A method for automatic calibration of MEMS accelerometer

Abstract

The present invention relates to a method for automatic calibration of a MEMS accelerometer. Taking the z-axis as an example, the method comprises the following steps: sequentially judging the horizontal state, static state and normal state of the MEMS accelerometer, if the MEMS accelerometer is in an abnormal state, the calibration action cannot be performed; utilizing the gravity characteristics of the earth, when the object is static, the three-axis resultant force F is the standard value of gravity acceleration, the z-axis is calibrated with the standard value, and the difference obtained is used to compensate the read z-axis original data; in a static state, the z-axis compensated data is monitored, if the absolute value of the difference between the z-axis compensated data and the standard value is greater than the expected value, it indicates that the MEMS accelerometer has temperature drift, and a re-calibration action needs to be triggered to solve the data offset caused by the temperature drift, if the absolute value of the difference between the z-axis compensated data and the standard value is less than the expected value, the calibration is performed according to the state of the MEMS accelerometer. The present invention can realize automatic calibration and temperature drift calibration of the MEMS accelerometer.

Description

Technical Field

[0001] The invention relates to the application field of inertial sensors, and in particular to an automatic calibration method for a MEMS accelerometer. Background Art

[0002] After the MEMS accelerometer is mounted and assembled, it will be affected by various stresses, which will cause data offset, and in severe cases it will even affect the normal use function. Therefore, in order to obtain the best customer experience, data card control will be carried out on the mounting production line and assembly production line, generally including horizontal, 6D, 45° tilt and other tests. If these test items need to be passed smoothly, manual data calibration must be performed. The use of manual data calibration will not only increase the production process and increase production costs, but also may cause calibration errors due to human factors. At the same time, MEMS accelerometers have the problem of temperature drift, and data offset will occur when the temperature rises or falls. Therefore, it is necessary to improve the existing calibration method of MEMS accelerometers. Summary of the invention

[0003] The purpose of the present invention is to provide a method for automatically calibrating a MEMS accelerometer during use, which can save the manual data calibration process, improve production efficiency, and solve the problems that the manual data calibration process is adopted in the calibration method of the MEMS accelerometer in the prior art described in the background technology, increase the production process, increase the production cost, human factors lead to calibration errors, and the MEMS accelerometer has temperature drift, which also leads to data offset.

[0004] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for automatic calibration of a MEMS accelerometer, which comprises the following steps:

[0005] S1, determine whether the MEMS accelerometer is in a horizontal state;

[0006] S2. If the MEMS accelerometer is in a horizontal state, determine whether the MEMS accelerometer is in a stationary state;

[0007] S3, if the MEMS accelerometer is in a stationary state, determine whether the MEMS accelerometer is in a normal state. If the MEMS accelerometer is in an abnormal state, the calibration action cannot be performed;

[0008] S4. Using the characteristics of the earth's gravity, when the object is perpendicular to the ground, the force detected by the vertical axis is the standard value, and when the object is stationary, the three-axis force F is the standard value. The z-axis is calibrated with the standard value, and the difference obtained is used to compensate the original data of the z-axis.

[0009] S5, monitoring the data after z-axis compensation in a stationary state. If the data deviation after z-axis compensation is greater than the expected value, it means that the MEMS accelerometer has temperature drift, and it is necessary to trigger a recalibration action to solve the data offset caused by the temperature drift. If the data after z-axis compensation is less than the expected value, proceed to step S6;

[0010] S6. When horizontal and stationary, the output data of the x-axis and y-axis are proportionally converged, and the z-axis output data is a simulated value calculated using the resultant force formula, and the ± sign is determined by the original data of the z-axis;

[0011] When the x-axis is vertical and stationary, and the z-axis has been calibrated, the y-axis outputs a converged value, and the z-axis output data uses the simulated value calculated using the resultant force formula, with the ± sign determined by the z-axis raw data;

[0012] When the y-axis is vertical and stationary, and the z-axis has been calibrated, the x-axis outputs a converged value, and the z-axis output data uses the simulated value calculated using the resultant force formula, with the ± sign determined by the z-axis raw data;

[0013] When the x-axis is vertical and stationary, and the z-axis is uncalibrated, the y-axis and z-axis outputs converge;

[0014] When the y-axis is vertical and stationary, and the z-axis is uncalibrated, the x-axis and z-axis outputs converge;

[0015] The above is the calibration method for the z-axis, and the calibration process for the x-axis and the y-axis corresponds to the above steps.

[0016] In the above scheme, in step S1, the method for judging whether the MEMS accelerometer is in a horizontal state is that, when performing data processing, first determine whether there is external calibration data. If there is external calibration data, it is necessary to compensate the external calibration data to the three-axis original data, and use the horizontal threshold 1 to judge the horizontal state of the MEMS accelerometer, that is, the data after the x-axis and y-axis compensation must be less than the horizontal threshold 1 at the same time, indicating that the MEMS accelerometer is in a horizontal state; if there is no external calibration data, the horizontal threshold 2 is used to judge the horizontal state of the MEMS accelerometer, that is, the data after the x-axis and y-axis compensation must be less than the horizontal threshold 1 at the same time, indicating that the MEMS accelerometer is in a horizontal state.

[0017] In the above scheme, the level threshold 1 is 100 mg, and the level threshold 2 is 200 mg.

[0018] In the above scheme, in step S2, the method for determining whether the MEMS accelerometer is in a stationary state is to collect multiple sets of three-axis raw data using the MEMS accelerometer. If the sum of the changes of these three-axis raw data is less than the stationary threshold, it means that the MEMS accelerometer is in a stationary state, that is, if |X 最大值 -X 最小值 |+|Y 最大值-Y 最小值| +|Z 最大值 -Z 最小值 |<static threshold, it means that the MEMS accelerometer meets the static condition.

[0019] In the above scheme, the resting threshold is 80 mg.

[0020] In the above scheme, in step S3, the method for determining whether the MEMS accelerometer is in a normal state is to calculate the three-axis resultant force value, and the calculation formula is: ;

[0021] If F is greater than the normal threshold, it means that the MEMS accelerometer is in a normal state; otherwise, it means that the MEMS accelerometer is in an abnormal state and the calibration action cannot be performed.

[0022] In the above scheme, the normal threshold is 250 mg.

[0023] In the above scheme, in step S5, the expected value is 100 mg.

[0024] In the above solution, in step S6, the output convergence value is 1 / 2 of the original data.

[0025] The present invention has positive effects: 1) The MEMS accelerometer automatic calibration method of the present invention can automatically calibrate the MEMS accelerometer, eliminating the manual calibration process, which can not only reduce labor consumption, but also improve the efficiency and accuracy of calibration and avoid the influence of human factors; 2) The MEMS accelerometer automatic calibration method of the present invention can not only make the calibrated MEMS accelerometer meet the production line level test standard, but also meet the test standards of 6D, 45° tilt and other test items; 3) The calibration method of the present invention can not only calibrate the MEMS accelerometer's own error, but also calibrate the temperature drift of the MEMS accelerometer, so that the output value of the MEMS accelerometer is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 Flowchart for calibration of the z-axis of a MEMS accelerometer.

[0027] Figure 2 Calibration flow chart for the x-axis of a MEMS accelerometer.

[0028] Figure 3 Calibration flow chart for the y-axis of a MEMS accelerometer. DETAILED DESCRIPTION

[0029] The technical scheme of the present invention is clearly and completely described below through embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] Example 1

[0031] In the automatic calibration method of the MEMS accelerometer of the present invention, when the zero bias value of any axis among the three axes x, y, and z is greater than the usage or test requirements of the customer, corresponding data calibration is required.

[0032] For example, assuming that the MEMS accelerometer is horizontally facing upward, the x-axis raw data is 150 mg, the y-axis raw data is -100 mg, and the z-axis raw data is 600 mg. When the MEMS accelerometer is placed horizontally, the theoretical values ​​of the x, y, and z-axis data of the accelerometer are 0 mg, 0 mg, and 1000 mg. The difference between the actual output value and the theoretical value is the zero bias, so the zero bias values ​​corresponding to the above raw data are 150 mg, -100 mg, and -400 mg, respectively. The production line horizontal test standard requires that the three-axis data bias value must be less than 100 mg, that is, -100 mg < x < 100 mg, -100 mg < y < 100 mg, 900 mg < z < 1100 mg. The use of raw data cannot meet the production line test standard. In the prior art, the MEMS accelerometer needs to be calibrated through a manual calibration process on the production line. If the calibration method of the present invention is used to automatically calibrate the z-axis, the horizontal test of the production line can be directly passed. The specific calibration method is as follows: Figure 1 As shown in the process, it specifically includes the following steps:

[0033] Turn on the power;

[0034] Because there is no customer calibration, the horizontal threshold 2 should be used as the judgment standard. Because in the above raw data, the x-axis raw data and the y-axis raw data are both less than 200 mg of the horizontal threshold 2, so the horizontal condition is met, and it is determined that the MEMS accelerometer is in a horizontal state;

[0035] Read multiple sets of x, y, and z axis raw data. Usually, at least 10 sets of x, y, and z axis raw data are read, and the change sum of these 10 sets of x, y, and z axis raw data is judged to be less than the static threshold. The judgment method is: judge |x 最大值 -x 最小值 |+|y 最大值 -y 最小值 |+|z 最大值 -z 最小值| Is it less than 80 mg? In this embodiment, the sum of the changes of 10 sets of x, y, and z three-axis raw data is less than the static threshold, so the static condition is met;

[0036] Because calibration has not been performed at this time, there is no need to consider compensating the calibration data to the three-axis raw data. The x-axis raw data and the y-axis raw data converge through proportionality, and the outputs are half of the original data, that is, the x-axis and y-axis outputs are 75mg and -50mg respectively, and the z-axis output is calculated using the following formula:

[0037]

[0038] At the same time, according to the direction of the original data of the z-axis, the z-axis output is about 996 mg, which is the simulation value, and the z-axis data compensation value is -396 mg (600-996).

[0039] The three-axis original data (150mg, -100mg, 600mg) are automatically calibrated to (75mg, -50mg, 996mg), thus meeting the production line level test standards (-100mg < x < 100mg, -100mg < y < 100mg, 900mg < x < 1100mg).

[0040] Example 2

[0041] Assume that when the x-axis of the MEMS accelerometer is vertical, the raw data of the x-axis is 850 mg, the raw data of the y-axis is 100 mg, and the raw data of the z-axis is 150 mg. When the MEMS accelerometer is placed horizontally, the theoretical values ​​of the x, y, and z-axis data of the accelerometer are 1000 mg, 0 mg, and 0 mg. The difference between the actual output value and the theoretical value is the zero bias, so the zero bias values ​​corresponding to the above raw data are -150 mg, -100 mg, and 150 mg, respectively. The production line horizontal test standard requires that the three-axis data bias value must be less than 100 mg, that is, 900 mg < x < 1100 mg, -100 mg < y < 100 mg, -100 mg < z < 100 mg, then the use of raw data cannot meet the production line test standard. In the prior art, the MEMS accelerometer needs to be calibrated through a manual calibration process on the production line. If the calibration method of the present invention is used to automatically calibrate the x-axis, the horizontal test of the production line can be directly passed. The specific calibration method is as follows: Figure 2 As shown in the process, it specifically includes the following steps:

[0042] Turn on the power;

[0043] Because there is no customer calibration, the horizontal threshold 2 should be used as the judgment standard. Because in the above raw data, the z-axis raw data and the y-axis raw data are both less than 200 mg, so the y-axis and z-axis horizontal conditions are met, and the MEMS accelerometer is in the x-axis vertical state;

[0044] Read multiple sets of x, y, and z axis raw data. Usually, at least 10 sets of x, y, and z axis raw data are read, and the change sum of these 10 sets of x, y, and z axis raw data is judged to be less than the static threshold. The judgment method is: judge |x 最大值 -x 最小值 |+|y 最大值 -y 最小值 |+|z 最大值 -z 最小值 | Is it less than 80 mg? In this embodiment, the sum of the changes of 10 sets of x, y, and z three-axis raw data is less than the static threshold, so the static condition is met;

[0045] Because calibration has not been performed at this time, there is no need to consider compensating the calibration data to the three-axis raw data. The z-axis raw data and the y-axis raw data are converged by proportion, and the output is half of the original data, that is, the z-axis and y-axis outputs are 75mg and 50mg respectively, and the x-axis output is calculated using the following formula:

[0046]

[0047] At the same time, according to the direction of the original x-axis data, the x-axis output is about 996 mg, which is the simulation value, and the x-axis data compensation value is -146 mg (850-996).

[0048] The three-axis original data (850mg, 100mg, 150mg) is automatically calibrated to (996mg, 50mg, 75mg), thus meeting the production line level test standards (900mg < x < 1100mg, -100mg < y < 100mg, -100mg < z < 100mg).

[0049] Example 3

[0050] Assume that when the y-axis of the MEMS accelerometer is reversed and vertical, the raw data of the x-axis is 50 mg, the raw data of the y-axis is -1500 mg, and the raw data of the z-axis is -150 mg. When the MEMS accelerometer is placed horizontally, the theoretical values ​​of the x, y, and z-axis data of the accelerometer are 0 mg, -1000 mg, and 0 mg. The difference between the actual output value and the theoretical value is the zero bias, so the zero bias values ​​corresponding to the above raw data are 50 mg, -500 mg, and -150 mg, respectively. The production line horizontal test standard is that the three-axis data bias must be less than 100 mg, that is, -100 mg < x < 100 mg, -1100 mg < y < -900 mg, -100 mg < z < 100 mg, then the use of raw data cannot meet the production line test standard. In the prior art, the MEMS accelerometer needs to be calibrated through a manual calibration process on the production line. If the calibration method of the present invention is used to automatically calibrate the y-axis, the horizontal test of the production line can be directly passed. The specific calibration method is as follows. Figure 3 As shown in the process, it specifically includes the following steps:

[0051] Turn on the power;

[0052] Because there is no customer calibration, the horizontal threshold 2 should be used as the judgment standard. Because in the above raw data, the x-axis raw data and the z-axis raw data are both less than 200 mg, so the x-axis and z-axis horizontal conditions are met, and the MEMS accelerometer is in the y-axis vertical state;

[0053] Read multiple sets of x, y, and z axis raw data. Usually, at least 10 sets of x, y, and z axis raw data are read, and the change sum of these 10 sets of x, y, and z axis raw data is judged to be less than the static threshold. The judgment method is: judge |x 最大值 -x 最小值 |+|y 最大值 -y 最小值 |+|z 最大值 -z 最小值 | Is it less than 80 mg? In this embodiment, the sum of the changes of 10 sets of x, y, and z three-axis raw data is less than the static threshold, so the static condition is met;

[0054] Because calibration has not been performed at this time, there is no need to consider compensating the calibration data to the three-axis raw data. The x-axis raw data and the z-axis raw data converge through proportionality, and the outputs are half of the original data, that is, the x-axis and z-axis outputs are 25mg and -75mg respectively, and the y-axis output is calculated using the following formula:

[0055]

[0056] At the same time, according to the direction of the original data on the y-axis, the y-axis output is approximately -997 mg, which is the simulation value, and the y-axis data compensation value is -503 mg (-1500+997).

[0057] The three-axis original data (50mg, -1500mg, -150mg) are automatically calibrated to (25mg, -997mg, -75mg), thus meeting the production line level test standards (-100mg < x < 100mg, -1100mg < y < -900mg, -100mg < z < 100mg).

[0058] Test Example 1

[0059] A production line level detection method for a MEMS accelerometer comprises the following steps:

[0060] S1. After the production line patch is installed, place it horizontally upward on the table;

[0061] S2. Turn on the machine and enter the engineering mode to observe the three-axis data of the accelerometer;

[0062] S3. When the absolute value of the zero bias of the x, y, and z axes is less than spec 100mg, the horizontal test item is judged to be qualified, otherwise, it is unqualified.

[0063] It can be seen from the above method that after the MEMS accelerometers of Examples 1-3 are calibrated by the calibration method of the present invention, the absolute values ​​of the zero bias of the x, y, and z axes are all less than spec 100 mg, so the MEMS accelerometers of Examples 1-3 are qualified in the horizontal detection after calibration.

[0064] Test Example 2

[0065] The 6D detection test of the MEMS accelerometer includes the following steps:

[0066] S1. After the production line patch is installed, place it horizontally upward on the table;

[0067] S2. Turn on the machine and enter the engineering mode to observe the three-axis data of the accelerometer;

[0068] S3. When the absolute value of the zero bias of the x, y, and z axes is less than the spec 100mg, the test is considered qualified, otherwise, it is unqualified;

[0069] S4. Place it horizontally downward on the table and observe the three-axis data of the accelerometer. When the absolute value of the zero bias of the x, y, and z axes is less than the spec 100mg, the test is considered qualified, otherwise, it is unqualified;

[0070] S5. The x-direction is perpendicular to the desktop, and the three-axis data of the accelerometer is observed. When the absolute value of the zero bias of the x-, y-, and z-axes is less than spec100mg, the test is considered qualified, otherwise, it is unqualified.

[0071] S6. Turn the x-axis perpendicular to the desktop and observe the three-axis data of the accelerometer. When the absolute value of the zero bias of the x-axis, y-axis, and z-axis is less than spec100mg, the test is considered qualified, otherwise, it is unqualified.

[0072] S7, y is perpendicular to the desktop, observe the three-axis data of the accelerometer, when the absolute value of the zero bias of the x, y, z axes is less than spec100mg, the test is considered qualified, otherwise, it is unqualified;

[0073] S8, y is perpendicular to the desktop in the reverse direction, and the three-axis data of the accelerometer is observed. When the absolute value of the zero bias of the x, y, and z axes is less than spec100mg, the test is considered qualified, otherwise, it is unqualified.

[0074] S9. When the above 6 direction tests are all qualified, it means that the 6D test item is qualified.

[0075] The principle is the same as that of test example 1. After the MEMS accelerometers of Examples 1-3 are calibrated by the calibration method of the present invention, the absolute values ​​of the zero bias of the x, y, and z axes are all less than spec 100 mg. Therefore, the MEMS accelerometers of Examples 1-3 are qualified in the 6D test after calibration.

[0076] Test Example 3

[0077] The 45° oblique angle detection test of the MEMS accelerometer includes the following steps:

[0078] S1. After the production line patch is installed, place it horizontally upward on the table;

[0079] S2. Turn on the machine and enter the engineering mode to observe the three-axis data of the accelerometer;

[0080] S3. Place it on the 45° inclined test fixture and observe the three-axis data of the accelerometer. The theoretical values ​​of the three axes are: x=500mg, y=500mg, z=707mg, and the control range is ±50mg. That is, if x is between 450 and 550mg, y is between 450 and 550mg, and z is between 657 and 757mg, the 45° inclined angle test item is judged to be qualified.

[0081] The principle is the same as that of detection examples 1 and 2. After the MEMS accelerometers of embodiments 1-3 are calibrated by the calibration method of the present invention, the output data of the three axes x, y, and z can all fall within the corresponding ranges, that is, x is between 450 and 550 mg, y is between 450 and 550 mg, and z is between 657 and 757 mg. Therefore, the MEMS accelerometers of embodiments 1-3 are qualified in the 45° angle detection after calibration.

[0082] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for automatic calibration of a MEMS accelerometer, characterized in that: It includes the following steps: S1, determine whether the MEMS accelerometer is in a horizontal state; S2. If the MEMS accelerometer is in a horizontal state, determine whether the MEMS accelerometer is in a stationary state; S3, if the MEMS accelerometer is in a stationary state, determine whether the MEMS accelerometer is in a normal state. If the MEMS accelerometer is in an abnormal state, the calibration action cannot be performed; S4. Using the characteristics of the earth's gravity, when the object is stationary, the three-axis resultant force F is the standard value of gravity acceleration. The z-axis is calibrated with the standard value, and the difference obtained is used to compensate the original data of the z-axis. S5, monitoring the z-axis compensated data in a stationary state, if the absolute value of the difference between the z-axis compensated data and the standard value is greater than the expected value, it means that the MEMS accelerometer has temperature drift, and it is necessary to trigger a recalibration action to solve the data offset caused by the temperature drift, if the absolute value of the difference between the z-axis compensated data and the standard value is less than the expected value, proceed to step S6; S6. When the MEMS accelerometer is horizontal and stationary, the output data of the x-axis and y-axis are proportionally converged, and the z-axis output data is a simulated value calculated using the resultant force formula, and the ± sign is determined by the original data of the z-axis; When the x-axis is vertical and stationary, and the z-axis is not calibrated, the y-axis outputs a converged value, and the z-axis output data uses the simulated value calculated using the resultant force formula, with the ± sign determined by the z-axis raw data; When the y-axis is vertical and stationary, and the z-axis is not calibrated, the x-axis outputs a converged value, and the z-axis outputs a simulated value calculated using the resultant force formula, with the ± sign determined by the z-axis raw data; When the x-axis is vertical and stationary, and the z-axis is calibrated, the y-axis and z-axis outputs converge; When the y-axis is vertical and stationary, and the z-axis is calibrated, the x-axis and z-axis outputs converge; In step S1, the method for judging whether the MEMS accelerometer is in a horizontal state is that, when performing data processing, firstly, it is judged whether there is external calibration data. If there is external calibration data, it is necessary to compensate the external calibration data to the three-axis original data, and use the horizontal threshold 1 to judge the horizontal state of the MEMS accelerometer, that is, the data after the compensation of the x-axis and the y-axis are both less than the horizontal threshold 1, indicating that the MEMS accelerometer is in a horizontal state; if there is no external calibration data, the horizontal threshold 2 is used to judge the horizontal state of the MEMS accelerometer, that is, the original data of the x-axis and the y-axis are both less than the horizontal threshold 2, indicating that the MEMS accelerometer is in a horizontal state; In step S2, the method for determining whether the MEMS accelerometer is in a stationary state is to collect multiple sets of three-axis raw data using the MEMS accelerometer. If the sum of the changes of these three-axis raw data is less than the stationary threshold, it means that the MEMS accelerometer is in a stationary state. That is, if |X 最大值 -X 最小值 |+|Y 最大值 -Y 最小值| +|Z 最大值 -Z 最小值 |<static threshold, it means that the MEMS accelerometer meets the static condition; In step S3, the method for determining whether the MEMS accelerometer is in a normal state is to calculate the three-axis resultant force value, and the calculation formula is: ; If F is greater than the normal threshold, it means that the MEMS accelerometer is in a normal state; Otherwise, it means that the MEMS accelerometer is in an abnormal state and cannot be calibrated.

2. The MEMS accelerometer automatic calibration method according to claim 1, characterized in that: The level threshold 1 is 100 mg, and the level threshold 2 is 200 mg.

3. The MEMS accelerometer automatic calibration method according to claim 1, characterized in that: The resting threshold is 80 mg.

4. The MEMS accelerometer automatic calibration method according to claim 1, characterized in that: The normal threshold is 250 mg.

5. The MEMS accelerometer automatic calibration method according to claim 1, characterized in that: In step S5, the expected value is 100 mg.

6. The MEMS accelerometer automatic calibration method according to claim 1, characterized in that: In step S6, the output convergence value is 1 / 2 of the original data.

7. A method for automatic calibration of a MEMS accelerometer, characterized in that: It includes the following steps: S1, determine whether the MEMS accelerometer is in a horizontal state; S2. If the MEMS accelerometer is in a horizontal state, determine whether the MEMS accelerometer is in a stationary state; S3, if the MEMS accelerometer is in a stationary state, determine whether the MEMS accelerometer is in a normal state. If the MEMS accelerometer is in an abnormal state, the calibration action cannot be performed; S4. Using the characteristics of the earth's gravity, when the object is stationary, the three-axis resultant force F is the standard value of gravity acceleration. The x-axis is calibrated with the standard value, and the difference obtained is used to compensate the original data of the x-axis. S5, monitoring the x-axis compensated data in a stationary state, if the absolute value of the difference between the x-axis compensated data and the standard value is greater than the expected value, it means that the MEMS accelerometer has temperature drift, and it is necessary to trigger a recalibration action to solve the data offset caused by the temperature drift, if the absolute value of the difference between the x-axis compensated data and the standard value is less than the expected value, proceed to step S6; S6. When the MEMS accelerometer is horizontal and stationary, the output data of the y-axis and z-axis are proportionally converged, and the x-axis output data is a simulated value calculated using the resultant force formula, and the ± sign is determined by the original x-axis data; When the z-axis is vertical and stationary, and the x-axis is not calibrated, the y-axis outputs a converged value, and the x-axis output data uses the simulated value calculated using the resultant force formula, with the ± sign determined by the x-axis raw data; When the y-axis is vertical and stationary, and the x-axis is not calibrated, the z-axis outputs converged values, and the x-axis output data uses the simulated value calculated using the resultant force formula, with the ± sign determined by the x-axis raw data; When the z-axis is vertical and stationary, and the x-axis is calibrated, the y-axis and x-axis outputs converge; When the y-axis is vertical and stationary, and the x-axis is calibrated, the z-axis and x-axis outputs converge; In step S1, the method for judging whether the MEMS accelerometer is in a horizontal state is that, when performing data processing, firstly, it is judged whether there is external calibration data. If there is external calibration data, it is necessary to compensate the external calibration data to the three-axis original data, and use the horizontal threshold 1 to judge the horizontal state of the MEMS accelerometer, that is, the data after the z-axis and y-axis compensation are both less than the horizontal threshold 1, indicating that the MEMS accelerometer is in a horizontal state; if there is no external calibration data, the horizontal threshold 2 is used to judge the horizontal state of the MEMS accelerometer, that is, the z-axis and y-axis original data are both less than the horizontal threshold 2, indicating that the MEMS accelerometer is in a horizontal state; In step S2, the method for determining whether the MEMS accelerometer is in a stationary state is to collect multiple sets of three-axis raw data using the MEMS accelerometer. If the sum of the changes of these three-axis raw data is less than the stationary threshold, it means that the MEMS accelerometer is in a stationary state. That is, if |X 最大值 -X 最小值 |+|Y 最大值 -Y 最小值| +|Z 最大值 -Z 最小值 |<static threshold, it means that the MEMS accelerometer meets the static condition; In step S3, the method for determining whether the MEMS accelerometer is in a normal state is to calculate the three-axis resultant force value, and the calculation formula is: ; If F is greater than the normal threshold, it means that the MEMS accelerometer is in a normal state; Otherwise, it means that the MEMS accelerometer is in an abnormal state and cannot be calibrated.

8. A method for automatic calibration of a MEMS accelerometer, characterized in that: It includes the following steps: S1, determine whether the MEMS accelerometer is in a horizontal state; S2. If the MEMS accelerometer is in a horizontal state, determine whether the MEMS accelerometer is in a stationary state; S3, if the MEMS accelerometer is in a stationary state, determine whether the MEMS accelerometer is in a normal state. If the MEMS accelerometer is in an abnormal state, the calibration action cannot be performed; S4. Using the characteristics of the earth's gravity, when the object is stationary, the three-axis resultant force F is the standard value of gravity acceleration. The y-axis is calibrated with the standard value, and the difference obtained is used to compensate the original data of the y-axis. S5, monitoring the data after y-axis compensation in a static state, if the absolute value of the difference between the data after y-axis compensation and the standard value is greater than the expected value, it means that the MEMS accelerometer has temperature drift, and it is necessary to trigger a recalibration action to solve the data offset caused by the temperature drift, if the absolute value of the difference between the data after y-axis compensation and the standard value is less than the expected value, proceed to step S6; S6. When the MEMS accelerometer is horizontal and stationary, the output data of the x-axis and z-axis are proportionally converged, and the y-axis output data is a simulated value calculated using the resultant force formula, and the ± sign is determined by the original data of the y-axis; When the x-axis is vertical and stationary, and the y-axis is not calibrated, the z-axis outputs a converged value, and the y-axis output data uses the simulated value calculated using the resultant force formula, with the ± sign determined by the y-axis raw data; When the z-axis is vertical and stationary, and the y-axis is not calibrated, the x-axis outputs a converged value, and the y-axis outputs a simulated value calculated using the resultant force formula, with the ± sign determined by the y-axis raw data; When the x-axis is vertical and stationary, and the y-axis is calibrated, the y-axis and z-axis outputs converge; When the z-axis is vertical and stationary, and the y-axis is calibrated, the x-axis and y-axis outputs converge; In step S1, the method for judging whether the MEMS accelerometer is in a horizontal state is to first determine whether there is external calibration data when processing data. If there is external calibration data, the external calibration data needs to be compensated to the three-axis original data, and the horizontal state of the MEMS accelerometer is judged by using the horizontal threshold 1, that is, the data after the x-axis and z-axis compensation are both less than the horizontal threshold 1, indicating that the MEMS accelerometer is in a horizontal state; if there is no external calibration data, the horizontal threshold 2 is used to judge the horizontal state of the MEMS accelerometer, that is, the original data of the x-axis and the z-axis are both less than the horizontal threshold 2, indicating that the MEMS accelerometer is in a horizontal state; In step S2, the method for determining whether the MEMS accelerometer is in a stationary state is to collect multiple sets of three-axis raw data using the MEMS accelerometer. If the sum of the changes of these three-axis raw data is less than the stationary threshold, it means that the MEMS accelerometer is in a stationary state. That is, if |X 最大值 -X 最小值 |+|Y 最大值 -Y 最小值| +|Z 最大值 -Z 最小值 |<static threshold, it means that the MEMS accelerometer meets the static condition; In step S3, the method for determining whether the MEMS accelerometer is in a normal state is to calculate the three-axis resultant force value, and the calculation formula is: ; If F is greater than the normal threshold, it means that the MEMS accelerometer is in a normal state; Otherwise, it means that the MEMS accelerometer is in an abnormal state and cannot be calibrated.

Images