An Intelligent Force Sensing and Control System for Microassembly/Micromanipulation
1. Project Significance Micro fabrication methods such as bulk-micro machining and surface-micro machining are commonly used to make devices for a wide variety of applications, including sensing, optical & wireless communications, digital display, and biotechnology. Despite the enormous research in creating new applications with MEMS, the research efforts at the backend, such as assembly and packaging, are relatively limited. One main reason for such limitation is that the manipulation of micro-sized objects is poorly understood: due to scaling effects, unmodeled forces that are insignificant at the macro scale become dominant at the micro scale. For example, when the part to be handled is less than one millimeter in size, adhesive forces between the manipulator and the part can be significant compared to gravitational forces. These adhesive forces arise primarily from surface tension, van der Waals, and electrostatic attractions and can be a critical factor for part manipulation. Furthermore, manufacturing processes which are capable of quickly and cheaply assembling MEMS devices have not been developed, partly because, at the micro-scale, structures are fragile and easily breakable. They typically break at the micro-Newton force range-- a range that cannot be felt by a human operator assembling microstructure with tweezers and microscopes, and is not reliably measurable by existing force sensors. As a result, it is extremely difficult to manipulate parts for assembly at that scale. Such obstacle is especially problematic to devices requiring assembly such as 3-D micro-mirrors used to switch optical signals in the all-optical communications network. It has been widely realized that there has not been a reliable micromanipulation sensor that can be used to accurately obtain micro-force data during assembly, and no force/impact control method can be used to effectively regulate the contact force/impact in the micromanipulation. The consequence of these efforts is that devices are often damaged during assembly, decreasing overall yield and driving up cost. For these reasons, research into automating the microassembly/micromanipulation process has increasingly focused on force sensing and control techniques.
2. Objectives The objectives of this proposed project are to pursue the most feasible and versatile solution in microassembly/micromanipulation, i.e., the use of piezo-sensors as an active feedback of force data during the process of assembly. Piezoelectric material such as polyvinylidene fluoride (PVDF) is highly sensitive to deflection from which force data can be extracted. By integrating the PVDF on a probe-tip used for contact micromanipulation and assembly, the force data can be provided to an intelligent learning controller. It will establish a micromanipulation model based on the on-line force information, which will be used to regulate the contact force/impact. These sensors will be used for automated microassembly, a process in which the computer controls micromanipulators, and contact force/impact can be regulated to maintain a safety margin during assembly. The resulting higher and faster yield will make batch fabrication & assembly of MEMS a reality.
3. Research Tasks To achieve these goals, several key problems will be investigated and solved. 1) Optimization in the design of the PVDF sensors for the highest sensitivity and signal to noise ratio in the micro-scale; 2) Development of on-line signal processing scheme for sensory data; 3) Integration of PVDF sensors with electronics using VLSI technology; 4) Development of a novel neuro-learning scheme for force/impact control, and automation of micromanipulation process with sensor feedback.
4. Experimental Studies
Setup
The experimental
testing and evaluation will be conducted in the Robotics and Automation
Laboratory at Michigan State University. In the laboratory, a micro robotic
system, as shown in Figure 1, is available for this project. It consists of a
3-DOF micromanipulator (SIGNATONE Computer Aided Probe Station), a 3-DOF
platform, a Mitutoyo FS60 optical microscope and a Sony SSC-DC50A CCD Color
Video Camera. The micro robot is controlled by a PC-based control system.
Several motion control softwares and user interfaces have been developed. They
provide a flexible and user-friendly platform to integrate with the force
sensing system and control schemes. To reduce vibrations, the micromanipulator
is installed on an active vibration isolated table.
Besides the robotics system, Figure 2 shows the prototypes of the developed PVDF sensors for the experimental studies.
Force-Guided
Assembly of Micro Mirrors
The micro mirror components are usually three-dimensional optomechanical
structures. They lie on the surface of substrate after fabricatio processes, as
shown in Figure 3(a). An assembly process is needed to lift it up to a upright
position as shown in Figure 3(b),(c). Preliminary experiment of the micro mirror
assemblied by the guidance of the contact force has been implemented. A
developed PVDF force sensor attaching a probe tip is mounted at the front of the
3-DOF micromanipulator. The assembly task includes multiple segments of actions.
Firstly, the probe tip will be moved to the position under a micro mirror. The
tip, then, will lift the mirror up to an upright position until it is locked by
the latches. Finally, the probe tip will go back to the initial position and be
ready for the next assembly process. The whole force-guided assembly process
will be performed automatically. The initial and final assembly statuses in
Figure 4 show a successful assembly of a MUMPs 43 micro mirror using the
developed force sensor and control scheme.
Figure 3: (a) Micro mirrors on a MUMPs 43 chip.
(b)
A force is applied at the mirror.
(c)
The automatic latches lock the mirror in upright position.
Figure
4: (a) A probe tip will be moved to
the position under a micro mirror.
(b) The automatic latches lock the
mirror in upright position after lifting up by a sensor tip using the
force-guided approach.
.
Figure 6: (a) The sensor tip is down from an initial point in free space to a
contact point on the constrained glass surface, the desired contact force is
.
.
.
.
Force-guided assembly:
Publication:
G. Y. Li and N. Xi, Calibration of a Micromanipulation System, IEEE 2002 International Conference on Intelligent Robotics and Systems, Sep. 30, 2002, Switzerland
Wen-Li Zhou, Allan P. Hui, Wen J. Li and Ning Xi, “Development of a Force-Reflection Controlled Micro Underwater Actuator” The proceedings of the 2001 IEEE/RSJ International Conference on Intelligent Robots and Systems, Maui, Hawaii, USA, Oct 29-Nov.03, 2001
Antony W.T. Ho, Imad. Elhajj, Wen J. Li, Ning Xi, and Tao Mei, “ A Bone Reaming System Using Micro Sensors For Internet Force-feedback Control” , Proceedings of the 2001 IEEE International Conference on Robotics and Automation, Seoul, Korea, May 21-26, 2001
Antony W. T. Ho, Wen J. Li, Imad Elhajj, Ning Xi "Development of a Bone Reaming System Using Micro Sensors for Internet Force-Feedback Control", Proceedings of the Workshop on Service Automation and Robotics, Hong Kong, June 2000.
King W. C. Lai, Carmen K. M. Fung, Wen J. Li, Imad Elhajj, Ning Xi "Internet Force-Feedback Control of Micro Actuators", Proceedings of International Symposium on Information Technology and Communication, Bangkok, August 2000.
King W. C. Lai, Carmen K. M. Fung, Wen J. Li, Imad Elhajj, Ning Xi "Transmission of Multimedia Information on Micro Environment via Internet", Proceedings of the IEEE International Conference on Industrial Electronics, Nagoya Japan, 2000.
Carmen K. M. Fung, King W. C. Lai, Wen J. Li, Yunhui Liu, Imad Elhajj, Ning Xi"Sensing and Action in a Micro Environment via Internet", Proceedings of 2000 International Conference on Information Society in the 21st Century (IS2000), November 5-8, 2000, Aizu, Japan.