The process of microassembly often involves the manipulation and grasping of complex micro-sized objects. These micro-objects are generally initially placed at random orientations in the plane when transferred to an assembly workstation. The part orientations must be corrected in order for the gripper to grasp them. Visual monitoring feedback is commonly used to assist in the re-orientation and manipulation process. However, due to the very limited field of view of imaging systems used in micro-assembly, the target micro-object may not remain within the viewable area during the reorientation processes. This paper presents an integrated controller algorithm to automatically rotate micro-objects to the desired orientation in the plane for grasping. This novel algorithm incorporates the technique of dynamic tracking to prevent the target micro-object from shifting outside of the viewable area during the re-orientation process. As the micro-object is rotated to correct its orientation, the algorithm will evaluate the required translational inputs in order to maintain the micro-object within the field of view. To minimize the delay due to feedback control, the proposed algorithm employs the principle of Kaiman filter estimation to track and project the motion at the next sampling interval. The proposed algorithm was evaluated experimentally with a microassembly operation using an in-house micro-manipulator. Results from the experiments showed that the proposed tracking algorithm was capable of tracking and maintaining the targeted micropart at the specified location with an error of within +/−15 pixels.

This content is only available via PDF.
You do not currently have access to this content.