six d.o.f. POMA 560 robot. Since CMLAN autonomously maps and generalizes a desired system function via learning on the sampled input/output pair nodes, CMLAN allows no knowledge in system modeling and other error sources. The CMLAN based direct inverse calibration avoids the complex procedure of identifying various system parameters such as geometric(kinematic) or nongeometric(dynamic) ones and generates the corresponding desired compensated joint commands directly to each joint for given target commands in the world coordinate. The generated net outputs automatically handles the effect of unknown system parameters and dynamic error sources. On-line sequential learning on the prespecified sampled nodes requires only the measurement of the corresponding tool tip locations for three d.o.f. manipulator but location and orientation for six d.o.f. manipulator. The proposed calibration procedure can be applied to any robot.

• Journal of the Korean Society for Precision Engineering
• /
• v.12 no.8
• /
• pp.73-85
• /
• 1995

The sensor system to measure the distance precisely from the center of the sensor system to the obstacle is needed to recognize the surrounding environments, and the sensor system is to be calibrated thoroughly to get the range information exactly. This study covers the calibration of the active range sensor which consists of camera and laser slit emitting device, and provides the equations to get the 3D range data. This can be possible by obtaining the extrinsic parameters of laser slit emitting device through image processing the slits measured during the constant distance intervals and the intrinsic parameters from the calibration of camera. The 3D range data equation derived from the simple geometric assumptions is proved to be applicable to the general cases using the calibration parameters. Also the exact 3D range data were obtained to the object from the real experiment.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1995.04a
• /
• pp.637-642
• /
• 1995

For measuring the error motion of ultra-precision spindle, eliminating the geometric errors is a must. Unless it is achieved, geometric errors will be dominant in data. Here, the roundness error and alignment error between spindle and sensor are to be removed. That's because typical error range of such spindle is muchless than geometric one. A capacitive transducer of cylidricalshape was developed, which takes full advantage of the spatial-averaging effect by using large area compared tpo the geometric error. This idea was first proposed by Chapman and here it is modified for better performance with nomical gap of 50 .mu. m and with newly designed guards which encompass the respective sensor to rectify the electrical field distribution in good shape. The measurement system is made to get the orbit of Ultra-Precision Air Spindle which is supposed to have its runout under 1 .mu. m. The Calibration data of this sensor is presented and the spindle orbit from 2000rpm to 5500rpm is showed. It is quite reasonable to use this sensor in the range of 60 .mu. m with an accuracy of several tens of nm.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.5
• /
• pp.127-135
• /
• 2015

In this paper, we propose the semi-auto camera calibration method including user input. The proposed method estimates the vanishing points using user defined reference lines and defines the constraint for reducing outlier in vanishing points estimation process. The proposed camera calibration method based on both algebraic and geometric method improves a calibration performance for difficult condition, which represents that existing method can't calibrate a image. Experimental results show that the proposed method calibration accuracy higher than existing method.

• International Journal of Highway Engineering
• /
• v.16 no.4
• /
• pp.119-126
• /
• 2014

PURPOSES: Used in transportation planning and traffic engineering, almost traffic simulation tools have input variable values optimized by overseas traffic flow attribution because they are almost developed in overseas country. Thus, model calibration appropriated for internal traffic flow attribution is needed to improve reliability of simulation method. METHODS : In this study, the traffic flow model calibration is based on expressways. For model calibration, it needs to define each expressway link according to attribution, thus it is classified by design speed, geometric conditions and number of lanes. And modified greenshield model is used as traffic flow model. RESULTS : The result of the traffic model calibration indicates that internal congested density is lower than overseas. And the result of analysis according to the link attribution indicates that the more design speed and number of lanes increase, the lower the minimum speed, the higher the congested density. CONCLUSIONS: In the traffic simulation tool developed in overseas, the traffic flow is different as design speed and number of lanes, but road segment don't affect traffic flow. Therefore, these results need to apply reasonably to internal traffic simulation method.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.10a
• /
• pp.355-360
• /
• 2002

To calibrate a non-redundantly actuated parallel mechanism, one can find actual kinematic parameters by means of geometrical constraint of the mechanism's kinematic structure and measurement values. However, the calibration algorithm for a non-redundant case does not apply fur a redundantly actuated parallel mechanism, because the angle error of the actuating joint varies with position and the geometrical constraint fails to be consistent. Such change of joint angle error comes from constraint torque variation with each kinematic pose (meaning position and orientation). To calibrate a redundant parallel mechanism, one therefore has to consider constraint torque equilibrium and the relationship of constraint torque to torsional deflection, in addition to geometric constraint. In this paper, we develop the calibration algorithm fir a redundantly actuated parallel mechanism using these three relationships, and formulate cost functions for an optimization algorithm. As a case study, we executed the calibration of a 2-DOF parallel mechanism using the developed algorithm. Coordinate values of tool plate were measured using a laser ball bar and the actual kinematic parameters were identified with a new cost function of the optimization algorithm. Experimental results showed that the accuracy of the tool plate improved by 82% after kinematic calibration in a redundant actuation case.

• Journal of Institute of Control, Robotics and Systems
• /
• v.8 no.9
• /
• pp.803-811
• /
• 2002

Machine vision systems are usually used to identify traffic lanes and then determine the steering angle of an autonomous vehicle in real time. The steering angle is calculated using a geometric model of various parameters including the orientation, position, and hardware specification of a camera in the machine vision system. To find the accurate values of the parameters, camera calibration is required. This paper presents a new camera-calibration algorithm using known traffic lane features, line thickness and lane width. The camera parameters considered are divided into two groups: Group I (the camera orientation, the uncertainty image scale factor, and the focal length) and Group II(the camera position). First, six control points are extracted from an image of two traffic lines and then eight nonlinear equations are generated based on the points. The least square method is used to find the estimates for the Group I parameters. Finally, values of the Group II parameters are determined using point correspondences between the image and its corresponding real world. Experimental results prove the feasibility of the proposed algorithm.