Entity

Time filter

Source Type

Sai Kung, Hong Kong

Hu Z.,Chinese University of Hong Kong | Chung R.,Chinese University of Hong Kong | Fung K.S.M.,ASM Assembly Automation Ltd.
Machine Vision and Applications | Year: 2010

Extended Gaussian image (EGI) and complex EGI (CEGI) have been widely used as the representation of 3D shapes for shape recognition and pose estimation. In this work, we extend the representations and present a new representation named enriched complex extended Gaussian image (EC-EGI). The representation follows the same framework of EGI and CEGI, which is to represent each surface patch of the target 3D shape as a weight at the associated spot on the surface of the Gaussian sphere. However, while the original CEGI uses a single complex number as the weight, the new representation uses three complex numbers, which are related to the centroid position of the surface patch in 3D. With the inclusion of more information in the new representation, not only could object pose be determined more accurately, but also some key ambiguities of shape representation that CEGI and EGI have also removed. The translation parameters in the pose estimation application could also be determined in a simpler and more accurate way. In addition, the Gaussian sphere partition problem of CEGI is no longer present. Experimental results on synthetic and real image data are shown to illustrate the performance of the proposed representation in pose estimation. © 2009 Springer-Verlag. Source


Sun Z.G.,Hong Kong Polytechnic University | Cheung N.C.,Hong Kong Polytechnic University | Zhao S.W.,Hong Kong Polytechnic University | Gan W.C.,ASM Assembly Automation Ltd.
Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science | Year: 2010

A control algorithm for the position tracking of a magnetic levitation system is presented in this article. The magnetic levitation system is well known for its non-linear dynamic characteristics and open-loop instability. The external disturbances will deteriorate the dynamic performance of the magnetic levitation system, and may give rise to system instability. This problem triggers enormous interests in designing various controllers for the non-linear dynamic system. In this article, a magnetic levitation system is first modelled. Then, a sliding mode controller is proposed, with a simple yet effective disturbance observer to perform disturbance rejection. Both the simulation results and the experimental results verify the validity of the robust controller. Source


Dong M.,Chinese University of Hong Kong | Chung R.,Chinese University of Hong Kong | Lam E.Y.,University of Hong Kong | Fung K.S.M.,ASM Assembly Automation Ltd.
IEEE Transactions on Electronics Packaging Manufacturing | Year: 2010

Die bonding in the semiconductor industry requires placement of solder bumps not on PCBs but on wafers. Such wafer bumps, which are much miniaturized from their counterparts on printed circuit boards (PCBs), require their heights meet rigid specifications. Yet the small size, the lack of texture, and the mirror-like nature of the bump surface make the inspection task a challenge. Existing inspection schemes generally reconstruct every bump surface. This work addresses by how much can the task be simplified if merely the bump heights are inspected against the specification. It is assumed that ball bumps are used as the wafer bumps. An imaging setup is described that lets the peaks of the ball bumps be distinguishable in the image data. A measure is also described that reveals how well the ball bumps meet the height specification without going through explicit 3-D reconstruction. The measurement, in the form of a 3 × 3 matrix extractable from the image data, is sensitive to variations in the bump heights, but not to 2-D uncertainties in soldering the bumps onto the wafer substrate, or small variations in the placement of the wafer in 3-D. Experimental results are shown to illustrate the effectiveness of the proposed system. © 2010 IEEE. Source


Ngai D.C.K.,ASM Assembly Automation Ltd. | Yung N.H.C.,University of Hong Kong
IEEE Transactions on Intelligent Transportation Systems | Year: 2011

In this paper, we present a learning method to solve the vehicle overtaking problem, which demands a multitude of abilities from the agent to tackle multiple criteria. To handle this problem, we propose to adopt a multiple-goal reinforcement learning (MGRL) framework as the basis of our solution. By considering seven different goals, either Q-learning (QL) or double-action QL is employed to determine action decisions based on whether the other vehicles interact with the agent for that particular goal. Furthermore, a fusion function is proposed according to the importance of each goal before arriving to an overall but consistent action decision. This offers a powerful approach for dealing with demanding situations such as overtaking, particularly when a number of other vehicles are within the proximity of the agent and are traveling at different and varying speeds. A large number of overtaking cases have been simulated to demonstrate its effectiveness. From the results, it can be concluded that the proposed method is capable of the following: 1) making correct action decisions for overtaking; 2) avoiding collisions with other vehicles; 3) reaching the target at reasonable time; 4) keeping almost steady speed; and 5) maintaining almost steady heading angle. In addition, it should also be noted that the proposed method performs lane keeping well when not overtaking and lane changing effectively when overtaking is in progress. © 2006 IEEE. Source


Trademark
ASM Assembly Automation Ltd. | Date: 2007-04-03

computer software used for detecting manufacturing defects in map sorters and die bonders.

Discover hidden collaborations