High-resolution absolute encoder

Abstract

A high-resolution encoder device for measuring and retrievably encoding a relative position of a first portion P and a second portion S comprises a sensing element of portion P, the sensing element presenting a variable characteristic for sensing, the variable characteristic varying over a length of portion P, n sensors for sensing the variable characteristic, the sensors being separated from one another and positioned in portion S, the sensors being configured to output a signal in response to the sensing, and a processing device connected to receive the signal from each of the n sensors and configured to hand over to ones of the relative positions forming vectors respectively having an entry from each sensor, the vectors defining the relative positions respectively.

Description

[Technical Field]

[0001] Cross-reference of related applications This application claims the benefit of priority under § 119(e) of U.S. Provisional Patent Application No. 63 / 145,002, filed on 3 February 2021, which is incorporated herein by reference in its entirety. [Background technology]

[0002] The present invention relates to an encoder device (encoder), and more specifically, but not limited to, a high-resolution absolute vector encoder.

[0003] Encoders are used to measure the angular position of a rotating element or the relative displacement of a sliding element. They are typically used in control systems, often called servo systems, where a motion controller is used to guide the moving element along a precise, desired path. For this purpose, encoder devices include an electronic interface that allows connection to the motion controller.

[0004] Encoders can come in two types: rotary and linear. Rotary encoders are designed to measure the angular position of a rotating element, such as a motor or the shaft of any rotating device. Linear encoders are designed to measure the relative motion of two sliding elements (for example, a sliding carriage mounted on a linear bearing relative to a stationary base).

[0005] In typical applications, a rotary encoder is mounted on the rear end of an electric motor shaft, providing positional information regarding the shaft's rotation angle to an electric motor controller. This positional information can be provided at high speed. The motor controller can then output current to the motor to rotate it toward the desired position.

[0006] In another common application, a linear encoder is mounted on the moving element of a linear motor and connected to the linear motor's motion controller.

[0007] Within the scope of this patent application, the term “encoder device” is used to refer to either a rotary encoder or a linear encoder. A linear encoder is constructed using the same components as a rotary encoder, by arranging these components along a straight path. The straight path is divided into a number of consecutive sections of equal length, referred herein to as periods. In a first implementation, the encoder components are arranged along one period in a manner similar to arranging these components along a circular path. In other implementations, the components can be arranged over several periods, provided that the distance of each component from the same position as the components in the first implementation is equal to an integer of the period.

[0008] In automated machinery, moving elements are often required to follow paths with extremely high precision and speed. To achieve this, encoder devices must be designed with high precision and capable of transferring positional information at high speed. For example, commercially available rotary encoders can achieve accuracy exceeding 0.01 degrees, and the transfer rate of rotation angle values ​​to the motion controller is typically between 8,000 and 30,000 values ​​per second.

[0009] Another parameter for designing an encoder device is its resolution. Resolution represents the smallest change in position that an encoder device can measure in one revolution or one unit of length, and is usually expressed as the number of position values ​​or linear distance per revolution. The smallest change in position defined by the resolution is usually smaller than the accuracy; that is, even if the output position value differs slightly from the actual position, the encoder device can produce a position value with more significant digits than the number of significant digits required for accuracy, and this error is smaller than the error defined by the accuracy characteristics of the encoder. High resolution allows motion controllers, also called servo controllers, to achieve precise and smooth control of moving elements.

[0010] Encoder devices can be absolute or incremental. Absolute encoders can measure angular or linear positions relative to a fixed reference position, while incremental encoders can only measure angular or linear movement from the start of motion. Therefore, when using incremental encoders in automated machinery, it is common to perform a reference position findout at the start of each machine operation. This findout is performed slowly in a specific direction until a limit switch or other device located at the reference position is activated. This findout procedure adds complexity to the system and slows the initial operation of the machine. Despite this drawback, incremental encoders are commonly used due to their simplicity and low cost. While machine manufacturers often prefer absolute encoders, they opt for incremental encoders because currently available absolute encoders are expensive.

[0011] It is desirable to provide an absolute encoder device that is easy to manufacture, low-cost, and delivers high accuracy and resolution.

[0012] Patent US9007057 by Villaret, issued on April 14, 2015, describes a simple absolute encoder device capable of providing high-resolution absolute position information. The device utilizes n analog sensors evenly spaced around a circumference. In the static part, a rotating disk is arranged on an annular track with alternating characteristic sections according to a specific pattern, allowing the sensors to sense the characteristics of adjacent track sections. As the disk rotates, different parts of the rotating disk come into contact with each sensor. The electrical signal from each sensor is first digitized and assigned a bit value of 1 or 0. Next, the bit values ​​of all sensors are combined into a digital word to create a unique code value for each angular range position of the rotating disk. In the second step, one of the n sensors is selected, and its analog output value is used to calculate a high-resolution position value.

[0013] The advantage of the above patent is the simplicity of the device. Because the sensors are evenly distributed along a circular line, the distance between sensors is relatively wide, and commercially available, standard-sized sensors can be used.

[0014] In the aforementioned patent, the term "sector" is defined as a corner portion of the circular track of the encoder rotating disk, and all sectors have approximately equal angular size. Each sector of the track is made of a material having a first or second characteristic according to a predefined pattern.

[0015] The first requirement of the aforementioned patent is that the code is monovalent, that is, the value of the code can only be obtained within the range of a single sector.

[0016] The second requirement is that the code should be Gray code. That is, during transit, only one bit of the digital code changes when moving from one sector to an adjacent sector. This is necessary to avoid errors in the code during transit.

[0017] Both of the above requirements can be achieved using a pattern designed according to the method described in Villaret's patent US8492704, granted on July 23, 2013.

[0018] These two requirements result in limitations in the actual implementation, as will be explained below.

[0019] According to US9007057, the achievable overall resolution of an encoder is approximately equal to the product of the resolution of the digital code and the resolution of the analog sensor signal.

[0020] To increase the resolution, it is necessary to increase either a) the resolution of the analog sensor reading or b) the resolution of the digital code.

[0021] Regarding a), the practical resolution of standard analog sensors and analog-to-digital converters is limited by electrical noise. In particular, a high-speed analog-to-digital converter is required because a high-resolution encoder is also needed to provide position values within a very short time. Therefore, increasing the resolution is very difficult and unrealistic.

[0022] Regarding b), according to Villaret's patent US8492704, an increase in the resolution of the digital code can be achieved by increasing the number of sensors and changing the pattern. However, increasing the number of sensors n results in a very inefficient resolution of the resulting code. That is, the number of codes is the number of possible code values 2 nIt becomes smaller than. For example, when using a sensor with n = 7, a practical pattern that results in 98 code values can be found. When using a sensor with n = 8, in a practical pattern, only 128 code values can be obtained. Therefore, even if the number of sensors increases, the improvement in resolution is slight. In another aspect, in the example of a sensor with n = 7, a practical pattern that defines a section spanning a relatively large number of sectors is found and can be easily implemented using relatively large-sized elements, such as magnets. As the number of sensors increases, the size of the pattern section becomes significantly smaller, making it unrealistic for implementation and requiring small components, which may be unrealistic especially in the case of magnets.

[0023] Therefore, in order to achieve further higher resolution, it is desirable to design an encoder without the above-mentioned constraints in pattern design.

Summary of the Invention

[0024] Therefore, an object of the present invention is to provide an encoder device that can provide higher resolution.

[0025] Therefore, a main object of the present invention is to provide a high-resolution encoder device having sensors that are suitably and sufficiently separated from each other so that preferably commercially available sensors can be used. The encoder device according to the present embodiment can utilize sensors that present analog outputs and can include memory and processing means to obtain a high absolute resolution that is not limited by the number of sensors.

[0026] The encoder according to the present embodiment can include two parts, and the two parts are relatively movable along a circular path in the case of a rotary encoder or along a linear path in the case of a linear encoder. Here, it should be understood that "linear" includes any curved shape and is not limited to a linear straight line. Further, the encoder includes a processing device that can receive a large number of analog signal inputs and output position values.

[0027] For the purposes of this document, the two encoder sections will be referred to as section P and section S for ease of understanding.

[0028] According to some embodiments of the present invention, a high-resolution encoder device is provided for measuring and retrievingly encoding the relative positions of a first portion P and a second portion S. a) A sensing element of a portion P, wherein the sensing element presents a variable characteristic for sensing, and the variable characteristic changes over the length of the portion P, b) n sensors for sensing variable characteristics, wherein the sensors are separated from each other and arranged in part S, and the sensors are configured to output a signal in response to sensing, c) A processing unit connected to each of n sensors and configured to receive signals from each of the relative positions, forming a vector having entries from each sensor, thereby defining the relative position of each, wherein the processing unit is configured to calculate the relative position from the vector.

[0029] The encoder may further include an analog-to-digital converter between the n sensors and the processor to digitize the signals from the n sensors before presenting the signals to the processor.

[0030] In this embodiment, the signal is

number

[0031] In this embodiment, the vector is an n-vector function

number

[0032] In this embodiment, the vector function V(X) is selected to satisfy the first and second mathematical conditions.

[0033] The first mathematical condition is that there is a one-to-one (injective) relationship between X and V(X), so the vector value V(X) obtained at position X represents a characteristic of that position.

[0034] Second mathematical condition: For any two positions X1 and X2, the norm

number

number

number

number

number

[0035] In this embodiment, the vector function V(X) has size p,

number

number

[0036] In this embodiment, the variable characteristics are Variable magnetic field strength resulting from permanent magnets of different polarities placed at several positions on the scale. Variable light intensity obtained by the variable transparency or reflectivity of various parts of the scale and numerous light sources, Variable light intensity obtained by the variable distance between the sensor and multiple light sources, Variable eddy current losses obtained by the irregular shape of the conductive part of the scale, and It is a member of the group consisting of variable inductance obtained by the variable magnetic properties of the rotor material.

[0037] One embodiment is a rotary encoder, where relative position includes relative angle.

[0038] Another embodiment is a linear encoder. One of its parts includes a straight or curved path, and its relative position is related to the length along the straight or curved path.

[0039] In one embodiment, the processor is configured to obtain an initial position estimate from the most recent known position.

[0040] The encoder can obtain the initial position by scanning all possible positions until the vector function satisfies the following conditions.

number

[0041] In this embodiment, the values ​​of the vector function are stored in a lookup table.

[0042] According to a second aspect of the present invention, a high-resolution encoding method for measuring and reproducibly encoding the relative positions of a first portion P and a second portion S, a) A variable characteristic for sensing is presented, the variable characteristic changes over the length of a portion P, and the sensing characteristic is presented via a sensing element. b) In part S, the variable characteristics are sensed at n discrete positions. c) Outputting signals from n discrete locations in accordance with sensing, d) Receiving signals from each of n discrete locations, and e) A high-resolution coding method is provided, which involves each sensor having an entry in a vector that forms a relative position, thereby defining the relative position, and the encoder then using the vector to measure and encode the relative position.

[0043] The embodiment may include digitizing the signals from n sensors before presenting the signals to the processor.

[0044] In this embodiment, the signal is

number

[0045] The embodiment uses the n vector function of the relative position X between the P part and the S part to determine each vector.

number

[0046] In this embodiment, the vector function V(X) is selected such that it satisfies two mathematical conditions.

[0047] The first mathematical condition is that there is a one-to-one (injective) relationship between X and V(X), so the vector value V(X) obtained at position X represents a characteristic of that position.

[0048] Second mathematical condition: For any two positions X1 and X2, the norm is as follows:

number

number

number

number

number

[0049] In this embodiment, the vector function V(X) has size p,

number

number

[0050] In this embodiment, the variable characteristics are as follows: Variable magnetic field strength resulting from permanent magnets of different polarities placed at several positions on the scale. Variable light intensity obtained by the variable transparency or reflectivity of various parts of the scale and numerous light sources, Variable light intensity obtained by the variable distance between the sensor and multiple light sources, variable eddy current loss obtained by the irregular shape of the conductive part of the scale, and This is one type of variable inductance obtained by the variable properties of the rotor material's magnetic properties.

[0051] The embodiment may include performing a function search of a vector function, and the function search is Select the location of part P, and place or simulate a sensing element at the location of part P. Select the location of part S, and place or simulate a sensor at the location of part S. Calculating or presenting signals from each sensor, and Test the signal and present candidate vector functions that satisfy the conditions, and This includes accepting a candidate vector function if the condition is met, and repeating the function search otherwise.

[0052] The relative position, in the case of a rotor, includes the relative angle. Alternatively, one of the parts includes a straight or curved path, and the relative position relates to the length along that path.

[0053] In embodiments, the processor may include obtaining an initial position estimate from the most recent known position.

[0054] The embodiment may include obtaining an initial position by scanning all possible positions until the vector function satisfies the following conditions:

number

[0055] The embodiment may include obtaining the value of a vector function from a lookup table.

[0056] Embodiments may include performing a start procedure, the initial start procedure including compiling a vector function from variable characteristics.

[0057] The starting procedure is further as follows: Selecting candidate characteristics and variations of variable characteristics, Setting a vector function using simulations of candidate characteristics and variations, and This may include verifying that the vector function satisfies the first and second conditions.

[0058] Unless otherwise defined, all technical and / or scientific terms used herein have the same meaning as those generally understood by those skilled in the art to which the present invention pertains. Similar or equivalent methods and materials may be used in the practice or testing of embodiments of the present invention, but exemplary methods and / or materials are described below. In case of any conflict, the patent specification, including definitions, shall prevail. Furthermore, the materials, methods, and examples are merely illustrative and not necessarily intended to be limiting.

[0059] Several embodiments of the present invention will be described herein, merely as examples, with reference to the accompanying drawings. While the drawings will be given in detail, it is emphasized that the illustrated details are intended to illustrate embodiments of the present invention. In this regard, the description with reference to the drawings will make it clear to those skilled in the art how embodiments of the present invention can be put into practice. [Brief explanation of the drawing]

[0060] [Figure 1] This is a simplified diagram showing a one-turn absolute encoder according to an embodiment of the present invention. [Figure 2a] This is a simplified diagram illustrating an example of variations of an analog signal as a function of rotation angle according to an embodiment of the present invention. [Figure 2b] All analog signals according to embodiments of the present invention are shown on the same plot. [Figure 3a] This shows variations of the norm of the difference N[V(X1)-V(X2)] for all values ​​of the fixed values ​​X1 and X2 according to embodiments of the present invention. [Figure 3b] This is a magnified view of the area around the fixed value of X1. [Figure 4] The norm and function F of the difference N[V(X1)-V(X2)] with respect to a fixed value X1 and X2 around X1 (55 degrees) are shown according to embodiments of the present invention. [Figure 5]This is a simplified block diagram showing a flowchart of a processing unit that implements the encoder algorithm according to an embodiment of the present invention. [Figure 6] A schematic diagram of a linear encoder according to an embodiment of the present invention is shown. [Figure 7] Figures a and b are simplified diagrams showing the mounting configuration of a magnetic encoder according to an embodiment of the present invention. [Modes for carrying out the invention]

[0061] This embodiment relates to a high-resolution absolute vector encoder.

[0062] The encoder according to this embodiment is provided for measuring the relative position X of two parts, part P and part S. Depending on the particular application, part P may be stationary, and part S may be stationary or moving along a path. Part P moves along a path.

[0063] A pattern function is defined and placed on the track of the P portion of the encoder. This track is linear in the case of a linear encoder, and circular in the case of a rotary encoder, and the characteristics of the track change according to the pattern defined by the pattern function. This pattern is such that it results in continuously variable characteristics along the track. The characteristics that change along the position on the track can be any one of many suitable characteristics that can be sensed by the appropriate sensor. For example, a) Variable magnetic polarization that can be detected by a Hall sensor or magnetic sensor, b) Variable transparency, reflectance, or light polarization that can be detected by an optical sensor. c) Variable conductivity detectable by an eddy current sensor, d) Inductance as perceived by electronic devices.

[0064] This enumeration is non-limiting, and other combinations of characteristics and sensors can be conceivable by those skilled in the art and should be considered within the scope of this patent.

[0065] To sense the characteristics of the track, a number of sensors are arranged at predetermined positions close to the track along the track of the encoder part S. During the relative movement of the two parts P and S along the path, the sensors move along and close to the patterned track. Each sensor outputs an analog signal that varies according to the characteristics of the adjacent track part and according to the relative position X of the two encoder parts. Each sensor output is amplified and / or conditioned by an electronic circuit and input to the processing device as an analog value A s (X).

[0066] Therefore, at each position X, s values are available, and the number s of the values of A s (X) gives a vector of dimension s

Number

[0067] According to this embodiment, the pattern and position of the S sensors are designed to give the following mathematical characteristics of the vector V. Condition a): Since there is a one-to-one (injective) relationship between X and V(X), the vector value V(X) obtained at the position X is the characteristic of that position. Condition b): For any two positions X1 and X2, there exists a threshold Δ such that the norm

Number

Number

Number

Number

number

[0068] Before the first operation, the encoder undergoes a calibration procedure, during which the value V(X) is converted to vector V c (X i This is recorded in the table and stored in the memory of the processing unit.

[0069] Based on the two characteristics described above, an algorithm is executed in the processing unit to evaluate the relative position X0 of P with respect to S.

[0070] At position X0, the value of the vector is

number

[0071] The search algorithm is created in two steps.

[0072] Step 1: Find position X as follows:

number

[0073] To perform this search, you need to use a previously recorded V c A table of (X) values ​​is used. Various search methods can be used to optimize this step.

[0074] Step 2: Evaluate

number

[0075] Step 3:

number

number

[0076] Step 3 is carried out according to condition b)_ defined above,

number

number

[0077] The acquired value of X1 represents the scale of the actual position X at this time and is output by the processing unit.

[0078] This algorithm is executed in the processing unit and, as will be described later, can be optimized to run at high speed and periodically, so that the encoder of this embodiment can output the measurement position at high speed. For example, using currently available microprocessors, a speed of 30,000 measurement positions per second can be achieved.

[0079] The first advantage of the encoder of the present invention is its simplicity. Because the encoder can be designed so that the sensors are located relatively far apart from each other, it can use low-cost, standard discrete electronic components.

[0080] The second advantage is that high precision is not required during the manufacturing process. In contrast, in the prior art cited above, the precision may be such that a predefined code is obtained, and the range of positions for each code may be approximately the same for each code in order to maintain the required precision or resolution.

[0081] A third advantage is that patterns can be designed using numerous elements of variable characteristics. Analog signals may exhibit rapidly increasing or decreasing sections, which are advantageous for high-resolution position measurements.

[0082] Before describing in detail at least one embodiment of the present invention, it should be understood that the present invention is not necessarily limited in its application to the structural details and arrangement of components and / or methods described in the following description and / or shown in the drawings and / or examples. Other embodiments of the present invention are possible, or it can be practiced or implemented in various ways.

[0083] Here, we refer to Figure 1, a simplified diagram of an embodiment of the encoder according to the present invention. The encoder rotor shown in Figure 1 as a disk 101 fixed to a rotating shaft 102 includes a track 103 having variable characteristics on its outer circumference. Here, the variable characteristics of the track are shown as the shape of a black ring with a variable radius width, indicated by arrow 104. The variable characteristics do not necessarily have to be annular in width, but are shown as such in this specification for simplification to more clearly show a characteristic that changes continuously as a function of the rotation angle of the disk.

[0084] Variable characteristics can take various forms, such as the following: • Variable magnetic field strength generated by permanent magnets of different polarities placed in multiple positions on the rotor. • Variable light intensity obtained by the variable transparency or reflectivity of various parts of the rotor and numerous light sources. • Variable light intensity obtained by the variable distance between the sensor and multiple light sources. • Variable eddy current losses obtained by the irregular shape of the rotating rotor and numerous high-frequency electric coils. • Variable inductance obtained by the variable magnetic properties of the rotor material.

[0085] The above enumeration is not limiting, and other types of characteristics and sensors may also be conceived within the scope of this patent.

[0086] The eight sensors 103a to 105h are static and positioned to sense the variable characteristics of the rotating disk 101. Each of the sensors 103a to 105h outputs an electrical analog signal A1 to A8. The sensor output changes continuously according to the rotation angle in accordance with the variable characteristics.

[0087] Analog signals A1 to A8 are input to an analog-to-digital converter 105, and their analog values ​​are then input to a processing unit (CPU, 106). The CPU 106 then periodically processes the values ​​of the signal values ​​A1 to A8 in a manner further described herein and outputs high-resolution position values.

[0088] Depending on the type of characteristics used, any specific geometric configuration of the encoder can be used, not limited to the configuration with adjacent disks and sensors used herein. Within the scope of this patent, any configuration can be used in which a number of sensors, each sensing the variable characteristics of the rotating part (rotor) of the encoder, output an analog electrical signal that changes according to the rotation angle of the shaft.

[0089] According to the encoder of this embodiment, signals A1 to A8,

number

[0090] Condition a) Since there is a one-to-one (injective) relationship between X and V(X), the vector value V(X) obtained at position X represents a characteristic of that position.

[0091] Condition b) For any two positions X1 and X2, the norm

number

number

number

number

number

[0092] The following refers to Figures 2A and 2B. These figures are two plots of typical sensor inputs according to this embodiment. Figure 2a shows an example of a single sensor signal input to the CPU after amplification and analog-to-digital conversion, while Figure 2b shows eight signals output by eight sensors on the same plot.

[0093] Before initial operation, the encoder of this embodiment undergoes a calibration procedure. During this calibration procedure, the vector V of the analog signal An is recorded and stored in a table in CPU memory with sufficient resolution over the entire range of positions over one rotation. The vector values ​​obtained from these tables in CPU memory are further V below. c It is called a vector derivative.

number

[0094] In the embodiments described, the selected norm N is the square root of the square of the signal difference.

number

[0095] Now, let's refer to Figure 3a, which shows the plot of norms against angular position. More specifically, the norms for a given position X1 = 55 degrees and position X2 across the entire range of rotation (360 degrees).

number

number

number

[0096] For such initial searches, a simple search procedure can scan the entire range in steps smaller than the minimum width of a segment, such as segment 301. However, since such a procedure is time-consuming, an alternative search procedure that can be performed quickly is described below.

[0097] Considering condition b), in this embodiment, the function F is selected and defined by scalar vector multiplication.

number

number

[0098] Referring now to Figure 4, this figure shows an exemplary plot illustrating variations of function F around position X1 = 55 degrees (406). In this graph, function F(403) is norm

number

[0099] value

number

[0100] For the specific F-function selected in this embodiment, when X=X1, the F-function reaches a null (0) value.

[0101] As a result of the initial search step, we obtained the first position X adjacent to X1, so we can write this using a first-order approximation as follows:

number

[0102] This equation

number

number

[0103] This calculation is repeated several times, each time using an estimated value X1 for X. These iterations converge rapidly, and after several iterations, a very accurate estimate of the actual position X1 is obtained.

[0104] This iterative algorithm can be executed quickly and periodically, allowing the CPU to output accurate, high-resolution position values ​​at high speed.

[0105] The maximum achievable precision δX is determined by the noise and the resolution of the analog-to-digital converter, affecting the signal A s Possible error δA s , and vector derivative

number

[0106] In the case of a vector error δV due to noise and resolution, the resulting position error is:

number

[0107] Therefore, to reduce positional errors, the encoder should be set to the highest possible value.

number

[0108] Other iterative algorithms are functions that are monotonic at least in the range X1–X.

number

number

[0109] The continuation of the algorithm executed on the CPU to accurately estimate the angular position of the encoder was described above. The continuation is described below, first

number

[0110] There are two cases to consider.

[0111] In the first case, for example, after powering on the encoder, the approximate position of the encoder is unknown. In this case, a startup procedure is defined, and the CPU program scans the entire range of angular positions in steps smaller than the minimum width of the segment. For example, segment 301 in Figures 2a and 2b. Such a procedure is time-consuming, but it is part of the encoder initialization procedure and does not affect the encoder's functionality.

[0112] In the second case, the first accurate estimate of the encoder angular position X can be obtained from the previous cycle. Because the encoder has a limited rotational speed C and the CPU algorithm runs in a very short cycle time, the actual angular position X1 of the encoder rotor will be very close to the previously measured position X. The function in the range (X - X1)

number

[0113] For example, the maximum speed of the encoder is C = 12,000 rpm, and the cycle time is dt = 30 microseconds. In this example, the angular position of the encoder can change by C.dt = 2.16 degrees within one cycle. Refer to Figure 4, the function

number

[0114] Furthermore, by adding an increment proportional to the velocity, a position X closer to the actual position of the encoder can be calculated. Because the maximum acceleration is limited in the physical system, extrapolating the velocity to the previously calculated position yields a starting position X closer to the algorithm in the second step, and increases the maximum velocity that can be applied to the encoder.

[0115] Referring to Figure 5, the simplified flowchart illustrates the encoder algorithm described above. The blocks in the flowchart are numbered from 1 to 11, and the processing of the CPU program represented by each block is explained below.

[0116] Block 1: When the encoder is powered on, the encoder processing unit starts from Block 1 and proceeds to Block 2.

[0117] Block 2: • The previous position variable PX is defined and labeled "Unknown". The variable C is defined for the angular velocity of the encoder and is set to zero, C=0. Proceed to Block 3.

[0118] Decision Block 3: • If the previous position PX is unknown, proceed to the startup procedure in Block 4. If the previous position PX is known, proceed to block 5.

[0119] Block 4: ·conditions

number

[0120] Blocks 5-11 describe the algorithm cycle, as indicated by the dashed contour line 501. The CPU program executes the cycle calculations rapidly and periodically.

[0121] Block 5: The algorithm's cycle is not the first one; the previous position is known, and that known position can be used to make an initial rough estimate of the current position. The velocity estimate C is also taken from the previous cycle.

number

[0122] Block 6: A more accurate estimate of the position, X1, is calculated using Equation 2 above.

number

[0123] Decision Block 7: To check if the estimated position X1 is at the required resolution, the difference...

number

[0124] The value ε is predefined and set to a value smaller than the required resolution of the encoder.

[0125] Alternatively,

number

[0126] The value δ is predefined to approximate the estimated noise level and the resolution of the analog-to-digital converter.

[0127] If δ is less than a predetermined value ε, proceed to block 9, and the value X1 is output by the encoder. Otherwise, proceed to block 8, where the value X1 is used as a new starting point for another iteration.

[0128] Block 8 The value X1 is copied to X as the new starting point for the next iteration. Thus, each iteration passes through blocks 6, 7, and 8 until the condition in block 7 is verified. The number of iterations is usually less than 5. Proceed to Block 6

[0129] Block 9 • The position value X1 is output by the encoder. Proceed to Block 10

[0130] Block 10 Velocity is calculated from the difference between the current position X1 and the previous position PX. Here, dt is the cycle time, i.e., the time difference between the calculations of PX and X. Proceed to Block 11

[0131] Block 11: • Position for the next cycle

number

[0132] As summarized below, various search algorithms can be conceivable in accordance with the principles of this disclosure.

[0133] a) When starting (power on), in the first algorithm cycle, condition

number

[0134] The algorithm described above relies on the two essential conditions for the variable characteristics mentioned earlier.

[0135] Condition a) Since there is a one-to-one (injective) relationship between X and V(X), the vector value V(X) obtained at position X represents a characteristic of that position. Condition b) For any two positions X1 and X2, the norm

number

number

number

number

[0137] In order to design a physical arrangement that produces such variable characteristics, for example, a computer can be used to construct a variable characteristic search method.

[0138] In a computer program, a norm and a function F are defined. A threshold Δ is defined to be significantly higher than the standard of the maximum noise level of an analog signal.

[0139] A variable maxdVcdt is defined and initialized to zero.

[0140] The search for candidate variable characteristics that satisfy conditions a and b can be executed as follows.

[0141] 1: In a simulation model, according to random dimensions or the number of other parameters, arrange a large number of active elements in part P of the encoder and sensors in part S of the encoder.

[0142] 2: Use a simulation program to calculate the resulting signal A s (X) for each sensor and the vector of all positions X [Number] as defined above, the encoder includes two parts P and S. Part P induces variable characteristics, and those characteristics change according to the relative movement of P and S. In order to generate variable characteristics, sensing elements can be arranged in part P according to a geometric pattern. Part S includes sensors arranged relatively far from this variable characteristic. Therefore, a sensor output signal that changes according to the relative position between P and S is obtained due to the variable characteristic.

[0143] 3: Function​​​​

Number

[0144] 4: Check whether condition b is verified. If not, return to 1.

[0145] 5:

Number

Number

Number

Number

[0146] The search can [[ID=�0]]

Number

[0147] In one of the above search methods, a random selection of several dimensions or parameters is used. Other methods can also be used to create a new variable characteristic function each time. For example, artificial intelligence and neural network technologies can be used more efficiently.

[0148] In the above embodiment, the vector V(X) is the vector

Number

[0149]

[0150] For example, by adding the signals of two sensors located symmetrically in the S part of a rotary encoder into one signal, the eccentricity of the shaft rotation can be corrected. [[ID=二十九]]<0000九七四>[Number]

[0151] The vector [[ID=4十一]]<0000九八二> can be designed by the above search method to verify the same characteristics a and b.

[0152] Referring to FIG. 6, a linear encoder according to this embodiment is shown. The encoder includes a linear scale 601 that constitutes the P part of the encoder. In the linear scale, a variable characteristic 602 is induced, which is variable along the length of the scale and is represented here as a black band of variable thickness. There can be many types of variable characteristics, such as the following as examples. • Variable magnetic field strength generated by permanent magnets of different polarities placed at multiple positions on a scale. • Variable light intensity obtained by the variable transparency or reflectivity of various parts of the scale and numerous light sources. • Variable light intensity obtained by the variable distance between the sensor and multiple light sources. • Irregular shape of the conductive part of the scale and variable eddy current loss obtained by eddy current sensors, • Variable inductance obtained by the variable magnetic properties of the rotor material.

[0153] The above enumeration is not limiting, and other types of characteristics and sensors may also be conceivable within the scope of this patent.

[0154] The encoder head 604 can constitute the S portion of the encoder. The encoder head and the scale move relative to each other in a straight line. The encoder head contains eight sensors (603a to 603h). Each sensor senses the variable characteristics of the scale depending on the position of the head and the position of the head relative to the scale.

[0155] Next, the electrical signals from these sensors are amplified and digitized by an A / D converter (605) and transmitted to the processing unit 606 (CPU).

[0156] The variable characteristic search method described above is used to design the variable characteristics of the encoder, and the encoder algorithm described above is implemented by the processing unit.

[0157] Therefore, absolute linear encoders can be designed with a simple physical design.

[0158] Referring now to Figures 7a and 7b, the implementation configuration of the rotary encoder is shown. Figure 7a shows an axial cross-section. The encoder includes a shaft 703 and a disk 702 fixed to this shaft and rotatable inside the fixed encoder body 701. Numerous permanent magnets 704a to 704j are attached to the disk 702 shown in Figure 2b at various positions. Referring again to Figure 7a, a printed circuit board (PCB) 706 fixed to the encoder body includes magnetic sensors such as 705a, 705e, etc. In this particular example, there are eight magnetic sensors 705a to 705h. The axial positions for 705a and 705e are shown in Figure 7a, and the radial positions for 705a to 705h are shown in Figure 7b. The magnetic sensors 705a to 705h are positioned axially close to the permanent magnets 704a to 704j. Due to this axial distance, the magnetic field induced in these sensors continuously changes with the rotation of the disk. The method for finding the above variable characteristics involves designing the size and position of the magnet to induce a variable magnetic field in the magnet and generating a magnetic sensor that has an electrical signal satisfying the above requirements a and b.

[0159] In this particular example, thirteen permanent magnets 704a to 704j are arranged around the circumference of the disk. Therefore, the random sizing and positioning of these permanent magnets implies 13 parameters. The variable characteristic search method described above can be performed periodically, and in each cycle, values ​​for the 13 parameters of permanent magnet position and size can be selected, and a simulation program can be run to analyze the sensor signal.

[0160] PCB 706 is fitted with an analog-to-digital converter and processing unit (not shown) for executing an encoder algorithm and outputting angular position values ​​via wire 707.

[0161] Therefore, it is possible to manufacture high-resolution absolute magnetic encoders with a very simple configuration at a low cost.

[0162] In this document, the terms "comprises," "comprising," "includes," "including," and "having," as well as their cognates, all mean "including but not limited to."

[0163] The phrase "consisting of" means "to include (or be equipped with) and be limited to."

[0164] As used herein, unless otherwise clearly indicated by the context, the singular forms “a,” “an,” and “the” refer to multiple objects.

[0165] Certain features of the Invention described in the context of separate embodiments for clarity may also be provided in combination in a single embodiment, and it is understood that this text should be interpreted as if such a single embodiment were clearly described in detail. Conversely, various features of the Invention described in the context of a single embodiment for brevity may also be provided separately or in any suitable secondary combination, or as appropriate for any other described embodiment of the Invention, and this text should be interpreted as if such separate embodiments or secondary combinations were clearly described in detail herein.

[0166] Certain features described in the context of various embodiments should not be considered essential features of those embodiments unless the embodiments would not function without those elements.

[0167] Although the present invention has been described in relation to its specific embodiments, it is obvious that many alternative, modified, and variant forms will be apparent to those skilled in the art. Therefore, it is intended to encompass all such alternative, modified, and variant forms that fall within the spirit and broad scope of the appended claims.

[0168] All publications, patents, and patent applications described herein are incorporated herein in whole by reference to the same extent that each individual publication, patent, or patent application is explicitly and individually indicated as being incorporated herein by reference. Furthermore, the citation or identification of any reference in this application should not be construed as an acknowledgment that such reference is available as prior art of the present invention. Where section headings are used, they should not necessarily be construed as restrictive.

Claims

1. A high-resolution encoder device for measuring and recognizing the relative positions of a first portion P and a second portion S of the encoder device, wherein the first portion P and the second portion S are movable relative to each other. a) A sensing element arranged along the length of the first portion P of the encoder device, wherein the sensing element presents a variable characteristic for sensing, and the variable characteristic changes over the length of the first portion P, b) n sensors for sensing the variable characteristics, wherein the sensors are separated from each other and arranged along the second portion S, and the sensors are configured to output a signal in response to sensing the variable characteristics, and c) A processing device connected to each of the n sensors to receive the signals and configured to pass them on to a relative position along the first portion P, forming a vector Vc(X) having entries from each sensor, thereby defining the relative positions, wherein the processing device is configured to calculate the relative positions from the vectors, and the processing device further determines that the vector function is subject to the following conditions, [Math 1] The following is obtained, where Δ is a predetermined threshold. The high-resolution encoder device comprises the processing unit, which is configured to acquire an initial position by scanning all possible positions until a position satisfying the condition.

2. The high-resolution encoder device according to claim 1, further comprising an analog-to-digital converter between the n sensors and the processing unit for digitizing the signals from the n sensors before presenting the signals to the processing unit.

3. The aforementioned signal is [Math 2] Vector A ｓ A high-resolution encoder device according to claim 1 or 2, which is input to the processing device as (X).

4. The aforementioned vector is an n-vector function [Math 3] The high-resolution encoder device according to claim 3, which is inserted into and defines the relative positions X of the portion P and the portion S.

5. The vector function V(X) is selected such that it satisfies the first and second mathematical conditions. The first mathematical condition is that there is a one-to-one (injective) relationship between X and V(X), so the vector value V(X) obtained at position X represents a characteristic of that position. Second mathematical condition: For any two positions X1 and X2, the norm [Math 4] and function [Math 5] There exists a threshold Δ such that such a threshold exists. [Math 6] or, function [Number 7] at a predetermined interval [Number 8] The high-resolution encoder device according to claim 4, which is monotonic over a certain period.

6. The vector function V(X) has size p, [Number 9] of [Number 10] It has the component, and the component B ｔ (X) is signal A ｓ A high-resolution encoder device according to claim 1, which is a mathematical combination of (X).

7. The aforementioned variable characteristics are Variable magnetic field strength resulting from permanent magnets of different polarities placed at several positions on the scale. Variable light intensity obtained by the variable transparency or reflectivity of various parts of the scale and numerous light sources, Variable light intensity obtained by the variable distance between the sensor and multiple light sources, Variable eddy current losses obtained by the irregular shape of the conductive part of the scale, and A high-resolution encoder device according to any one of claims 1 to 6, which is a member of the group consisting of variable inductance obtained by the variable magnetic properties of a rotor material.

8. A rotary encoder, wherein the relative position includes a relative angle, according to any one of claims 1 to 7, high-resolution encoder device.

9. A linear encoder, wherein one of the first and second parts is configured to follow a straight or curved path, and the relative position is related to the length along the straight or curved path, according to any one of claims 1 to 8.

10. The high-resolution encoder device according to any one of claims 1 to 9, wherein the processing device is configured to acquire an initial position estimate from the most recent known position.

11. A high-resolution encoder device according to any one of claims 1 to 10, wherein the values ​​of a vector function are stored in a lookup table.

12. A high-resolution encoding method for measuring and encoding the relative positions of a first portion P and a second portion S, wherein the first portion P and the second portion S are movable relative to each other, and the method is a) A variable characteristic for sensing is presented, the variable characteristic changes over the length of the first portion P, and the variable characteristic is presented via sensing elements arranged over the length of the first portion P. b) In the second part S, the variable characteristics are sensed at n discrete positions. c) Outputting signals from the n discrete locations in accordance with the sensing, d) Receiving the signal from each of the n discrete locations, e) Each of the relative positions is assigned a vector Vc(X) having entries from each sensor, thereby defining each of the relative positions, and the encoder uses the vector to measure and encode the relative positions, and f) The vector function is subject to the following conditions, [Math 11] The following is obtained, where Δ is a predetermined threshold. A high-resolution coding method, comprising obtaining an initial position by scanning all possible positions until a position satisfying a certain condition.

13. The high-resolution coding method according to claim 12, further comprising digitizing the signals from the n discrete sensors before providing the signals to the processing unit.

14. The aforementioned signal, [Math 12] Vector A ｓ The high-resolution coding method according to claim 12 or 13, comprising inputting (X) to a processing device.

15. Each of the aforementioned vectors is an n-vector function of the relative position X of the portion P and the portion S. [Number 13] The high-resolution coding method according to claim 14, which includes inserting into

16. The aforementioned vector function V(X) is selected such that it satisfies two mathematical conditions. The first mathematical condition is that there is a one-to-one (injective) relationship between X and V(X), so the vector value V(X) obtained at position X represents a characteristic of that position. Second mathematical condition: For any two positions X1 and X2, the norm [Number 14] and function [Number 15] There exists a threshold Δ such that such a threshold exists. [Number 16] or, function [Number 17] at a predetermined interval [Number 18] The high-resolution coding method according to claim 15, which is monotonic over a certain period.

17. The aforementioned vector function V(X) has size p, [Number 19] Ingredients [Number 20] It has the above component B ｔ (X) is signal A ｓ The high-resolution coding method according to claim 16, which is a mathematical combination of (X).

18. The aforementioned variable characteristics are Variable magnetic field strength resulting from permanent magnets of different polarities placed at several positions on the scale. Variable light intensity obtained by the variable transparency or reflectivity of various parts of the scale and numerous light sources, Variable light intensity obtained by the variable distance between the sensor and multiple light sources, Variable eddy current losses obtained by the irregular shape of the conductive part of the scale, and A high-resolution coding method according to any one of claims 12 to 17, which is a member of the group consisting of variable inductance obtained by the variable magnetic properties of a rotor material.

19. This includes performing a function search for a vector function, wherein the function search is performed Select the position of the portion P, and place or simulate a sensing element at the position of the portion P. Select the position of the portion S, and place or simulate a sensor at the position of the portion S. Calculate or present signals from each sensor, and Test the aforementioned signal to present candidate vector functions that satisfy the aforementioned conditions, and A high-resolution coding method according to any one of claims 16 to 18, wherein the candidate vector function is accepted if the above condition is met, and the search for the function is repeated if the above condition is not met.

20. The high-resolution coding method according to any one of claims 12 to 19, wherein the relative position includes a relative angle.

21. The high-resolution coding method according to any one of claims 12 to 19, wherein the first part and one of the second parts are configured to follow a straight path or a curved path, and the relative positions are related to the length along the straight path or curved path.

22. A high-resolution coding method according to any one of claims 12 to 21, comprising obtaining an initial position estimate from the most recent known position.

23. A high-resolution coding method according to any one of claims 12 to 22, comprising obtaining the value of a vector function from a lookup table.

24. The high-resolution coding method according to claim 16, comprising performing a start procedure, wherein the start procedure includes compiling the vector function from the variable characteristics.

25. The aforementioned starting procedure further, Selecting candidate characteristics and variations of the variable characteristics, Setting the vector function using the simulation of the candidate characteristics and their variations, and The high-resolution coding method according to claim 24, comprising verifying that the vector function satisfies the first and second conditions.

Images