Classifications

• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
  • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
  • G06F3/03—Arrangements for converting the position or the displacement of a member into a coded form
  • G06F3/033—Pointing devices displaced or positioned by the user, e.g. mice, trackballs, pens or joysticks; Accessories therefor
  • G06F3/038—Control and interface arrangements therefor, e.g. drivers or device-embedded control circuitry
  • G06F3/0383—Signal control means within the pointing device
• • G—PHYSICS
  • G06—COMPUTING OR CALCULATING; COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
  • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
  • G06F3/03—Arrangements for converting the position or the displacement of a member into a coded form
  • G06F3/0304—Detection arrangements using opto-electronic means
  • G06F3/0317—Detection arrangements using opto-electronic means in co-operation with a patterned surface, e.g. absolute position or relative movement detection for an optical mouse or pen positioned with respect to a coded surface

Definitions

• the present invention relates generally to optical navigation systems and methods of sensing movement using the same.
• Data input devices such as computer mice, touch screens, trackballs and the like, are well known for inputting data into and interfacing with personal computers and workstations. Such devices allow rapid relocation of a cursor on a monitor, and are useful in many text, database and graphical programs.
• a user controls the cursor, for example, by moving the mouse over a surface to move the cursor in a direction and over distance proportional to the movement of the mouse.
• Computer mice come in both optical and mechanical versions. Mechanical mice typically use a rotating ball to detect motion, and a pair of shaft encoders in contact with the ball to produce a digital signal used by the computer to move the cursor.
• One problem with mechanical mice is that they are prone to inaccuracy and malfunction after sustained use due to dirt accumulation, etc. hi addition, the movement and resultant wear of the mechanical elements, particularly the shaft encoders, necessarily limit the useful life of the device.
• Cypress/SLM ref.: SLMP0352-PCT mice have become very popular because they provide a better pointing accuracy and are less susceptible to malfunction due to accumulation of dirt.
• CMOS complementary metal-oxide- semiconductor
• speckle a complex interference pattern
• Laser-based light generation has a high electrical-to-light conversion efficiency, and a high directionality that enables a small, efficient illumination footprint tailored to match a footprint of the array of photodiodes.
• speckle patterns allow tracking operation on virtually any rough surfaces (broad surface coverage), while maintaining the maximum contrast even under unfavorable imaging condition, such as being "out-of-focus”.
• An alternative approach for measuring linear displacements uses an optical sensor having one-dimensional (ID) arrays of photosensitive elements, such as photodiodes, commonly referred to as a comb-array.
• the photodiodes within a ID array may be directly wired in groups to enable analog, parallel processing of the received signals, thereby reducing the signal processing required and facilitating
• the present invention provides a solution to these and other problems, and offers further advantages over conventional devices and methods of using the same.
• FIG. 1 (prior art) is a schematic block diagram of a linear, one-dimensional (ID) comb-array in a four (4) photosensitive elements per period configuration and the associated cosine and sine templates;
• FIGs. 2A -2D are matrices showing cosine and sine assignments for a two- dimensional (2D) comb-array according to an embodiment of the present invention
• FIGs. 3 A and 3 B are schematic block diagrams of a 2D comb-array constructed from the matrices of FIGs. 2 A -2D and having photosensitive elements
• FIGs. 4A and 4B are diagrams comparing two orthogonal (or ID x ID)
• FIG. 5 is a schematic block diagram of an optical navigation system having a speckle-based 2D comb-array according to an embodiment of the present invention
• FIGs. 6 A and 6B are graphs of circular trajectories at various speeds and over various surfaces for an optical navigation system with a 2D comb-array according to an embodiment of the present invention versus actual movement of the system;
• FIG. 7 is a schematic block diagram of a 2D comb-array having
• FIG. 8 is a schematic block diagram of an optical sensor having two 2D comb-arrays arranged in quadrants according to an embodiment of the present invention.
• FIG. 9 is a schematic block diagram of an optical sensor having a 2D array of hexagonal photosensitive elements, wired to detect ID motion along three different axes according to another embodiment of present invention.
• FIG. 10 is a dot pattern for a non-periodic phyllotactic array of photosensitive elements according to yet another embodiment of the present invention.
• FIG. 11 is a Voronoi diagram showing the photosensitive element pattern for the phyllotactic array of FIG. 10;
• FIG. 12 is a schematic block diagram of an optical sensor having a hexagonal 2D array of hexagonal photosensitive elements according to another embodiment of the present invention.
• FIG. 13 is a schematic block diagram of an optical sensor having a square 2D array wired for 4-axis motion detection according to an embodiment of the present invention.
• the present invention relates generally to optical navigation systems, and more particularly to optical sensors for sensing relative lateral movement between the sensor and a surface on or over which it is moved.
• Optical navigation systems can include, for example, an optical computer mouse, trackballs and the like, and are well known for inputting data into and interfacing with personal computers and workstations.
• Cypress/SLM ref.: SLMP0352-PCT invention may be practiced without these specific details. Irx other instances, well- known structures, and techniques are not shown in detail or are shown in block diagram form in order to avoid unnecessarily obscuring an. understanding of this description.
• Reference in the description to "one embodiment” or “an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention.
• the appearances of the phrase “one embodiment” in various places in the specification do not necessarily all refer to the same embodiment.
• the teirn "to couple” as used herein may include both to directly connect and to indirectly connect through one or more intervening components.
• the optical sensor of the present invention is a speckle-based sensor, which senses movement based on displacement of a complex intensity distribution pattern of light, known as speckle.
• Speckle is essentially the complex interference pattern generated by scattering of coherent light off of a rough surface and detected by an intensity photosensitive element, such as a photodiode, with a finite angular field-of-view (or numerical aperture).
• the optical sensor includes a two-dimensional (2D) array that com ⁇ bines the displacement measurement accuracy of a 2D correlator with the signal processing simplicity and efficiency of a linear or one-dimensional (ID) comb-array.
• the 2D array may be either a periodic, 2D comb-array, which includes a number of regularly spaced photosensitive elements having ID or 2D periodicity, a quasi-periodic 2D array (such as a Penrose tiling), or a non-periodic 2D array, which has a regular pattern but doesn't include periodicities.
• a 2D comb-array it is meant a planar array of a number of regularly spaced and electrically connected photosensitive elements
• ID correlation it is instructive to compare the signal processing for image correlation to a comb-array technique in one-dimension (ID).
• DFT discrete Fourier transform
• ID comb-array A linear or ID comb-array is an array having multiple photosensitive elements that are connected in a periodic manner, so that the array acts as a fixed template that interrogates one spatial frequency component of the signal.
• An embodiment of one such ID comb-array is shown in FIG. 1 and described in greater detail below. The connection of multiple photosensitive elements in a periodic
• Cypress/SLM ref.: SLMP0352-PCT manner enables the comb-array to serve effectively as a correlator at one spatial frequency K (defined by a pitch of the photosensitive elements in the array and the collection optics).
• K defined by a pitch of the photosensitive elements in the array and the collection optics.
• the comb signal now viewed as a function of ttie displacement x, is:
• C is a slowly varying amplitude and K ⁇ 2 ⁇ /4/iV the selected spatial frequency.
• the factor e lKm can be thought as the phase that encodes the initial alignment of the selected spatial frequency component and the template.
• a ID comb-array is essentially a ID correlation at one spatial frequency.
• K (K x , K y ).
• V Xiy Ce iK ⁇ "' ⁇ x) e iK ⁇ " ⁇ y) (8)
• (K x , K y ) ⁇ (2 ⁇ A/N, 2 ⁇ B/N) is the selected 2D spatial frequency.
• the comb signal is simply the product of harmonic functions of the x and y displacements. Notice that the comb-array signal is periodic and peaks whenever the template is spatially in-phase with the image spatial frequency.
• the next step is to determine the 2D array configuration that generates the four signals shown in (9) above.
• FIG. 1 shows a general configuration (along one axis) of a ID comb-arra ⁇ y 102 of photosensitive elements, such as photodiodes 104, wherein the combination of interlaced groups of photosensitive elements serves as a periodic filter on spatial frequencies of light-dark signals produced by the speckle (or non-speckle) images.
• the ID comb-array 102 consists of a number of photodiode sets or periods, each having four of photodiodes 104, labeled h.ere as A,
• Comparing the phase of the in-phase and quadrature signals permits determination of the magnitude and direction of motion of the ID comb-array 102 relative to a scattering surface.
• the in-phase C 0111 and the quadrature S out signals are obtained by taking the underlying speckle pattern and processing them according to the cosine and sine templates, 112 and 114 respectively.
• the system is designed so that an optical "light-dark" signal pattern, i.e., speckle, has a size substantially equal to the period of the comb-array - four (4) photodiodes 104 or pixels in the embodiment of FIG. 1.
• the in-phase signal current is obtained from
• FIG. 2A shows the matrix of the CC or cos (K x x) cos (Ky y) signal for a 2D comb-array having photosensitive elements
• FIG. 2B shows the matrix for the GS signal
• FIG. 2C shows the matrix for the SC signal
• FIG. 2D shows the matrix for the SS signal.
• a 2D comb-array can now be constructed from the above matrices, as shown in FIGs. 3A and 3B.
• the 2D comb-array 302 has multiple photosensitive elements 304 arranged or grouped into cells 306, each cell having photosensitive
• Photosensitive elements 304 within a cell 306 with the same letter and same number, as shown in the detail of FIG. 3 B, as well as corresponding elements of all cells in the 2D comb-array 302 with the same number, are electrically connected or wired- sum to yield eight signals Al through D2.
• the eight wired-sum signals are further combined with differential amplifiers 308 to give the following four signals:
• the coordinate system or the array can be rotated by 45° to get
• the 2D comb-array offers a simplicity of design and several further advantages over the conventional 2D correlation and/or multi-axis ID comb-array, including: (i) faster signal processing; (ii) reduced power consumption; (iii) high angular accuracy; and (iv) performance that is independent of a direction movement relative to an array orientation.
• the TD comb-array has significantly faster signal processing than correlation because it generates much less data to process, and consequently much simpler algorithms to execute. For example, zero-crossing detection algorithm can be employed to determine the displacements. To specify a displacement in a plane, two real numbers are needed, namely the x and y translations. In a conventional correlation-based optical mouse, these two real numbers are determined from
• Cypress/SLM ref. SLMP0352-PCT successive image correlation. Because each image in the correlation-based approach typically comprises about 10 3 pixels, a large amount of data needs to be processed just to determine the two x- and ⁇ -translation values. In contrast, the 2D comb-array produces only four (4) positive real numbers, which are equivalent to just two (2), signed real numbers. In a sense, parallel processing is built into the inter-connection architecture of the 2D comb-array. By "wiring" the processing into the architectuure, the remaining external computation becomes relatively simple and can be accomplished quickly. Simple computation translates to smaller signal processing circuitry, while faster processing allows high velocity tracking and increased resources to implement sophisticated digital signal processing (DSP) algorithms ttiat can boost tracking performance of an optical navigation system using the optical sensor of the present invention even further.
• DSP digital signal processing
• the 2D comb-array is expected to consume less electric power than a correlation-based device because it has much less data to process, and consequently much simpler algorithms to implement. This is a highly desirable feature for power- sensitive applications such as a wireless optical mouse.
• the electric power consumption can be further reduced by combination with efficient laser illumination, such as in laser speckle based mice.
• the angular accuracy of a 2D comb-array can be scaled much easier than that of a conventional 2D correlator mouse.
• the minimum angle that can be detected by a 2D sensor is inversely proportional to the number of photosensitive elements in a row or a column. Improving angular accuracy depends generally on an increase in the number of photosensitive elements of the array. This constitutes a heavy penalty for a 2D correlator mouse, because the quantity of data to be processed goes izp quadratically with the number of elements in a row or a column.
• tfcie the quantity of data to be processed goes izp quadratically with the number of elements in a row or a column.
• Cypress/SLM ref.: SLMP0352-PCT quantity of data or number of signals to be processed in a 2D comb-array mouse is independent of the number of elements. That is, the number of differential signals output from the 2D comb-array is always equal to four in a 2D comb-array having a configuration similar to that shown in FIGs. 3A and 3B, and therefore the angular accuracy is limited only by the size of the array that can be implemented.
• the performance of the 2D comb- array is independent of the direction movement relative to the array.
• the performance of the 2D comb-array 402 is superior to an optical sensor 404 having multiple linear or ID comb-arrays 406 since each point in the image, on average, traverses a much longer path 408 inside the active area of the
• 2D comb-array 402 in all directions than a path 410 in the ID comb-array 406, and therefore contributes more to the displacement estimation.
• the embodiments of the 2D comb-array described heretofore operate with symmetric (e.g. square) pixel geometries, matching the "light-dark" signal pattern, i.e., speckle, to the period of the 2D comb-array is more easily achieved, resulting in improved signal contrast and higher front-end SNR than can be achieved with conventional ID comb arrays that typically employ highly “asymmetric" pixel shapes.
• the optical navigation system 502 generally includes an optical head 504 having a light source 506, such as a VCSEL (Vertical Cavity
• illumination optics including a first or collimating lens 508 to collimate a diverging light beam, imaging optics including a second or imaging lens 510 to map or image an illuminated portion of a rough, scattering surface 512 to a ZD comb-array 514 at the image plane of the second lens.
• the illumination optics are configured to illuminate the surface 512 at a predetermined incident angle selected to permit lift detection, by which the device ceases to track the motion if the separation of the of the optical head 504 or data input device from the surface 512 exceeds a predetermined separation.
• the imaging optics may include an aperture 516 at the back focal plane of the second lens 510 to provide a telecentric imaging system that preserves good speckle pattern integrity during motion and to match an average size of the speckle to a period of the 2D comb- array.
• FIGs. 6A and 6B The results for circular trajectories at various speeds and over two different surfaces, validating the disclosed approach are shown in FIGs. 6A and 6B.
• the graphs of FIG. 6A illustrate the circular trajectories produced when the optical head was moved four times in a circle having a radius of 1 cm over a white surface at speeds of 1 cm/s, 10 cm/s, 25 cm/s and 40 cm/s.
• FIG. 6B illustrate the circular trajectories produced when the optical head was moved over a wood grain surface at these same speeds.
• the dashed reference In FIGs. of 6A and 6B, the dashed reference
• Cypress/SLM ref.: SLMP0352-PCT circles are indicated by the reference number 602, and the traces or circular trajectories produced by the optical navigation system indicated by solid black lines. The numbers along the axes are in arbitrary units.
• an optical navigation system with a sensor using a 2D comb-array of the present invention is capable of sensing movement over patterned and un-patterned surfaces at speeds of up to 40 cm/s and with path errors of typically less than 5%. Subsequent testing has demonstrated accurate tracking performance for a wide variety of surfaces and a broad range of motions.
• a 2D comb-array in accordance with the present invention may also include a combination of the above generalizations or embodiments.
• 2D comb-array has other than 4x4 elements-
• the 2D comb-array 702 includes a number
• certain elements 704 within each cell 706, and corresponding elements of all cells in the 2D comb-array 702 are coupled to one of thirty-six (36) output lines.
• the 36 ⁇ ired-sum signals are further combined with weight factors in accordance with the matrices 708 to produce four output signals - CC, CS, SC and SS. Details of the matrices 708 used to produce each of these four signals are shown in detail in the tables below.
• the optical sensor can include multiple 2D comb-array or sub-arrays of a given spatial frequency or different spatial frequencies.
• FIG. 8 shows a schematic block diagram of an optical sensor 802 having two 2D comb-array-pairs arranged in quadrants 804, 806, 808 and 810 according to an embodiment of the present invention. Diagonally opposing quadrants 804 and 806 are connected and form a first single array-pair or first 2D comb-array. Opposing quadrants 808 and 810 are connected and form a second single array-pair or second 2D comb-array.
• elements within each cell 812 in a quadrant 804, 806, 808 and 810 as well as corresponding elements of all cells in the array-pair are coupled to form sixteen (16) wired-sum signals 814.
• the 16 wired- sum signals 814 are further combined with differential amplifiers 816 to produce eight (8) signals, CCl, GSl, SCl, SSl from the first 2D comb-array, and CC2, CS2, SC2, SS2 from the second 2D comb-array.
• the strengths of the signals from either of the 2D comb-arrays or array-pairs may decrease because the selected spatial frequency component is weak at some particular location on the surface, or because contributions from various parts of the array add coherently to zero.
• the square symmetry arrangement of the optical sensor 802 enables simple and efficient illumination of all photosensitive elements 818 in the optical sensor.
• any 2D array can be used to extract velocity vectors along any finite set of arbitrary axes.
• comb-arrays that had ID or 2D periodicity have been considered, but other patterns can be used having no periodicity or grouping into cells.
• a 2D array can have a sunflower floret pattern, or even a pseudo-randomly positioned number of photosensitive elements, and still extract ID or 2 D motion information. All of the following examples are particular cases of the general concept, i.e., use of a ID or 2D array to capture 1 -dimensional or 2-dimensional motion. Also described are a few new optical sensor configurations.
• the speckle pattern is a superposition of 2D spatial frequencies that contains a range of spatial frequencies determined by the optical properties of the system.
• ID motion that is motion along a selected axis or in a selected direction, is detected by selecting a particular distribution of light that is part of the speckle pattern, and observing how it changes with motion along the selected axis or in the selected direction.
• the distribution is one that doesn't change its shape with motion, other than a multiplicative constant (whose value yields the amount or magnitude of the motion). This means that if the distribution is ⁇ (x,y) it is desirable to keep a constant shape after some translation in some
• the elements in ID sensor be responsive to a translation in any direction, but extracts only the component of motion along the selected axis or in the selected direction.
• Cypress/SLM ref.: SLMP0352-PCT navigation system may be moved over a surface in both an x and a y direction, but extract only the component of motion along the x- axis.
• multiplicative constant Preferably, it is possible to extract the motion sought from the constant ⁇ .
• eigenfunctions of the translation operators are complex exponentials in the direction of translation and can have any functional form whatsoever in the perpendicular direction. That is, for translation in the x-direction, the eigenfunction is:
• f x is the spatial frequency in the x direction
• f y is the spatial frequency in the y direction
• Cypress/SLM ref. SLMP0352-PCT of the fact that eigenfunctions are orthogonal under inner product. So given a distribution that contains an eigenfunction of particular interest, it is possible to find out how much of it is present by taking the inner product with the eigenfunction sought.
• the translation operator is Hermitian under the inner product:
• eigenfunction sought with coefficient of proportionality c, c can be extracted by taking the inner product with the eigenfunction of interest:
• Equation 17 is integrated to obtain a value for complex c. After moving the sensor or array, the above integration is repeated to get a new value for complex c, d , which should be equal to c ⁇ .
• the eigenvalue, ⁇ is obtained by taking the ratio
• d x can be extracted from the computed value of ⁇ in a straightforward manner.
• ⁇ (x,y) exp(2 ⁇ x£.).
• the main and quadrature signals can be computed from the following equation:
• c ⁇ c r + ic, f f S(x, y)V ⁇ [exp C-l ⁇ i (xf x + ⁇ p 0 ))] dxdy + (21) i ⁇ ⁇ S(x,y)3 ⁇ ex.p(-2 ⁇ i(xf x + ⁇ Q ) ⁇ dxdy
• the eigenvalue ⁇ is then computed from two successive measurements, as described above with reference to equations 17, 18 and 19, and the distance moved in the x-direction extracted from the following equation:
• Cypress/SLM ref.: SLMP0352-PCT is most readily accomplished by performing a coordinate rotation so that an axis or direction x runs along the desired direction, and then all the formulas above apply in the rotated coordinate system.
• the array will include a number of photosensitive elements, each at some position (x,, ⁇ ) within the array, and having a total photosensitive element area
• a 1 trying to cover an array area B 1 so the ratio A 1 1 B 1 is the fill factor of the array.
• each photosensitive element is given a weighting coefficient w, , ,
• z means two different things in this expression: the first subscript means it's for the imaginary part of c; the second is the index of the photosensitive element.
• Cypress/SLM ref.: SLMP0352-PCT although the signal processing required for handling trie extra information is more complicated, offsetting some of the advantages realized by using a speckle-based optical sensor having a 2D array.
• There is some flexibility in how to provide this redundancy including: (i) several spatially separated arrays (spatial redundancy); (ii) different spatial frequencies for a given direction (spatial frequency redundancy); and (iii) multiple axes using more than 2 different directions (directional redundancy) - in which case, of course, the computed motion values are no longer necessarily orthogonal.
• any single motion component detected will be called a “signal/quadrature pair" (SQ-pair). If 2 or more SQ-pairs are to be detected an initial decision must be made as to how to the photosensitive elements in the array are used, hi particular, the elements in the array can be connected so that: (i) any given photosensitive element feeds only one SQ-pair; (ii)
• Which approach is used involves in part a tradeoff between power budget and IC or chip size, since signal splitters and buffers consume both, while not using the information from every photosensitive element in every SQ-pair can reduce signal-to-noise ratio (SNR).
• SNR signal-to-noise ratio
• An ideal photosensitive element configuration for detecting pure, ID motion will be considered first.
• the optimum configuration of the 2D array is having a photosensitive element everywhere with the weighting coefficient as follows:
• the cosine and sine have interleaved zeros. If the weighting coefficient is zero, then there is no need to waste a photosensitive element there, so if the photosensitive elements are spaced 1/4 of a period apart, the photosensitive elements for the main and quadrature signals can be interleaved, with each photosensitive element only contributing to a single signal. So one preferred 2D array for ID motion detection is photosensitive elements arranged in a number of vertical stripes, with alternating signals going to main and quadrature signals. Moreover, the stripes of photosensitive elements also do not need to be made continuous in the vertical direction, because the eigenfunction sought is continuous in the y direction.
• Cypress/SLM ref.: SLMP0352-PCT shows a hexagonal array 902, interleaved and wired to produce three (3) ID SQ- pairs along axes 904, 906, and 908, separated by 120°.
• a grid of hexagonal photosensitive elements 904 is wired to produce ID motion the along three different axes 904, 906, 908.
• Photosensitive elements 904 associated with each axis for detecting the in-phase signal are indicated by the same numbers, 1, 2, or 3.
• Alternating rows of photosensitive elements 904 for detecting the quadrature signal are indicated by like numbers, 1', 2', and 3'.
• In-phase and quadrature signals from photosensitive elements arranged along each axes 904, 906, 908, are wire-summed, with alternate signs indicating the in-phase (+) and quadrature (-) signals.
• any interrupted array of photosensitive elements arises from the fact that the signal is sampled, and therefore is susceptible to aliasing.
• an interrupted array will pick up any spatial frequencies that are at multiples of the underlying period of the photosensitive elements.
• the effect of aliasing can be reduced by using the signal from every photosensitive element, i.e., splitting each element output and sending a copy to every ID SQ-pair.
• This increases the sampling rate (since every photosensitive element is used, rather than every 2 n or 3 r element in a row), and also means using a smoother sampling function (since the sampling function is not a series of delta functions, but rather is convolved with a step function, which suppresses higher harmonics).
• some confounding aliased contributions are to be expected if the optical pattern contains strong periodicities, e.g., from a woven cloth or patterned surface.
• a way to reduce the susceptibility to aliasing would be to use a completely non-periodic array, and in particular, an array that has no strong peaks at any spatial
• Cypress/SLM ref. SLMP0352-PCT frequency.
• a particularly interesting aperiodic pattern is the so-called phyllotactic array, or "sunflower” array. It has a couple of nice properties: it is based on the Golden
• Ratio which is the most irrational of all numbers, meaning it minimizes the height of higher harmonics in its spectrum. It is also fairly easy to generate. In polar coordinates, theyth point is located at
• FIG. 10 An embodiment of a dot pattern for a phyllotactic array 1002 having 200 elements 1004 is shown in FIG. 10.
• the optimum photosensitive element size is the Voronoi diagram for the photosensitive element centers (a Wigner-Seitz cell).
• the photosensitive element 1102 pattern for the array 1104 would thus look like that shown in FIG. 11.
• Such a phyllotactic array 1002 has no strong peaks in its Fourier spectrum, but has roughly average photosensitive element size 1004. It would thus be resistant to aliasing when used on a patterned surface.
• a possible limitation to a 3 -axis 2D array, such as shown in FIG. 9, is that for any one axis, the distribution of photosensitive elements that contribute to each
• Cypress/SLM ref.: SLMP0352-PCT axis of motion is relatively sparse.
• a desirable feature of the 2D comb-arrays described previously is that by grouping sums of photosensitive elements appropriately, each element can be made to contribute to both axes of motion, without having to put a separate weighting coefficient on each - which is fairly expensive in terms of chip surface area and power consumption.
• each photosensitive element contributes to all three axes of motion, but there are still only a small number of weighting coefficients that are applied only after the output of groups or rows of elements are summed.
• this embodiment is the 3-axis analog of the 2D comb-array described above.
• FIG. 12 a schematic diagram is shown of a hexagonal array 1202, wired for 3 -axis motion detection.
• each hexagon represents a single photosensitive element 1204, such as a photodiode.
• Each photosensitive element 1204 in the array 1202 is couple to at least one signal line in each of three groups of signal lines 1206, 1208 and 1210 to detect motion in a direction perpendicular to the signal line.
• vertically oriented the signal lines 1206 are for detecting horizontal motion.
• each of the lines in the groups of signal lines 1206, 1208 and 1210 appear in either a solid or dashed pattern.
• the solid lines are main or in-phase signal lines for the group, while the dashed lines are for quadrature signal lines.
• the signs + and - indicate the weighting coefficient of +1 and-1 respectively.
• Cypress/SLM ref.: SLMP0352-PCT If a photosensitive element 1204 is crossed by a line that means that the element will contribute to a signal that is encoded by the line. In the embodiment shown, every photosensitive element 1204 is crossed by three lines associated Avith the three different groups of signal lines 1206, 1208 and 1210; that means that each element contributes to a signal for each of the three axes. For example, the photosensitive element 1204 at the very top of the hexagon array 1202 contributes to a main signal of the group of signal lines 1206 with a weighting coefficient -H, to the main signal of the group of signal lines 1208 with a weighting coefficient +1, and to the main signal of the group of signal lines 1210 with a weighting coefficient +1.
• the photosensitive element just below and to the right contributes to the quadrature signal of the group of signal lines 1206 with a weighting coefficient -1 , the quadrature signal of the group of signal lines 1208 with a weighting coefficient +1, and the main signal of the group of signal lines 1210 with a weighting coefficient +1. And so forth.
• the overall pattern is periodic; a heavy black line outlines a unit cell 1212 of the periodic pattern, and there are only 16 elements in the unit cell.
• Cypress/SLM ref.: SLMP0352:-PCT together giving 16 output signals.
• the signal from each flavor can be split three ways, the weighting coefficient appropriate to each of the 3 signals applied, and then combine the output signals into main and quadrature signals for each of the three
• the above embodiment enables the acquisition of 3 -axis information that can be combined to give resistance to fading in any single axis, and enable the use of each photosensitive element 1204 in all three axes, giving better SNR and better resistance to aliasing than arrays in previous speckle-based optical sensors.
• FIG. 13 shows a square, 2D array 1302 and wiring diagram for four axes of motion detection.
• the connection of photosensitive elements 1304 is similar to that describe with reference to FIG. 12, but there are now four directions, and each photosensitive element contributes to either main (in-phase) or quadrature signals in each of the four directions.
• a first group of signal lines 1306 couple to all photosensitive elements to detect motion in a horizontal direction.
• a second group of signal lines 1308 is connected to detect vertical movement, a third group of signal lines 1310 is connected to detect movement in a direction -45° from vertical, and a fourth group of signal lines 1312 is connected to detect movement in a direction +45° from vertical. It is noted that the signal lines in the groups of signal lines 1310 and 1312 are spaced more closely together than in groups of signal lines 1306 and
• Cypress/SLM ref. SLMP0352-PCT 1308. This is an indication that they are detecting a different, higher spatial frequency than the groups of signal lines 1306 and 1308.
• each of the lines in the groups of signal lines 1306, 1308, 1310 and 1312 appear in either a solid or dashed pattern.
• the solid lines are main or in-phase signal lines for the group, while the dashed lines are for quadrature signal lines.
• the signs + and - indicate the weighting coefficient of +1 and -1 respectively.
• a heavy black line outlines a unit cell 1314 of the periodic pattern, and there are 16 photosensitive elements in the unit cell. So there are basically 16 distinct "flavors" of photosensitive element 1304, each element characterized by the weighting coefficient that applies for each of the axes and whether it goes to main or quadrature for that axis. Again, the heavy black dots indicate photosensitive elements of the same flavor or making the same contribution to each axis.
• the signal from ea_ch flavor is split four ways and routed, with the appropriate weighting coefficient, to main and quadrature signals for each of the four axes.
• this concept can be generalized to any periodic array; by superimposing multiple periodic grids on the array. For example, one can extract multiple direction vectors, using all detectors in each calculation, without having to add individual weighting coefficients. One could also add more directions to this array based on other periodicities within the array; the number of flavors of cell climbs significantly, of course.
• the method employs a two-dimensional array of pixels or photosensitive elements connected in a fashion that enables simplified, signal processing of a given
• Various embodiments of the pixel-connection scheme can be implemented to allow processing of different (or multiple) spatial frequencies. This method allows ID speckle-based displacement measurements with lower power required from signal processing electronics than in 2D correlation type devices, a ⁇ id without compromising measurement accuracy as in prior devices using linear, ID comb-arrays.

Landscapes

• Engineering & Computer Science (AREA)
• General Engineering & Computer Science (AREA)
• Theoretical Computer Science (AREA)
• Human Computer Interaction (AREA)
• Physics & Mathematics (AREA)
• General Physics & Mathematics (AREA)
• Position Input By Displaying (AREA)
• Length Measuring Devices By Optical Means (AREA)
• Indicating Or Recording The Presence, Absence, Or Direction Of Movement (AREA)
• Optical Transform (AREA)

Priority Applications (3)

Applications Claiming Priority (4)

Publications (3)

Family

ID=36260724

Family Applications (1)

Country Status (4)

Cited By (3)

Families Citing this family (86)

Citations (3)

Family Cites Families (48)

• 2005
  • 2005-10-28 US US11/261,316 patent/US7138620B2/en not_active Expired - Lifetime
  • 2005-10-29 EP EP05851276.5A patent/EP1946353B1/en not_active Expired - Lifetime
  • 2005-10-29 JP JP2007539224A patent/JP2008519331A/ja active Pending
  • 2005-10-29 WO PCT/US2005/039189 patent/WO2006050229A2/en not_active Ceased
• 2006
  • 2006-10-18 US US11/582,776 patent/US7459671B2/en not_active Expired - Lifetime

Patent Citations (3)

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events