George K. Knopf

Rapid fabrication of tooling for microfluidic devices via laser micromachining and hot embossing

This paper presents a new method for rapid fabrication of polymeric micromold masters for the manufacture of polymer microfluidic devices. The manufacturing method involves laser micromachining of the desired structure of microfluidic channels in a thin metallic sheet and then hot embossing the channel structure onto poly(methyl methacrylate) PMMA substrate to form the mold master. The channeled layer of the microfluidic device is then produced by pouring the polydimethylsiloxane (PDMS) elastomer over the mold and curing it. The method is referred to as LHEM (laser micromachining, hot embossing and molding). Polymers like PDMS are preferred over silicon as the material for building microfluidic devices because of their biocompatibility properties as well as because of their lower cost. The proposed manufacturing method involves fewer processing steps than the conventional soft lithography process and enables manufacture of non-rectangular channels in microfluidic devices. To test the method, a mold for a micro capillary electrophoresis microfluidic chip was fabricated. The experimental results confirmed that high quality (Ra 10 to 100 nm) molds can be fabricated quickly and inexpensively. Advantages and limitations of the proposed method are discussed in the concluding section of the paper.

Fuzzy neural network approach to control systems

A mathematical model for an adaptive fuzzy neuron is proposed. Each neuron within a network corresponds to a fuzzy inference rule. These neurons may learn from experience via the adaptation of synaptic modifiers. The parallel structure of a fuzzy neural network controller enables complex decisions to be made in real-time. A simplified example of a neural network for controlling the steering of an automobile is used to illustrate these notions.<<ETX>>

Bioelectronic Imaging Array Based on Bacteriorhodopsin Film

A photoreceptor array that exploits the light sensitive bacteriorhodopsin (bR) films has been manufactured on a flexible indium-tin-oxide (ITO) coated plastic film using electrophoretic sedimentation technique (EPS). The effective sensing area of each photoreceptor is 2 × 2 mm2, separated by 1 mm and arranged in a 4 × 4 array. A switched integrator with gain on the order of 1010 is used to amplify the signal to a suitable level. When exposed to light, the differential response characteristic is attributed to charge displacement and recombination within bR molecules, as well as loading effects of the attached amplifier. The peak spectral response occurs at 568 nm and is linear over the tested light power range of 200 ¿ W to 12 mW. The response remains linear at other tested wavelengths, but with reduced amplitude. Initial tests have indicated that responsivity among all photoreceptors is greater than 71% of the average value, 465.25 mV/mW. The differential nature of the signal generated by bR makes it a suitable sensing material for vision applications such as motion detection. The prototype array demonstrates this property by employing Reichardts delay-and-correlate algorithm. Furthermore, fabricating sensor arrays on flexible substrates introduces a new design approach that enables non-planar imaging surfaces.

Visualization of randomly ordered numeric data sets using spherical self-organizing feature maps

Abstract Scientific data visualization requires a variety of mathematical techniques to transform multivariate data sets into simple graphical objects, or glyphs, that provide scientists and engineers with a clearer understanding of the underlying system behaviour. The spherical self-organizing feature map (SOFM) described in this paper exploits an unsupervised clustering algorithm to map randomly organized N -dimensional data into a lower three-dimensional (3D) space for visual pattern analysis. Each node on the spherical lattice corresponds to a cluster of input vectors that lie in close spatial proximity within the original feature space, and neighbouring nodes on the lattice represent cluster centres with a high degree of vector similarity. Simple metrics are used to extract associations between the cluster units and the input vectors assigned to them. These are then graphically displayed on the spherical SOFM as either surface elevations or colourized facets. The resulting colourized graphical objects are displayed and manipulated within 3D immersive virtual reality (IVR) environments for interactive data analysis. The ability of the proposed algorithm to transform arbitrarily arranged numeric strings into unique, reproducible shapes is illustrated using chaotic data generated by the Lozi, Henon, Rossler, and Lorenz attractor functions under varying initial conditions. Implementation of the basic data visualization technique is further demonstrated using the more common Wisconsin breast cancer data and multi-spectral satellite data.

Adaptive reconstruction of free-form surfaces using Bernstein basis function networks

Abstract Many computer graphics and computer-aided design (CAD) applications require mathematical models to be generated from measured coordinate data. The three-dimensional surface model of a complex object can be created by fitting the data with low-order parametric surface patches, and adjusting the control parameters such that the constituent patches meet seamlessly at their common boundaries. Most neural networks that are designed for function approximation produce a solution that is not easily transferable to CAD software. An adaptive technique to reconstruct a smooth surface from Bezier patches is presented in this paper. The approach utilizes a neural network that performs a weighted summation of Bernstein polynomial basis functions. The number of basis neurons is related to the degree of the Bernstein polynomials and is equivalent to the number of control parameters. Free-form surfaces are reconstructed by simultaneously updating networks that correspond to the separate patches. A smooth transition between adjacent Bezier surface patches is achieved by imposing positional and tangential continuity constraints on the weights during the adaptation process. The final weights of the various networks correspond to the control points of the stitched Bezier surface, and can therefore be used directly in many commercial CAD packages. This method has been used to create a closed surface by adaptively fitting two adjacent patches to synthetic range data and a complex open surface by fitting four patches to experimentally measured range data.

Search-Guided Sampling to Reduce Uncertainty of Minimum Deviation Zone Estimation

Integrating computational tasks in coordinate metrology and its effect on the inspection’s uncertainty is studied. It is shown that implementation of an integrated inspection system is crucial to reduce the uncertainty in minimum deviation zone (MDZ) estimation. An integrated inspection system based on the iterative search procedure and online MDZ estimation is presented. The search procedure uses the Parzen Windows technique to estimate the probability density function of the geometric deviations between the actual and substitute surfaces. The computed probability density function is used to recognize the critical points in the MDZ estimation and to identify portions of the surface that require further iterative measurements until the desired level of convergence is achieved. Reduction of the uncertainty in the MDZ estimation using the developed search method compared to the MDZ estimations using the traditional sampling methods is demonstrated by presenting experiments including both actual and virtual inspection data. The proposed search method can be used for assessing any geometric deviations when no prior assumptions about the fundamental form and distribution of the underlying manufacturing errors are required. The search method can be used to inspect and evaluate both primitive geometric features and complicated sculptured surfaces. Implementation of this method reduces inspection cost as well as the cost of rejecting good parts or accepting bad parts. DOI: 10.1115/1.2798114

Surface reconstruction using neural network mapping of range-sensor images to object space

Range sensors that employ structured-light triangulation techniques often require calibration procedures, based on the sys- tem optics and geometry, to relate the captured image data to object coordinates. A Bernstein basis function (BBF) neural network that directly maps measured image coordinates to object coordinates is described in this paper. The proposed technique eliminates the need to explicitly determine the sensors optical and geometric pa- rameters by creating a functional map between image-to-object co- ordinates. The training and test data used to determine the map are obtained by capturing successive images of the points of intersec- tion between a projected light line and horizontal markings on a calibration bar, which is stepped through the object space. The sur- face coordinates corresponding to the illuminated pixels in the im- age are determined from the neural network. An experimental study that involves the calibration of a range sensor using a BBF network is presented to demonstrate the effectiveness and accuracy of this approach. The root mean squared errors for the x and y coordinates in the calibrated plane, 0.25 and 0.15 mm, respectively, are quite low and are suitable for many reverse engineering and part inspec- tion applications. Once the network is trained, a hand carved wooden mask of unknown shape is placed in the work envelope and translated perpendicular to the projected light plane. The surface shape of the mask is determined using the trained network.

Photoelectric Monolayers Based on Self-Assembled and Oriented Purple Membrane Patches

The fabrication of dry ultrathin photoelectric layers is described in this paper. The self-assembled monolayer of oriented purple membrane (PM) patches from bacteriorhodopsin (bR) is created on a bio-functionalized gold (Au) surface using a biotin molecular recognition technique. The biotin enables the extracellular side of the bR PMs to be accurately labeled. During the biochemical immobilization process, the biotinylated alkylthiols modify the Au surface using HS terminals of the thiols and affix the labeled bR to the functionalized surface using streptavidin-biotin interactions. An optically transparent indium tin oxide (ITO) electrode is then placed on top of the finished assembly to complete the circuit of the photocell for testing and performance verification. An experimental study shows that the proposed self-assembly method can produce a 12.33 nm thick photoelectric layer that generates nearly 0.54 mV/(mW ·cm2) when exposed to a 568-nm laser beam. In contrast, other immobilization techniques such as electric field sedimentation produce dry bR films that are more than 10 μm thick. These ultrathin photoelectric layers can be used to create nanoscale light sensors and photocells.

Design, kinematic modeling and performance testing of an electro-thermally driven microgripper for micromanipulation applications

Microgripping systems incorporate miniature end-effectors used to manipulate micro-sized objects such as tiny mechanical parts, electrical components, biological cells and bacteria. This paper presents a thorough study of the design, kinematics and static/dynamic performances, including electro-thermo performance characteristics, of the new microgripping system. The developed microgripper had a monolithic design which consisted of a combination of an in-plane electro-thermally driven microactuator and a compliant tweezing mechanism. The kinematics of the microgripper was studied as a transformation of input linear actuation motions into output tweezing displacements and compared with microgripper prototypes fabricated from 25 µm thick nickel foil by using laser micromachining technology. The static, dynamic and electro-thermal characteristics of the system performance were analyzed with respect to actual actuation motions, tweezing displacements, voltage, power, electric resistance and overall temperature under constant applied current within a range of {20, 40, ..., 160} mA. Maximum tweezing displacements of 47.5 µm (tweezing gap of 94.9 µm) were achieved under an applied current of 160 mA for a fabricated microgripper having a transform coefficient K = 1.731. The repeatability and reliability of the fabricated microgripper were also tested along with the capability to grip, hold and release a 110 µm diameter glass bead proving that this microgripper can be utilized as a grasping end-effector for micromanipulation, microrobotic and microassembly applications.

Adaptive reconstruction of bone geometry from serial cross-sections

Abstract Many biomedical applications, such as the design of customized orthopaedic implants, require accurate mathematical models of bone geometry. The surface geometry is often generated by fitting closed parametric curves, or contours, to the edge points extracted from a sequence of evenly spaced planar images acquired using computed tomography (CT), magnetic resonance imaging (MRI), or ultrasound imaging. The Bernstein basis function (BBF) network described in this paper is a novel neural network approach to performing functional approximation tasks such as curve and surface fitting. In essence, the BBF architecture is a two-layer basis function network that performs a weighted summation of nonlinear Bernstein polynomials. The weight values generated during network training are equivalent to the control points needed to create a smooth closed Bezier curve in a variety of commercially available computer-aided design software. Modifying the number of basis neurons in the architecture is equivalent to changing the degree of the Bernstein polynomials. An increase in the number of neurons will improve the curve fit, however, too many neurons will diminish the networks ability to generate a smooth approximation of the cross-sectional boundary data. Additional constraints are imposed on the learning algorithm in order to ensure positional and tangential continuity for the closed curve. A simulation study and real world experiment are presented to show the effectiveness of this functional approximation method for reverse engineering bone structures from serial medical imagery.