Kelvin K. L. Cheo

Neural network approach for linearization of the electrostatically actuated double-gimballed micromirror

In this paper, a hierarchical circuit based approach is used for the development of a reduced-order macro-model for a double-gimballed electrostatic torsional micromirror. The nonlinearity and cross-axis coupling of the micromirror subjected to the differential driving scheme are investigated using the proposed macro-model. The simulation results are used to train a feed-forward neural network which carries out a function approximation of the relation between the desired location and the required driving voltages. The trained neural network is then coded into MAST AHDL. System-level simulation of the micromirror together with the neural network is performed in the SABER/spl trade/ simulator. It is found that using a feed-forward neural network, the linearity of the micromirror is greatly improved, the steady state of the cross-axis coupling is reduced to a negligible level and the transient response of the cross-axis coupling is also suppressed. This implies that introducing a feed-forward neural network would be useful to simplify the design of the feedback control system for the double-gimballed electrostatic torsional micromirror.

Hyperspectral imaging using a microelectrical-mechanical-systems-based in-plane vibratory grating scanner with a single photodetector

We present a single-photodetector-based hyperspectral imaging system that utilizes a microelectrical-mechanical-systems-driven diffraction grating for fast spatial scanning and two synchronized steering mirrors for slow spectral scanning. This configuration allows high-speed scanning without degradation in optical performance resulting from dynamic non-rigid-body deformation of the platform. The proposed operational principle is demonstrated with a prototype device developed using silicon microfabrication technology. The proposed spectral imaging system has the potential to achieve low cost, small form factor, and high-speed operation.

A high-speed MEMS grating laser scanner with a backside thinned grating platform fabricated using a single mask delay etching technique

A novel micro-electromechanical system (MEMS) technology-based grating laser scanner with a backside thinned grating platform has been successfully developed for high-speed laser scanning applications. The grating platform is thinned by a round cavity and reinforced by a circular frame, which are fabricated using a single mask delay etching (SMDE) technique. The SMDE technique, which utilizes the well-know loading effects of the deep reactive ion etching (DRIE) process, is a simple and low-cost methodology to regulate the etching rate of a prescribed area. It can be used in a silicon-on-insulator (SOI) micromachining process to form multilevel structures in a silicon device layer through a multi-step DRIE process from a wafers backside. This paper presents the design, simulation, fabrication process and characterization of the high-speed MEMS grating scanner as well as the principle and applications of the SMDE technique. When illuminated with a 635 nm wavelength incident laser beam, the prototype scanner with a 1 mm diameter diffraction grating is capable of scanning at 50.192 kHz with an optical scan angle of 14.1°.

Neural Network Approach for Linearization of the Electrostatically Actuated Double-Gimballed Micromirror

In this paper, a hierarchical circuit based approach is used for the development of a reduced-order macro-model for a double-gimballed electrostatic torsional micromirror. The nonlinearity and cross-axis coupling of the micromirror subjected to the differential driving scheme are investigated using the proposed macro-model. The simulation results are used to train a feed-forward neural network which carries out a function approximation of the relation between the desired location and the required driving voltages. The trained neural network is then coded into MAST AHDL. System-level simulation of the micromirror together with the neural network is performed in the SABER™ simulator. It is found that using a feed-forward neural network, the linearity of the micromirror is greatly improved, the steady state of the cross-axis coupling is reduced to a negligible level and the transient response of the cross-axis coupling is also suppressed. This implies that introducing a feed-forward neural network would be useful to simplify the design of the feedback control system for the double-gimballed electrostatic torsional micromirror.

Post-Corrections of Image Distortions in a Scanning Grating-Based Spectral Line Imager

In this letter, a post-correction algorithm, which is capable of compensating typical image distortions in a scanning grating-based spectral line imager, is presented in detail. A prototype spectral line imager with a single-pixel detector is built for demonstration purposes. A double-layered vibratory grating scanner and a galvano-scanning mirror are utilized in the imager for fast spatial scanning and slow spectral scanning, respectively. There are three major imager distortions exist in the prototype system, which are spatial smile distortion, spectral keystone distortion, and nonuniform sampled spatial positions due to resonant spatial scanning. They are then corrected using a post-correction algorithm, which is based on reverse ray-tracing and Delaunay Triangulation mapping. The scanning grating-based spectral imager with the post-correction algorithm has the potential to achieve low-cost, small form-factor, and high-speed operation.

A spectral line imager based on a MEMS vibratory grating scanner

A single-pixel spectral line imager based on a highspeed MEMS grating scanner is demonstrated to possess the capability of resolving 104 spectral bands and 250 spatial pixels. Delauney Triangulation is used in the signal processing to correct the spectral image distortions.

A high-speed electrostatic double-layered vibratory grating scanner with very high optical resolution

This paper demonstrates the design, fabrication and characterization of a micromachined electrostatic double-layered vibratory grating scanner for high speed, high optical resolution laser scanning applications. The prototype scanner with a 2mm diameter diffraction grating is capable of scanning a laser beam at around 21.591kHz with an optical scan angle of 33.5°, resulting a θopticalD product (product of optical scan angle and diameter of the diffraction grating) of 67 deg·mm. The optical resolution of the prototype scanner is measured to be 916 pixels per-unidirectional-scan while scanning in atmosphere. While scanning in vacuum condition, the optical resolution can be estimated to be 1460 pixels per-unidirectional-scan, which is suitable for SXGA (1280×1024) resolution scanned beam displays.

Built-in optical angular position sensing mechanism for high-resolution vibratory grating scanner

A built-in angular sensing mechanism unique to the vibratory grating scanner is demonstrated through utilizing the zeroth-order beam, of which the intensity variation is affected by the angular position of the grating platform. With the first-order diffracted beam used for optical scanning, the stationary zeroth-order beam can be used for angular position sensing.

MEMS-driven diffraction gratings for rapid scanning of laser beams with very high optical resolution

Miniaturized low-power, high-speed scanners are tremendously useful in a variety of applications. Besides MEMS micromirrors, the in-plane vibratory grating scanner is a development in this area which possesses several unique features. The in-plane scanning mechanism minimizes the dynamic non-rigid-body out-of-plane deformation of the mirror surface, allowing for higher-resolution displays. The dispersive element permits splitting the incoming beam into its constituents for analysis. Coupling a grating platform to an in-plane moving structure is also useful for real-time motion measurement which would otherwise be difficult to pick-up. The past developments till the current design are explored in this paper. Possible alternative applications besides image display, for example spectral imaging and realtime motion sensing are also described.

Synchronized laser scanning of multiple beams by MEMS gratings integrated with resonant frequency fine tuning mechanisms

This paper presents an effective method to achieve synchronized laser scanning of multiple beams by using MEMS diffraction gratings with their resonant frequency fine tuning mechanisms. Multiple gratings are actuated in-plane by a common electrostatic comb-driven resonator and their resonant frequencies can be fine-tuned to compensate the micromachining process errors. Continuous and reversible resonant frequency tuning was achieved. The resonant frequency of one diffraction grating gradually dropped from 19870 Hz to 19588 Hz with its tuning voltages increased from 0V to 5V. Finally, synchronized laser scanning of multiple beams was demonstrated using stroboscopic method.