Angular motion measurement method combining an angular acceleration sensor with a gyroscope

Abstract

The application discloses an angular motion measurement method of an angular acceleration sensor combined with a gyroscope, which considers different error characteristics of the angular acceleration sensor and the gyroscope, designs two different sensor data combination schemes, and performs information fusion on angular acceleration and angular velocity at a posture layer and a data layer respectively, so that the precision of angular velocity measurement is improved, and the precision of carrier posture calculation is improved; compared with a method of improving the precision of angular velocity measurement by using a gyroscope array, the method has simpler installation mode and stronger adaptability to the environment. Meanwhile, the introduced angular acceleration sensor has low cost and is convenient for large-scale application.

Description

Technical Field

[0001] This invention belongs to the field of automatic control technology, specifically relating to a method for measuring angular motion using a combination of an angular acceleration sensor and a gyroscope. Background Technology

[0002] Attitude estimation technology is crucial and fundamental for path planning and motion control of vehicles such as autonomous vehicles and robots. Therefore, researchers have conducted extensive studies on how to obtain accurate attitude information of these vehicles. In the field of attitude calculation, gyroscopes, as the most widely used sensors, can be used to integrate the angular velocities they acquire to obtain the attitude angle data of the vehicle's motion.

[0003] Laser and fiber optic gyroscopes offer high precision, and integration can yield highly accurate attitude calculations. However, these gyroscopes are expensive and bulky, hindering large-scale applications. MEMS gyroscopes are small, low-power, low-cost, and easy to mass-produce, leading to their increasing use in autonomous driving and robotics. However, their precision is limited, and errors can be significant over long periods. Meanwhile, angular acceleration sensors, as inertial devices, can directly acquire angular acceleration data of the vehicle's motion. Their working principle and error characteristics differ from gyroscopes. The acquired angular acceleration data can be integrated twice to obtain attitude information of the vehicle's motion. Therefore, a combination of angular acceleration sensors and gyroscopes can be used, leveraging the complementary characteristics of the two sensors to improve the accuracy of angular motion measurements. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide an angular motion measurement method that combines an angular acceleration sensor and a gyroscope, which can improve the accuracy of angular motion measurement by fusing information from both sensors.

[0005] An angular motion measurement method combining an angular acceleration sensor and a gyroscope, wherein an attitude layer information fusion method or a data layer information fusion method can be selected as needed to measure angular motion;

[0006] The attitude layer information fusion method includes:

[0007] Step 11: Using a gyroscope and an angular acceleration sensor, collect angular velocity and angular acceleration data of the carrier during its motion, and remove random noise and spike noise from the data by filtering.

[0008] Step 12: Align the data from the gyroscope and angular accelerometer;

[0009] Step 13: For the gyroscope, the quaternion method is used to describe the carrier's attitude angles. Then, using the angular velocity data obtained in Step 12, the attitude update differential equation is solved to obtain the carrier's attitude angle data ψ. gyro;

[0010] Step 14: For the angular acceleration sensor, integrate the angular acceleration data obtained in step 12 to obtain the angular velocity data at the corresponding moment;

[0011] Step 15: The quaternion method is used to describe the carrier's attitude angles. Then, using the angular velocity data obtained in Step 14, the attitude update differential equation is solved to obtain the carrier's attitude angle data ψ. AA ;

[0012] Steps 16 and 13 utilize the attitude angle data ψ obtained from the gyroscope. gyro And the attitude angle data ψ obtained by the angular acceleration sensor in step 15 AA The measurement equations for the extended Kalman filter are constructed as follows:

[0013] z att =ψ gyro -ψ AA =H att ·δx att +v att

[0014] Among them, z att The observable δx in the Kalman filter equation is... att H is the state variable in the Kalman filter equation. att It is the observation matrix of the Kalman filter equation, v att It is the observation noise of the Kalman filter equation;

[0015] Step 17: Using the measurement equation for attitude angles constructed in Step 16, perform Kalman filtering to obtain an error estimate for attitude angle measurement; use the filtering result to perform error feedback correction on the attitude angle data obtained based on the gyroscope; repeat steps 11 to 17 to obtain continuous angular motion data of the carrier motion.

[0016] The data layer information fusion method includes:

[0017] Step 21: Using a gyroscope and an angular acceleration sensor, collect the angular velocity ω of the carrier during its motion. gyro Combine angular acceleration data and remove random noise and spike noise from the data;

[0018] Step 22: Align the data from the gyroscope and angular accelerometer;

[0019] Step 23: For the angular acceleration sensor, first integrate the angular acceleration data obtained in step 22 to obtain the angular velocity data ω at the corresponding moment. AA ;

[0020] Step 24: Based on the angular velocity data ω obtained by integrating angular acceleration in step 23. AA And the aligned angular velocity data ω from step 22 gyro The measurement equations for the extended Kalman filter are constructed as follows:

[0021] z ang =ω gyro -ω AA =H ang ·δx ang +v ang

[0022] Among them, z ang The observable δx in the Kalman filter equation is... ang H is the state variable in the Kalman filter equation. ang It is the observation matrix of the Kalman filter equation, v ang It is the observation noise of the Kalman filter equation;

[0023] Step 25: Using the measurement equation for angular velocity constructed in Step 24, perform Kalman filtering to obtain an error estimate for angular velocity measurement; use the filtering result to perform error feedback correction on the angular velocity data obtained based on the gyroscope.

[0024] Step 26: Use the quaternion method to describe the carrier attitude angle, and then use the angular velocity data obtained in step 25 to solve the attitude update differential equation to obtain the attitude angle data ψ of the carrier motion; repeat steps 21 to 26 to obtain continuous angular motion data of the carrier motion.

[0025] The present invention has the following beneficial effects:

[0026] This invention, based on angular velocity data measured by a gyroscope, introduces angular acceleration data directly measured by an angular acceleration sensor, and designs an information fusion method between the angular acceleration sensor and the gyroscope. This method considers the different error characteristics of the angular acceleration sensor and the gyroscope, and designs two different sensor data combination schemes to fuse angular acceleration and angular velocity information at the attitude layer and data layer respectively, improving the accuracy of angular velocity measurement and thus improving the accuracy of carrier attitude calculation. Compared with the method of using a gyroscope array to improve angular velocity measurement accuracy, the method of this invention is simpler to install and more adaptable to the environment. At the same time, the introduced angular acceleration sensor has a lower cost, facilitating large-scale application. Attached Figure Description

[0027] Figure 1 This is a flowchart of the combination of angular acceleration sensor and gyroscope attitude layer of the present invention.

[0028] Figure 2This is a flowchart of the combination of the angular acceleration sensor and the gyroscope data layer of the present invention.

[0029] Figure 3 The results show the attitude calculations of the method of this invention and a pure gyroscope in three axes.

[0030] Figure 4 The results show the attitude calculation errors of the method of this invention and a pure gyroscope in three axes. Detailed Implementation

[0031] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0032] To address the issue of error divergence in existing low-cost gyroscopes during long-term operation, this invention utilizes angular acceleration sensors with different error characteristics to fuse angular acceleration data with angular velocity data. This invention provides an angular motion measurement method combining an angular acceleration sensor and a gyroscope, encompassing two different combination modes: an attitude layer and a data layer. The attitude layer combination mode allows the two sensors to operate independently, simplifying implementation and reducing susceptibility to sensor malfunctions. The data layer combination mode combines data at the raw data level, eliminating raw data errors and achieving higher accuracy. The two combination modes can be selected based on actual operating conditions. The specific processes of the attitude layer and data layer combination modes are as follows: Figure 1 and Figure 2 As shown, it includes:

[0033] Step 11: Using a gyroscope and an angular acceleration sensor, collect angular velocity and angular acceleration data of the carrier during its motion, and remove random noise and spike noise from the original sensor data by filtering.

[0034] Step 12: Since the sampling frequencies of the two inertial sensors are different, data alignment of the two sensor data in step 11 is achieved by data interpolation.

[0035] for Figure 1 The illustrated combination mode of the angular acceleration sensor and gyroscope attitude layer shows that the two sensors work independently to calculate the carrier attitude, and a Kalman filter is used to combine the carrier attitude calculated by the two sensors. The steps are as follows:

[0036] Step 13: For the gyroscope, the quaternion method is used to describe the carrier's attitude angles. Then, using the angular velocity data obtained in Step 12, the attitude update differential equation is solved to obtain the carrier's attitude angle data ψ. gyro .

[0037] Step 14: For the angular acceleration sensor, first integrate the angular acceleration data obtained in Step 12 to obtain the angular velocity data at the corresponding moment, as shown in the following formula:

[0038]

[0039] Where ΔT is the discrete sampling period of the angular acceleration sensor. The data are angular accelerations at time k along three axes. These are the angular velocity data at time k along the three axes.

[0040] Step 15: The quaternion method is also used to describe the carrier's attitude angles. Then, using the angular velocity data obtained in Step 14, the attitude update differential equation is solved to obtain the carrier's attitude angle data ψ. AA .

[0041] Steps 16 and 13 utilize the attitude angle data ψ obtained from the gyroscope. gyro And the attitude angle data ψ obtained by the angular acceleration sensor in step 15 AA Both have the same physical meaning, but different data sources. Therefore, subtracting them constructs the measurement equation for the extended Kalman filter, as shown in the following formula:

[0042] z att =ψ gyro -ψ AA =H att ·δx att +v att

[0043] Among them, z att The observable δx in the Kalman filter equation is... att H is the state variable in the Kalman filter equation. att It is the observation matrix of the Kalman filter equation, v att This is the observation noise in the Kalman filter equation.

[0044] Step 17: Using the measurement equation for the attitude angle constructed in Step 16, perform Kalman filtering to obtain an error estimate for the attitude angle measurement. Use the filtering result to perform error feedback correction on the attitude angle data acquired based on the gyroscope. Repeating steps 11 to 17 will obtain continuous angular motion data of the carrier motion.

[0045] for Figure 2 The angular acceleration sensor and gyroscope data layer combination mode shown directly fuses the information from the two sensors at the raw data level using a Kalman filter to obtain more accurate angular velocity data. This data is then further used to calculate the carrier's attitude. The steps are as follows:

[0046] Step 21: Using a gyroscope and an angular acceleration sensor, collect angular velocity and angular acceleration data of the carrier during its motion, and remove random noise and spike noise from the original sensor data by filtering.

[0047] Step 22: Since the sampling frequencies of the two inertial sensors are different, data alignment of the two sensor data in step 21 is achieved by data interpolation.

[0048] Step 23: For the angular acceleration sensor, first integrate the angular acceleration data obtained in step 22 to obtain the angular velocity data at the corresponding moment, as shown in the following formula:

[0049] ω AA (k)=ω AA (k-1)+α AA (k)ΔT

[0050] Where ΔT is the discrete sampling period of the angular acceleration sensor, and α AA (k) represents the angular acceleration data collected at time k, ω AA (k) represents the angular velocity data at time k.

[0051] The angular velocity data ω obtained by integrating angular acceleration in steps 24 and 23. AA And the angular velocity data ω acquired by the gyroscope in step 22. gyro Both have the same physical meaning, but different data sources. Therefore, subtracting them constructs the measurement equation for the extended Kalman filter, as shown in the following formula:

[0052] z ang =ω gyro -ω AA =H ang ·δx ang +v ang

[0053] Among them, z ang The observable δx in the Kalman filter equation is... ang H is the state variable in the Kalman filter equation. ang It is the observation matrix of the Kalman filter equation, v ang This is the observation noise in the Kalman filter equation.

[0054] Step 25: Using the measurement equation for angular velocity constructed in Step 24, perform Kalman filtering to obtain an error estimate for the angular velocity measurement. Use the filtering result to perform error feedback correction on the angular velocity data acquired from the gyroscope, improving the measurement accuracy of the angular velocity data.

[0055] Step 26: The quaternion method is used to describe the carrier's attitude angles. Then, using the angular velocity data obtained in step 25, the attitude update differential equation is solved to obtain the carrier's attitude angle data ψ. Repeating steps 21 to 26 will obtain continuous angular motion data of the carrier.

[0056] This invention underwent experimental testing on a standard laboratory three-axis turntable to verify its accuracy. During the experiment, the turntable was equipped with an inertial navigation system including a low-cost gyroscope and an angular acceleration sensor. Angular motion data was acquired by controlling the rotation of the turntable in three axes. A high-precision laser inertial navigation system was also mounted on the turntable to provide the true angular motion values. The measurement results of the pure gyroscope measurement method and the angular motion measurement method (data layer) combining the angular acceleration sensor and gyroscope proposed in this invention are as follows: Figure 3 and Figure 4 As shown, the specific error indicators are as follows:

[0057]

[0058]

[0059] Experimental results show that, compared with the attitude calculation results of a pure gyroscope, the angular motion measurement method combining an angular acceleration sensor and a gyroscope proposed in this invention reduces the maximum residual of the attitude angle calculation results by 35.04% and the average residual of the attitude angle calculation results by 20.81%, thereby improving the accuracy of angular motion measurement.

[0060] In summary, the above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method of measuring angular motion by combining an angular acceleration sensor with a gyroscope, characterized by, According to the need, the attitude layer information fusion method or the data layer information fusion method is selected to measure the angular motion; The attitude layer information fusion method comprises: Step 11, collecting the angular velocity and angular acceleration data of the carrier motion by using the gyroscope and the angular acceleration sensor respectively, and removing the random noise and the peak noise in the data by filtering; Step 12, aligning the data of the gyroscope and the angular acceleration sensor; Step 13, for the gyroscope, the quaternion method is used to describe the carrier attitude angle, and then the angular velocity data obtained in step 12 is used to solve the attitude update differential equation to obtain the attitude angle data of the carrier motion ; Step 14, for the angular acceleration sensor, the angular acceleration data obtained in step 12 is integrated to obtain the angular velocity data at the corresponding moment; Step 15, describe the carrier attitude angle by using quaternion method, then solve the attitude update differential equation by using the angular velocity data obtained in step 14 to obtain the attitude angle data of the carrier motion ; Step 16, step 13 uses the attitude angle data obtained by the gyroscope and step 15 uses the attitude angle data obtained by the angular acceleration sensor , the measurement equation is constructed as an extended Kalman filter, and the formula is as follows: wherein, is an observation of the attitude layer fusion Kalman filter equation, is a state quantity of the attitude layer fusion Kalman filter equation, is an observation matrix of the attitude layer fusion Kalman filter equation, is an observation noise of the attitude layer fusion Kalman filter equation; Step 17, performing Kalman filtering by using the measurement equation about the attitude angle constructed in step 16 to obtain the error estimation of the attitude angle measurement; the error feedback correction is performed on the attitude angle data obtained based on the gyroscope by using the filtering result; and steps 11 to 17 are repeated to obtain the continuous angular motion data of the carrier motion; The data layer information fusion method comprises: Step 21, collecting the angular velocity and angular acceleration data of the carrier motion by using the gyroscope and the angular acceleration sensor respectively, and removing the random noise and the peak noise in the data; Step 22, aligning the data of the gyroscope and the angular acceleration sensor; Step 23, for the angular acceleration sensor, first utilize the angular acceleration data obtained in step 22 to integrate to obtain the angular velocity data at the corresponding time ; Step 24, based on the angular velocity data acquired by integrating the angular acceleration in step 23 , and the aligned angular velocity in step 22 , construct the measurement equation of the extended Kalman filter, the formula is as follows: wherein, is an observation of the data layer fusion Kalman filter equation, is a state quantity of the data layer fusion Kalman filter equation, is an observation matrix of the data layer fusion Kalman filter equation, is an observation noise of the data layer fusion Kalman filter equation; Step 25, performing Kalman filtering by using the measurement equation about the angular velocity constructed in step 24 to obtain the error estimation of the angular velocity measurement; and the error feedback correction is performed on the angular velocity data obtained based on the gyroscope by using the filtering result. Step 26, describe the carrier attitude angle by using quaternion method, and then solve the attitude update differential equation by using the angular velocity data obtained in step 25 to obtain the attitude angle data of the carrier motion ; repeat steps 21 to 26 to obtain continuous angular motion data of the carrier motion.

Images