Data processing: measuring – calibrating – or testing – Measurement system – Orientation or position
Reexamination Certificate
2003-04-18
2004-09-07
Barlow, John (Department: 2863)
Data processing: measuring, calibrating, or testing
Measurement system
Orientation or position
C702S151000, C033S0010PT
Reexamination Certificate
active
06789042
ABSTRACT:
FIELD OF INVENTION
This invention relates to a position sensor for the sensing of the absolute position of a movable body without requiring counting from a reference mark. In particular the invention relates to a barcode system where a linear array of detectors is used to read and process the absolute position of a surface to which the barcode is attached to high accuracy. The barcode is coded with values that are used to find the absolute position of the barcode marking within the length of the barcode, and has the ability to be interpolated to provide a resolution of measurement better than that of the barcode marking (or spacing) pitch, or indeed better than that of the individual detector pitch of the linear array of detectors.
BACKGROUND
Conventionally, position sensors have been used for sensing the position of a body that is movable relative to the sensor. These sensors typically consist of a detector unit and a graduated scale of material with contrasting bars formed of alternating transparent and opaque bars, or bars of alternating high and low reflectivity, the displacement of which is detected by the detector unit.
The scale is typically illuminated by a source of electromagnetic radiation (EMR), typically UV, visible or IR light, that in turn generates an image in in the form of a pattern on one or more arrays of photodetectors sensitive to the EMR. Such arrays include CCD devices, VLSI vision chips, one and two dimensional photodetector arrays and lateral effect photodiodes (commonly referred to as PSD's or position sensitive devices). The output of the one or more arrays is processed to produce a measure of the position of the movable body.
Such sensors commonly provide a signal based on the incremental position of the scale, and absolute position is determined by counting from a known reference position. The accuracy of incremental sensors is often substantially improved by the use of well known techniques such as quadrature interpolation. Such techniques generally require a non-varying bar pitch.
Alternatively, the sensor may provide a signal based on absolute position by the use of barcodes applied to the scale. These barcodes generally do not have a constant bar pitch as each set of barcodes are unique for each position to be sensed. Such absolute position sensors generally do not provide the position measurement accuracy provided by incremental sensors as they cannot use the aforementioned quadrature interpolation techniques.
If an absolute position sensor is required to have high accuracy, two separate scales and arrays of detectors are generally required. The first measures coarse absolute position by interrogation of a barcode, and the second provides a fine relative position by quadrature interpolation of a constant bar pitch pattern.
On the other hand International Patent Application No. PCT/AU99/00590 discloses the use of a single scale for the measurement of coarse absolute position and also for the fine relative position by quadrature interpolation of a regular bar pattern. The angle encoder sensor disclosed in this specification is composed of a pseudo-random bar code scale of constant bar pitch and a varying bar width or, alternatively, special forms of bar codes with varying bar pitch for coarse absolute positioning. This arrangement however has a number of disadvantages. The use of a barcode consisting of varying bar widths for coarse absolute position measurement is known to have difficulties in image processing due to the width interpretation of the barcodes due to imperfections of the individual pixels that make up the detectors in the array, and the variations in the quality of the barcode markings. Also, as the sensor requires a finite time to integrate an image, movement of the barcode relative to the sensor during this time period produces smearing of the image, further degrading the quality of the image to be processed. A further problem is the variation of the image signal produced on the array. This can be due to non-homogeneous illumination, non-homogenous surface properties, or due to the combination of the positioning of all the components that transmit, reflect or repropogate, and collect the EMR. All these effect the accuracy to which the image processing can identify the width of the individual bars of a barcode.
For the barcode of constant width but varying pitch, the image processing requires an image threshold level to be used. The setting of a single image threshold level for the entire array for the processing of the barcode has several disadvantages. The evaluation of the binary states of all of the bits incident on the array by the use of a single image threshold level is susceptible to error because the image levels on the array are due, not only to the barcode pattern, but also the effects of non-homogeneous illumination, non-homogenous surface properties, and the combination of the positioning of all the components that transmit, reflect or repropogate, and collect the EMR.
SUMMARY OF INVENTION
The present invention consists of a position sensor comprising a body at least partially surrounded by a housing, the body having a grating element attached thereto or integral therewith, the grating element comprising a surface, the surface comprising a coded distribution of regions of high and low reflectivity or transmissibility, the sensor also comprising at least one EMR source and at least one array of EMR sensitive detectors, the source irradiating the surface and the array receiving incident EMR reflected from or transmitted through the surface, the source and the array fixed with respect to the housing, a pattern thereby produced by incident EMR on the array resulting from the regions of high and low reflectivity or transmissibility on the surface of the grating element, the coded distribution having a sequence of bits, each bit comprising only one region of high reflectivity or transmissibility and only one region of low reflectivity or transmissibility, each bit representing a binary state 1 or a binary state 0 depending on whether it respectively corresponds to a transition from regions of high to low, or alternatively a transition from regions of low to high, reflectivity or transmissibility on the surface of the grating element, the bits being arranged at a substantially uniform code pitch, the incident EMR on the array therefore determining a substantially spatially periodic intensity pattern of incident EMR on the array, characterised in that the coded distribution comprises over 50% of dominant bits of either a binary state 1 or 0, and less than 50% of regressive bits, the dominant and regressive bits arranged consecutively as a pseudo-random binary code, the pseudo-random binary code sampled as a series of n consecutive bits and also determining the substantially spatially periodic intensity pattern incident on the array, the coded distribution arranged such that any said series of n consecutive bits also comprises over 50% dominant bits and less than 50% regressive bits, the pattern on the array processed by a processor to derive the absolute position of the coded distribution with respect to the housing, and hence provide a measure of the absolute position of the body with respect to the housing.
Preferably the substantially spatially periodic intensity pattern incident on the array is interpolated by Fourier analysis by the processor.
Preferably each sample of n consecutive bits is unique over the range of absolute position of the body with respect to the housing.
Preferably the transition from regions of high to low, or alternatively a transition from regions of low to high, reflectivity or transmissibility on the surface of the grating element is arranged in the form of a Manchester code.
Preferably the substantially spatially periodic intensity pattern incident on the array is processed to remove at least one regressive bit before being interpolated by Fourier analysis by the processor to further increase the accuracy of measurement of absolute position of the body with respect to the housing.
Prefera
Bishop Innovation Limited
Cherry Stephen
Rothwell Figg Ernst & Manbeck p.c.
LandOfFree
Position encoder using statistically biased pseudorandom... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Position encoder using statistically biased pseudorandom..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Position encoder using statistically biased pseudorandom... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3228558