Entity

Time filter

Source Type


Li J.,University of Electronic Science and Technology of China | Chen M.,University of Electronic Science and Technology of China | Jin X.,University of Electronic Science and Technology of China | Chen Y.,University of Electronic Science and Technology of China | And 3 more authors.
Optik | Year: 2011

A multiple axes 3-D laser scanning system consisting of a portable 3-D laser scanner, a industrial robot and a turntable is demonstrated. By using a criterion sphere, a robot tool center point (TCP) calibration approach is proposed to calibrate the relation between the laser 3-D scanner and the robot end-effector. In this approach, two different translational motions of robot are first made to determine the rotation part, and then at least three different rotational motions are made to determine the translation part. Meanwhile, by using the criterion sphere, a turntable approach is proposed to calibrate the pose of the turntable relative to the robot. In this approach, several rotational angles of turntable and two different heights of the sphere are made to determine the rotational axis of turntable. Experiment is performed on a portable laser scanner mounted on an industrial robot ABB IRB4400 with a turntable. The experiment results show that the two proposed calibration algorithms are stable and flexible. The application of 3-D measurement is also given to demonstrate the effectiveness and stability of the multiple axes 3-D laser scanning system. © 2010 Elsevier GmbH. All rights reserved. Source


Yang Y.,Tsinghua University | Song Y.,Tsinghua University | Liang W.,Tsinghua University | Wang J.,Tsinghua University | Qi L.,InterSmart Robotic Systems Co.
Jiqiren/Robot | Year: 2010

To improve the removal control for robot grinding process, we propose a modeling method based on SVM (support vector machine) regression. By analyzing a group of measurable variables relevant to grinding removal, such as robot's speed, contact force and curvature of the workpiece's surface, a regression model is built using machine learning method to predict the grinding removal. In this way, the analysis on a series of complicated dynamic variables could be avoided. The experimental results show that this method could achieve good performance. The prediction accuracy of the model reaches higher than 90%, which basically meets the demand of practical grinding. Source


Wu S.,University of Connecticut | Kazerounian K.,University of Connecticut | Gan Z.,InterSmart Robotic Systems Co. | Sun Y.,InterSmart Robotic Systems Co.
Machining Science and Technology | Year: 2014

Robotic belt grinding is an effective process for manufacturing workpieces with complicated free-form geometries. However, due to the relatively low stiffness in the system, more sophisticated modeling and control strategies are called for. This article presents a novel model for estimation of the material removal in the robotic belt grinding process. In particular, two process parameters, robot velocity and contact force between the workpiece and the contact wheel, are analyzed in the presented process model. A superposition method is introduced to estimate the pressure distribution in the contact area. The presented method greatly reduces the computation time compared to finite element analysis (FEA) methods and provides explicit equations for real-time system analysis. Additionally, a shape-dependent model is proposed to estimate the material removal. The model introduces local coefficients to denote the material removal ability of the system at certain locations. This developed methodology can essentially adapt to workpieces with complicated geometries. Experimental results verified the effectiveness and accuracy of the model. © 2014 Taylor & Francis Group, LLC. Source


Lv H.,Tsinghua National Laboratory for Information Sciences and Technology | Song Y.,Tsinghua National Laboratory for Information Sciences and Technology | Jia P.,Tsinghua National Laboratory for Information Sciences and Technology | Gan Z.,InterSmart Robotic Systems Co. | Qi L.,InterSmart Robotic Systems Co.
2010 IEEE International Conference on Information and Automation, ICIA 2010 | Year: 2010

Robotic belt grinding system has good prospect to release hand-grinder from their dirty and noisy working environment. However, as a kind of non-rigid processing system, it is a challenge to model its processes precisely for free-form surface because its performance is unstable due to a variety of factors, such as belt wear and belt replacement. In order to adapt to the variability, an adaptive modeling approach based on echo state network (ESN) is presented, whose major idea is to exhaust information from new data by using sliding window technique to select training samples. With machine learning paradigm this approach is more flexible than traditional ones which often base on formula and experimental curves. Experimental results of grinding turbine blades demonstrate this approach is workable and effective. ©2010 IEEE. Source


Yang Y.,Tsinghua University | Song Y.,Tsinghua University | Wang J.,Tsinghua University | Gan Z.,InterSmart Robotic Systems Co. | Qi L.,InterSmart Robotic Systems Co.
Proceedings - 2010 3rd IEEE International Conference on Computer Science and Information Technology, ICCSIT 2010 | Year: 2010

The performance of a model, which is trained with offline data, is highly relied on the conditions in which the system is working. When the working conditions change, the prediction accuracy of the model will be reduced significantly. To solve this problem, we propose an adaptive SVR modeling method based on vector-field-smoothed (VFS) algorithm. This method can adapt the model quickly to new working conditions by using only a few adaptive samples. Also, it can extend the feature subspace which the model covers so as to enhance the generalization ability of the model. The experimental results show that the model using this method can achieve a much better performance than the original model, as well as the model using other adaptive SVR modeling method. © 2010 IEEE. Source

Discover hidden collaborations