Precision calibration method of tilt angle sensor

Abstract

The invention relates to a precision calibration method of a tilt angle sensor. The precision calibration method comprises the following steps: 1) carrying out preliminary calibration on a triaxial accelerometer of the tilt angle sensor by adopting a six-position calibration algorithm; 2) acquiring an included angle between a driving shaft of the tilt angle sensor and a horizontal plane by adopting a three-axis ATAN resolving algorithm; and 3) re-calibrating the tilt angle sensor by adopting a quadratic angle interpolation algorithm. Compared with the prior art, the device has the advantages of double-shaft + / -90-degree full-scale high accuracy, high calibration efficiency and the like.

Description

Technical Field

[0001] The present invention relates to the field of intelligent sensors, and in particular to a method for calibrating the accuracy of an inclination sensor. Background Art

[0002] Generally, a biaxial inclination sensor calculates the angle between a single axis of an accelerometer and the horizontal plane by using the projection vector of the gravitational acceleration on the single axis of the accelerometer. However, with this design method, the measurement range of the biaxial inclination sensor generally cannot reach a biaxial measurement range of ±90°, or even if the biaxial measurement range of ±90° is achieved, the accuracy of the sensor is very poor in the range of ±80° to ±90°. For a high-precision (typical value of 0.01°) inclination sensor product, the accuracy can only reach the level of 1° to 2° in the range of ±80° to ±90°. At the same time, due to the large measurement range of ±90°, there are too many calibration points in the existing linear interpolation algorithm during debugging, resulting in low calibration efficiency, and reducing the calibration points cannot reach the expected accuracy level. Summary of the Invention

[0003] To solve the technical problems in the background art, the present invention provides a method for calibrating the accuracy of an inclination sensor, including the following steps:

[0004] 1) Initially calibrate the triaxial accelerometer of the inclination sensor by using a six-position calibration algorithm;

[0005] 2) Obtain the angle between the active axis of the inclination sensor and the horizontal plane by using a triaxial ATAN calculation algorithm;

[0006] 3) Re-calibrate the inclination sensor by using a quadratic angle interpolation algorithm.

[0007] Further, the step 1) specifically includes the following steps:

[0008] 11) Construct a triaxial accelerometer error model;

[0009] 12) Align the six faces of the inclination sensor with the horizontal plane in turn, and obtain the triaxial acceleration values of the accelerometer at these six positions;

[0010] 13) Use the least squares method to estimate the zero bias matrix and the coefficient matrix of the error model to complete the initial calibration. Further, in the step 11), the expression of the triaxial accelerometer error model is:

[0011] Acc = K·TrueAcc + Offset

[0012]

[0013] Among them, Acc is the output matrix of the three-axis accelerometer, K is the coefficient matrix, TrueAcc is the true value matrix of acceleration, Offset is the zero deviation matrix, Acc x 、Acc y 、Acc z are respectively the outputs of the three-axis accelerometer on the X-axis, Y-axis and Z-axis, sen x 、sen y 、sen z are respectively the axis sensitivities of the three-axis accelerometer on the X-axis, Y-axis and Z-axis, TrueAcc x 、TrueAcc y 、TrueAcc z are respectively the true values of the three-axis accelerometer on the X-axis, Y-axis and Z-axis, X offset 、Y offset 、Z offset are respectively the zero biases of the three-axis accelerometer on the X-axis, Y-axis and Z-axis, Δ y-x 、Δ z-x 、Δ x-y 、Δ z-y 、Δ x-z 、Δ y-z are respectively the cross-axis errors.

[0014] Furthermore, in the step 12), if the acceleration values output by two of the three axes of the three-axis accelerometer are 0, it indicates that these two axes are both parallel to the horizontal plane. If the acceleration value output by the third axis of the three-axis accelerometer is +1g or -1g, it indicates that the third axis is perpendicular to the horizontal plane, thereby automatically judging the axis position.

[0015] Furthermore, in the step 2), by introducing the Z-axis data of the three-axis accelerometer, the atan algorithm is used to obtain the included angles between the X-axis and Y-axis of the three-axis accelerometer and the horizontal plane respectively.

[0016] Furthermore, the atan2f() function is used to calculate the included angles between the X-axis and Y-axis of the three-axis accelerometer and the horizontal plane respectively, then there are:

[0017]

[0018] Among them, θ x 、θ y are respectively the included angles between the X-axis and Y-axis as the active axes and the horizontal plane, and X, Y, Z are the acceleration values of the X-axis, Y-axis and Z-axis of the three-axis accelerometer after preliminary calibration respectively.

[0019] Furthermore, the step 3) specifically includes the following steps:

[0020] 31) Set multiple angular point intervals θ0 and corresponding calibration points according to the measurement range of the inclination sensor;

[0021] 32) Determine multiple calibration intervals based on the calibration points, and perform quadratic parabola function fitting on all calibration intervals to obtain the quadratic parabola function corresponding to each calibration interval;

[0022] 33) Use the quadratic parabola interpolation algorithm to calibrate the angles between the X-axis and Y-axis and the horizontal plane obtained in step 2) again.

[0023] Further, the angular point interval θ0 is set to 10° or 15°.

[0024] Further, in step 33), for the positive half measurement range [0°, 90°] of the inclination sensor, there is:

[0025] When the included angle θ ∈ [(n - 1)θ0, nθ0], n is a positive integer and Calibrate with the quadratic parabola function corresponding to the calibration interval θ ∈ [(n - 1)θ0, (n + 1)θ0];

[0026] When the included angle θ ∈ [90 - 2θ0, 90], calibrate with the quadratic parabola function corresponding to the calibration interval [90 - 2θ0, 90].

[0027] Further, step 3) also includes the following steps:

[0028] 34) After completing the re-calibration, set the factory zero point of the inclination sensor respectively, and finally complete the accuracy calibration of the inclination sensor.

[0029] Compared with the prior art, the present invention has the following advantages:

[0030] I. High accuracy in the full measurement range: The present invention uses a six-position calibration algorithm combined with a quadratic angle interpolation algorithm for double calibration respectively, which can make the accuracy of the inclination sensor at large angles in the range (±60° to ±90°) equivalent to that at 0° to ±60°, ensuring high precision of the inclination sensor in the full measurement range of the two axes.

[0031] II. High calibration efficiency: The present invention uses a quadratic angle interpolation algorithm instead of the original linear interpolation algorithm, that is, considering the possible non-linear interval of the input and output of the accelerometer, which is more in line with the actual situation, and can also increase the calibration point spacing, greatly reducing the number of calibration points, and the time-consuming for calibrating the inclination sensor can be shortened by about half. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a flowchart of the method of the present invention;

[0033] Figure 2 Flow chart of a six-position calibration algorithm for automatically determining position

[0034] Figure 3 Flow chart of a triaxial ATAN angle calculation algorithm

[0035] Figure 4 Flow chart of a quadratic angle interpolation calibration Detailed implementation manners

[0036] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. This embodiment is implemented on the premise of the technical solution of the present invention, and the detailed implementation manners and specific operation processes are given. However, the protection scope of the present invention is not limited to the following embodiments.

[0037] It should be noted that in this specification, similar reference numerals and letters denote similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0038] In the description of this embodiment, it should be noted that the orientation or positional relationship indicated by the terms "upper", "lower", "inner", "bottom", etc. is based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the inventive product is usually placed during use. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention.

[0039] The terms "first", "second", etc. are only used for distinguishing descriptions and cannot be construed as indicating or implying relative importance.

[0040] In the description of this embodiment, it should also be noted that unless otherwise clearly defined and limited, the terms "arranged", "connected", "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in this embodiment can be understood according to specific situations.

[0041] To make the purpose, technical solution and advantages of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0042] Embodiment

[0043] As Figure 1 shown, the present invention provides a method for calibrating the accuracy of an inclination sensor. The method includes the following steps:

[0044] 1) Collect the original data (original three-axis acceleration values) of the three-axis accelerometer of the inclination sensor, filter the original data to remove noise and high-frequency interference. For the filtered three-axis acceleration values, use the six-position calibration algorithm for automatically determining the position to perform preliminary calibration of the axis zero point, cross-axis error, and axis sensitivity. The six-position calibration is actually zero-bias compensation and coordinate system conversion processing for the accelerometer in the position state where the inclination sensor is located;

[0045] As Figure 2 shown below, the six-position calibration algorithm for automatically determining the position will be introduced.

[0046] The six-position calibration method specifically obtains the output data of the three-axis accelerometer at the +1g and -1g positions (a total of 6 positions) on the X-axis, Y-axis, and Z-axis in sequence, and then calculates the zero bias and rotation matrix of each axis to complete the calibration. The specific process is as follows:

[0047] 11) Construct a three-axis accelerometer error model, and the expression of this model is:

[0048] Acc=K·TrueAcc+Offset

[0049]

[0050] where Acc is the three-axis accelerometer output matrix, K is the coefficient matrix, TrueAcc is the acceleration true value matrix, Offset is the zero-point deviation matrix, Acc x 、Acc y 、Acc z are the outputs of the three-axis accelerometer on the X-axis, Y-axis, and Z-axis respectively, sen x 、sen y 、sen z are the axis sensitivities of the three-axis accelerometer on the X-axis, Y-axis, and Z-axis respectively, TrueAcc x 、TrueAcc y 、TrueAcc z are the true values of the three-axis accelerometer on the X-axis, Y-axis, and Z-axis respectively, X offset 、Y offset 、Z offset are the zero biases of the three-axis accelerometer on the X-axis, Y-axis, and Z-axis respectively, Δ y-x 、Δ z-x 、Δ x-y 、Δ z-y 、Δ x-z 、Δ y-z are the cross-axis errors respectively.

[0051] 12) Align the six faces of the inclination sensor with the horizontal plane in sequence, and obtain the triaxial acceleration values of the accelerometer at these six positions;

[0052] In consideration of the fact that during actual operation, the calibration personnel cannot determine which axis is in the +1g or -1g position, the present invention determines the triaxial acceleration values output by the triaxial accelerometer when the inclination sensor receives a calibration command. The directions of the two axes with acceleration values close to 0 are parallel to the horizontal plane, and the direction of the axis with an acceleration value close to +1g or -1g is perpendicular to the horizontal plane. Thus, it is determined which axis is in the +1g or -1g position currently, avoiding incorrect calibration. During automatic calibration, it is only necessary to align the six faces of the inclination sensor with the horizontal plane in sequence, and send the commands "P1", "P2", "P3", "P4", "P5", and "P6" to represent the first to sixth positions in this process, and then the acquisition of the triaxial acceleration value data at the six positions can be completed.

[0053] 13) Use the adjoint matrix method to obtain the inverse matrix (i.e., the rotation matrix) of the coefficient matrix, use the least squares method to estimate and obtain the bias matrix and the coefficient matrix, and save them to the non-volatile storage unit such as the EEPROM of the inclination sensor to complete the preliminary calibration. When actually measuring the angle, it is only necessary to multiply the triaxial acceleration value output by the accelerometer minus the bias by the rotation matrix to obtain the true value of the triaxial acceleration, and use this as the output value after the preliminary calibration of the accelerometer.

[0054] 2) According to the triaxial acceleration values after the preliminary calibration in step 1), by introducing the acceleration data of the Z axis, use the arctangent algorithm to calculate the angles between the X axis and the Y axis and the horizontal plane respectively. At this time, the output angle of the inclination sensor can already reach a certain calibration accuracy. Subsequently, use the quadratic parabola interpolation algorithm for the set angle calibration points to further improve the accuracy of the output angle. After the re-calibration by the quadratic parabola interpolation algorithm, the accuracy of the inclination sensor can reach a typical value of ±0.01° and a maximum value of ±0.02° within the range of 0 to ±85°;

[0055] As Figure 3 shown, the triaxial ATAN solution algorithm adopted by the present invention is as follows:

[0056] Existing dual-axis inclination sensors generally use the SIN algorithm (sine algorithm) for dual-axis angle calculation. It has the advantages of convenient calculation and relatively independent variables between the two axes. However, it also has the disadvantages of low accuracy for large-range angles and too many calibration points. In the present invention, the ATAN algorithm is used for calculation based on the acceleration values of the three axes of the accelerometer. First, the sum of the squares of the two passive axes is calculated, then it is square-rooted, and finally, the atan2f() function is used with the acceleration value of the active axis and the root mean square value of the two passive axes as parameters to calculate the angle between the active axis and the horizontal plane. There is:

[0057]

[0058] In the formula, θ x 、θ y are the angles between the X-axis and the Y-axis and the horizontal plane when they are the active axes respectively. X, Y, and Z are the acceleration values of the X-axis, Y-axis, and Z-axis of the three-axis accelerometer after preliminary calibration.

[0059] 3) Place the inclination sensor product on a horizontal marble platform and perform secondary calibration on the inclination sensor;

[0060] As Figure 4 shown, the present invention uses the quadratic parabola interpolation algorithm for the set angle calibration points to realize the secondary calibration of the inclination sensor. The main steps include setting the calibration point interval and calibration range, calculating the coefficients of the quadratic parabola function, and using the quadratic angle interpolation algorithm for secondary calibration, specifically including the following steps:

[0061] 31) Set the angle point interval θ0 of the quadratic angle interpolation algorithm and the corresponding calibration points according to the range of the inclination sensor;

[0062] When selecting the angle point interval θ0, it should be divisible by the range. In this example, the range of the inclination sensor is [-90°, +90°]. In addition, it also needs to be determined according to the calibration efficiency requirements and the actual input and output of the accelerometer. After experimental testing, the angle point interval θ0 is preferably 10° or 15°. When the angle point interval θ0 is set to 15°, the calibration points are set to ±15°, ±30°, ±45°, ±60°, ±75°, ±90° for a total of 12 according to the angle point interval θ0; when the angle point interval θ0 is 10°, the calibration points are set to ±10°, ±20°, ±30°, ±40°, ±50°, ±60°, ±70°, ±80°, ±90° for a total of 18.

[0063] To ensure calibration accuracy, in existing linear interpolation calibration algorithms, the calibration point interval is usually set to 5° or less, and the calibration points are set to ±5°, ±10°, ±15°, ±20°, ±25°, ±30°, ±35°, ±40°, ±45°, ±50°, ±55°, ±60°, ±65°, ±70°, ±75°, ±80°, ±85°, ±90°, a total of 36 and more. It can be seen that the quadratic angle interpolation algorithm adopted by the present invention can greatly reduce the number of calibration points and increase the calibration efficiency on the premise of improving calibration accuracy.

[0064] 32) Determine multiple calibration intervals according to the calibration points, and perform quadratic parabola function fitting on all calibration intervals to obtain the quadratic parabola function corresponding to each calibration interval, and store the parameters of the quadratic parabola function in a non-volatile storage unit such as the EEPROM of the tilt sensor;

[0065] Taking the angle point interval of 15° as an example, there are a total of 10 calibration intervals, 5 in each of the positive and negative half ranges, which are [0°, 30°], [15°, 45°], [30°, 60°], [45°, 75°], [60°, 90°], [-30°, 0°], [-45°, -15°], [-60°, -30°], [-75°, -45°], [-90°, -60°]. Each interval covers three calibration points, so quadratic parabola fitting can be performed with these three calibration points.

[0066] Taking the angle point interval of 10° as an example, there are a total of 16 calibration intervals, 8 in each of the positive and negative half ranges, which are [0°, 20°], [10°, 30°], [20°, 40°], [30°, 50°], [40°, 60°], [50°, 70°], [60°, 80°], [70°, 90°], [-20°, 0°], [-30°, -10°], [-40°, -20°], [-50°, -30°], [-60°, -40°], [-70°, -50°], [-80°, -60°], [-90°, -70°].

[0067] Theoretically, the input-output change is linear within the range. However, since the measurement characteristics of the three-axis accelerometers used in each tilt sensor are different during production and manufacturing, there may be multiple non-linear intervals. At this time, if linear interpolation is used for calibration, it will affect the measurement accuracy. Therefore, in order to better fit the actual input-output characteristics of the accelerometer, the present invention adopts a quadratic angle interpolation algorithm for re-calibration.

[0068] 33) Use quadratic parabola interpolation to re-calibrate the angles between the X-axis and Y-axis calculated by the arctangent algorithm as the active axes and the horizontal plane, and finally obtain an accurate angle output value.

[0069] The angle θ between the X-axis and the horizontal plane after preliminary calibration x For example (θ y Same), first determine the angle θ x Is it out of the calibration range (-90° to 90° in this example), if yes, then end directly, if not, then determine the angle θ x In which calibration interval, the calibration is then performed according to the quadratic parabola function corresponding to the calibration interval. In addition, the angle θ x It is also possible that the calibration interval is located in two adjacent calibration intervals at the same time. In this case, the present invention selects the calibration interval with the smaller first calibration point for calibration;

[0070] When the angle θ x When it is within the first to the third-to-last calibration interval (taking the angle point interval of 15° as an example, the first calibration interval corresponding to the positive and negative half range is [0°, ±30°], the second calibration interval is [±15°, ±45°], and the third-to-last calibration interval is [±30°, ±60°], that is, 0 to ±60°), according to the angle θ x The quadratic parabola function corresponding to the calibration interval is calibrated, that is:

[0071] When θ x ∈[0°,15°], select the quadratic parabola function corresponding to the first calibration interval [0°,30°]; when θ x ∈[15°,30°], select the quadratic parabola function corresponding to the second calibration interval [15°,45°]; when θ x ∈[30°,45°], select the quadratic parabola function corresponding to the third calibration interval [30°,60°], and the same applies to the negative half range;

[0072] When the angle θ x When it is in the second-to-last calibration interval (taking the angle point interval of 15° as an example, the second-to-last calibration interval corresponding to the positive and negative half range is [±45°, ±75°]), calibration is performed according to the quadratic parabola function corresponding to the calibration interval, that is:

[0073] When θ x ∈[45°,60°], select the quadratic parabola function corresponding to the penultimate calibration interval [45°,75°], and the same applies to the negative half range;

[0074] When the angle θ x When it is in the penultimate calibration interval (taking the angle point interval of 15° as an example, the penultimate calibration interval corresponding to the positive and negative half range is [±60°, ±90°]), calibration is performed according to the quadratic parabola function corresponding to the calibration interval, that is:

[0075] When θ x ∈ [60°, 90°], select the quadratic parabola function corresponding to the penultimate calibration interval [60°, 90°]. The same applies to the negative half range.

[0076] 4) After completing the re - calibration, set the factory zero points of the X - axis and Y - axis of the inclinometer sensor respectively to complete the accuracy calibration of the inclinometer sensor.

[0077] In summary, the present invention adopts a combined calibration method, which can effectively improve the accuracy of the inclinometer sensor in a large - range angle. After introducing the Z - axis acceleration value variable, by applying the combination of the six - position algorithm + ATAN angle calculation + quadratic parabola interpolation algorithm, and adopting the method of preliminary calibration combined with re - calibration, the detection accuracy is greatly improved. It not only does not increase the calibration time, but instead shortens the calibration time of the existing calibration method by more than half, achieving the effect of 1 + 1 + 1>3.

[0078] The preferred specific embodiments of the present invention have been described in detail above. It should be understood that those of ordinary skill in the art can make many modifications and variations based on the concept of the present invention without creative work. Therefore, all technical solutions that can be obtained by those skilled in the art in the technical field of the present invention based on the concept of the present invention through logical analysis, reasoning, or limited experiments on the basis of the prior art should be within the protection scope determined by the claims.

Claims

1. A method for calibrating the accuracy of an inclination sensor, characterized in that, It includes the following steps: 1) Initially calibrate the triaxial accelerometer of the tilt sensor using a six-position calibration algorithm; 2) Obtain the angle between the active axis of the tilt sensor and the horizontal plane using a triaxial ATAN calculation algorithm; 3) Recalibrate the tilt sensor using a quadratic angle interpolation algorithm.

2. The accuracy calibration method of an inclination sensor according to claim 1, characterized in that, The specific steps of step 1) include the following steps: 11) Construct a triaxial accelerometer error model; 12) Align the six faces of the tilt sensor with the horizontal plane in sequence, and obtain the triaxial acceleration values of the accelerometer at these six positions; 13) Use the least squares method to estimate the bias matrix and coefficient matrix of the error model to complete the initial calibration.

3. A method for calibrating the accuracy of an inclination sensor according to claim 2, characterized in that, In step 11), the expression of the triaxial accelerometer error model is: Acc = K·TrueAcc + Offset Among them, Acc is the output matrix of the triaxial accelerometer, K is the coefficient matrix, TrueAcc is the matrix of the true acceleration values, Offset is the zero-offset matrix, Acc x 、Acc y 、Acc z are respectively the outputs of the triaxial accelerometer on the X-axis, Y-axis, and Z-axis, sen x 、sen y 、sen z are respectively the axis sensitivities of the triaxial accelerometer on the X-axis, Y-axis, and Z-axis, TrueAcc x 、TrueAcc y 、TrueAcc z are respectively the true values of the triaxial accelerometer on the X-axis, Y-axis, and Z-axis, X offset 、Y offset 、Z offset are respectively the zero biases of the triaxial accelerometer on the X-axis, Y-axis, and Z-axis, Δ y-x 、Δ z-x 、Δ x-y 、Δ z-y 、Δ x-z 、Δ y-z are respectively the cross-axis errors.

4. A method for calibrating the accuracy of an inclination sensor according to claim 2, characterized in that In step 12), if the acceleration values output by two of the three axes of the triaxial accelerometer are 0, it indicates that these two axes are both parallel to the horizontal plane. If the acceleration value output by the third axis of the triaxial accelerometer is +1g or -1g, it indicates that the third axis is perpendicular to the horizontal plane, and the axis position is automatically judged in this way.

5. A method for calibrating the accuracy of an inclination sensor according to claim 1, characterized in that, In step 2), by introducing the Z-axis data of the triaxial accelerometer, use the arctangent algorithm to obtain the angles between the X-axis and Y-axis of the triaxial accelerometer and the horizontal plane respectively.

6. A method for calibrating the accuracy of an inclination sensor according to claim 5, characterized in that, Use the atan2f() function to calculate the angles between the X-axis and Y-axis of the triaxial accelerometer and the horizontal plane respectively, then there are: where θ x and θ y are the angles between the X-axis and the Y-axis and the horizontal plane when they are the driving axes respectively, and X, Y, and Z are the acceleration values of the X-axis, Y-axis, and Z-axis of the triaxial accelerometer after preliminary calibration.

7. A method for calibrating the accuracy of an inclination sensor according to claim 1, characterized in that, The specific steps of step 3) include the following steps: 31) Set multiple angle point intervals θ0 and corresponding calibration points according to the measurement range of the tilt sensor; 32) Determine multiple calibration intervals according to the calibration points, and perform quadratic parabola function fitting on all calibration intervals to obtain the quadratic parabola function corresponding to each calibration interval; 33) Use the quadratic parabola interpolation algorithm to recalibrate the angles between the X-axis and Y-axis obtained in step 2) and the horizontal plane.

8. A method for calibrating the accuracy of an inclination sensor according to claim 7, characterized in that The angle point interval θ0 is set to 10° or 15°.

9. The accuracy calibration method of an inclination sensor according to claim 7, characterized in that In step 33), for the positive half measurement range [0°, 90°] of the tilt sensor, there are: When the included angle θ ∈ [(n - 1)θ0, nθ0], where n is a positive integer and calibration is performed using the quadratic parabola function corresponding to the calibration interval θ ∈ [(n - 1)θ0, (n + 1)θ0]; When the angle θ ∈ [90 - 2θ0, 90], calibrate using the quadratic parabola function corresponding to the calibration interval [90 - 2θ0, 90].

10. A method for calibrating the accuracy of an inclination sensor according to claim 7, characterized in that Step 3) also includes the following steps: 34) After completing the recalibration, set the factory zero point of the tilt sensor respectively, and finally complete the accuracy calibration of the tilt sensor.

Images