Pedestrian Navigation Method Using Consumer-Grade IMU with Badge Type

Abstract

The present invention belongs to the technical field of micro-electromechanical systems, and specifically discloses a badge-type consumer-grade IMU pedestrian navigation method, including: determining whether a pedestrian is in a static state; obtaining the pedestrian's motion parameters at the current moment, correcting the gyroscope and accelerometer, clearing the pedestrian's speed and performing S5; obtaining the angular velocity measured by the corrected gyroscope, and projecting the acceleration measured by the corrected accelerometer to a navigation system, and updating the pedestrian's motion parameters at the current moment according to the acceleration under the navigation system; compensating and updating the gyroscope correction according to the acceleration calculation under the navigation system obtained in S3, and performing S5; the method has the following advantages: the inertial navigation method based on the MEMS inertial measurement unit (MIMU) has the advantages of high-precision positioning effect in a short-flight scenario, is independent of external information, and is not easily affected by interference, which makes up for the shortcomings of radio positioning and is combined with the radio positioning method to improve the accuracy of pedestrian navigation positioning in an indoor environment.

Description

Technical Field

[0001] The present invention relates to the technical field of micro-electromechanical systems, and in particular to a badge-type consumer-grade IMU pedestrian navigation method. Background Art

[0002] With the rapid development of communications and computer technologies, the demand for location-based services continues to increase. This has led to an ever-expanding range of applications for location-based services, encompassing a wide range of fields and critical systems, including military, healthcare, manufacturing, and logistics. Furthermore, as urban buildings become increasingly complex and large-scale, indoor spaces are the primary location for human activity, accounting for approximately 80% of the time spent indoors. Therefore, accurately locating indoor users and enabling collaborative navigation has become a hot topic for seamless indoor and outdoor collaborative navigation.

[0003] Because satellite signals are extremely weak when reaching the ground over long distances, and are affected by building obstructions and multipath effects, GNSS positioning accuracy is drastically reduced or even unusable, failing to meet the needs of indoor location services. Currently, domestic and foreign researchers have proposed indoor positioning technologies and application systems based on Bluetooth (BLE), wireless network (WLAN), ultra-wideband (UWB), infrared, radio frequency technology (RFID), ultrasound, laser, and visual SLAM to address complex indoor environments. However, these technologies are often faced with a dilemma between cost and accuracy, and are susceptible to non-line-of-sight (NLOS) errors and multipath effects in the environment, resulting in a sharp decline in accuracy in complex indoor environments.

[0004] Therefore, to address the problem of accuracy degradation in radio positioning caused by multipath effects and non-line-of-sight errors indoors, a badge-type consumer-grade IMU pedestrian navigation method is proposed. Summary of the Invention

[0005] The present invention aims to provide a badge-type consumer-grade IMU pedestrian navigation method to solve or improve the above-mentioned problems of multipath effect and non-line-of-sight error of existing radio positioning in complex indoor environments.

[0006] In view of this, a first aspect of the present invention is to provide a pedestrian navigation method using a badge-type consumer-grade IMU.

[0007] A first aspect of the present invention provides a badge-type consumer-grade IMU pedestrian navigation method, which obtains the pedestrian's angular velocity and acceleration through a gyroscope and accelerometer for navigation. The navigation method includes the following steps: S1: obtaining the pedestrian's current angular velocity and acceleration, and judging whether the pedestrian is in a static state based on a preset judgment threshold. If so, proceed to S2; if not, proceed to S3; S2: obtaining the pedestrian's motion parameters at the current moment, correcting the gyroscope and accelerometer according to the pedestrian's angular velocity and acceleration when in a static state, clearing the pedestrian's speed and proceeding to S5; S3: obtaining the angular velocity measured by the corrected gyroscope, and projecting the acceleration measured by the corrected accelerometer to a navigation system based on the angular velocity, and updating the pedestrian's motion parameters at the current moment based on the acceleration in the navigation system; S4: calculating the deflection between the gravitational acceleration and the average acceleration in the navigation system over a period of time greater than one step period of the pedestrian, and compensating and updating the gyroscope based on the deflection; S5: outputting the motion parameters, navigating according to the motion parameters, and returning to S1.

[0008] The badge-type consumer-grade IMU pedestrian navigation method provided by the present invention determines the pedestrian's state through a preset judgment threshold, which is respectively a stationary state and a non-stationary state. Given the periodicity of a pedestrian's gait, compensation is performed based on the pedestrian's single-step motion cycle. Specifically, the gyroscope and accelerometer are corrected based on the pedestrian's angular velocity and acceleration when the pedestrian is stationary. Error compensation is performed by comparing the projected mean value of the ratio of the single-step cycle time (the time it takes to take one step during walking) in the navigation system with the nonlinearity of the gravitational acceleration.

[0009] Compared with the traditional complementary filtering method, the proposed improved complementary filtering method effectively filters out the random noise of the specific force itself according to the periodic motion law of pedestrians, and compensates for the attitude error when measuring the acceleration projection by observing the gravity acceleration accelerometer, thereby improving the pedestrian navigation accuracy.

[0010] In addition, the technical solution provided by the embodiment of the present invention may also have the following additional technical features:

[0011] In any of the above technical solutions, the motion parameters include: the pedestrian's speed, displacement and motion posture.

[0012] In any of the above technical solutions, before the pedestrian's speed is cleared, S2 also includes: calculating the horizontal attitude angle based on the acquired acceleration; calculating the pedestrian's quaternion at the current moment through the horizontal attitude angle; and calculating the attitude matrix including attitude information based on the quaternion.

[0013] In this technical solution, since the gyroscope of consumer-grade MEMS inertial devices is not sensitive to the angular velocity of the Earth's rotation, other methods are needed to calculate the heading angle (other devices such as magnetometers are required, which are not used in this invention and are not described here). As for the horizontal attitude angle, it can be calculated using the three-axis acceleration measured by the accelerometer. That is, the three-axis acceleration measured by the accelerometer is obtained by projecting the gravity acceleration. Therefore, the horizontal attitude angle can be obtained through the projection relationship.

[0014] When the current attitude angle is obtained, the current quaternion can be calculated based on the attitude angle information for subsequent corresponding navigation solutions;

[0015] After obtaining the quaternion, the attitude matrix containing attitude information is calculated for subsequent corresponding navigation solutions.

[0016] In any of the above technical solutions, the step of projecting the acceleration measured by the corrected accelerometer to the navigation system based on the angular velocity specifically includes: calculating the angular velocity of the human body movement using the angular velocity measured by the gyroscope, specifically using the following formula: The pedestrian's quaternion at the current moment is updated using the angular velocity of the human body's movement, specifically using the following formula: Calculate and update the attitude matrix using the updated quaternion The acceleration is projected into the navigation system using the updated attitude matrix as the projection coefficient, specifically as follows: in, is the angular velocity of human motion, gB is the three-axis gyro bias obtained in the static state, ω is the measured angular velocity, α is the corrected compensation update value, b is the carrier system, n is the navigation system, is the antisymmetric matrix formed by the three-axis angular velocity, Q i is the quaternion at the i-th moment, dt is the navigation time, f n is the acceleration of the navigation system, g n is the acceleration due to gravity, is the attitude matrix, and

[0017] In this technical solution, after obtaining the determined angular velocity of motion, the quaternion is updated by integrating the actual angular velocity change, and the attitude matrix is ​​updated according to the updated quaternion;

[0018] Since the acceleration measured by the accelerometer is the acceleration of the object under the load system, and the carrier coordinate system is in motion and cannot be used as a reference for position changes, it is necessary to project the acceleration measured by the accelerometer into a fixed navigation coordinate system through the attitude matrix and remove the influence of gravity acceleration in the navigation coordinate system to obtain the actual object motion acceleration.

[0019] In any of the above technical solutions, the step of updating the motion parameters of the pedestrian at the current moment based on the acceleration in the navigation system includes: obtaining the pedestrian's velocity at the current moment by calculating the acceleration in the navigation system; obtaining the pedestrian's displacement based on the velocity at the current moment; and obtaining the motion posture at the current moment based on the updated posture matrix, and calculating using the following formula: in, is the heading angle, θ is the pitch angle, γ is the roll angle, is the principal value of the heading angle calculated by the attitude matrix, θ 主 is the principal value of the pitch angle calculated by the attitude matrix, γ 主 is the main value of the roll angle calculated by the attitude matrix, and the motion attitude includes: θ and γ.

[0020] In this technical solution, after obtaining the posture matrix, the previous posture information of the object can be calculated by the following formula, that is, the rotation transformation relationship between the carrier coordinate system where the object is located and the fixed navigation coordinate system.

[0021] In any of the above technical solutions, the step S4 includes: cross-multiplying the mean value of the specific force projection with the gravitational acceleration in the navigation system to obtain the error compensation of the pedestrian in the non-static state, specifically: The error compensation is projected onto the carrier system through the current attitude matrix, specifically: The gyroscope’s corrected compensation update value α is obtained through error compensation calculation under the load system, specifically: Among them, s1 is the number of data in 500ms, e b is the compensation vector under the load system, is the pitch angle compensation value, is the roll angle compensation value, is the heading angle compensation value, is the mean value of the projection of the comparison force in the navigation system within a period of time t, e n is the non-collinearity error, The calculation example of the feedback coefficient is K = [0.00001 0.00001 0.00001].

[0022] In this technical solution, since the error of inertial navigation will accumulate and diverge over time under long-term working conditions, and the acceleration changes and fluctuates greatly in the acceleration stage at the beginning of the movement and the deceleration stage before the end of the movement, the acceleration error is the dominant factor at this time. Therefore, the traditional complementary filtering method is improved here to address this problem, that is, the projection mean of the acceleration information measured by the accelerometer within 500ms in the navigation system is compared with the gravity acceleration in the navigation system. The degree of non-collinearity of these two vectors reflects the error caused by the navigation error accumulated over time and the acceleration fluctuation. This error is mainly caused by inaccurate posture. Therefore, the angular error formed by the non-collinearity is used as a zero bias compensation closed loop in the solution to eliminate the posture error. The feedback coefficient reflects the error influence caused by the acceleration and deceleration stages, and suppresses the acceleration and deceleration errors, thereby realizing error compensation in pedestrian navigation and improving navigation accuracy.

[0023] The non-collinear error between the projection mean of the specific force in the navigation system and the gravitational acceleration needs to be closed into the navigation through a feedback coefficient. The feedback coefficient mainly reflects the idea of ​​pattern recognition. The error size under the acceleration and deceleration state is judged through the optimization scheme, and the feedback coefficient size is designed accordingly to achieve the best compensation effect.

[0024] In any of the above technical solutions, the judgment threshold includes an angular velocity threshold and an acceleration threshold, and the judgment is made using the following rules: Among them, j belongs to [1, s2], f j is the current accelerometer measurement value, g is the acceleration of gravity, ω j is the gyro measurement value, ε acc The acceleration condition for determining whether it is static is 0.5m / s 2 , ε gyro The angular velocity condition for determining whether it is static is 1.5° / s.

[0025] In this technical solution, the average of the measurement values ​​of the MEMS gyroscope and MEMS accelerometer within 100ms is selected to remove a certain amount of random noise interference.

[0026] In any of the above technical solutions, the gyroscope is corrected as follows: when the pedestrian is in a static state, the mean of the three-axis angular velocity measured by the gyroscope is converted into deviations in the three-axis directions; the deviations in the three-axis directions are aggregated as the three-axis gyroscope zero bias of the pedestrian in the current static state, and the three-axis gyroscope zero bias is eliminated and corrected in the next non-static angular velocity measurement of the pedestrian; and the three-axis gyroscope zero bias is calculated using the following formula:

[0027] In this technical solution, the average of the three-axis angular velocity measured by the gyroscope is input during the static process. This average is the three-axis constant bias of the gyroscope. This bias reflects that when the object is stationary, it still maintains a certain angular velocity in motion. This non-actual motion is not what we want to exist. Therefore, in the static state, this part of the bias is calculated, and then this part of the interference is removed in the dynamic state.

[0028] In any of the above technical solutions, the accelerometer is corrected by: when the pedestrian is in a static state, obtaining the three-axis acceleration measured by the accelerometer; comparing the three-axis acceleration with the gravity acceleration to obtain the deviation on each axis; correcting the scale of the accelerometer based on the deviation on each axis; and calculating using the following formula: where BD_f is the calculated accelerometer scale.

[0029] In this technical solution, due to the scaling error of the accelerometer device itself, there is a scaling error between the measured acceleration and the actual acceleration. However, since an object is only affected by gravity when it is static, the error caused by the scaling problem can be eliminated by comparing the three-axis acceleration measured by the static accelerometer with the acceleration due to gravity.

[0030] In any of the above technical solutions, the horizontal attitude angle is calculated based on the acquired acceleration using the following formula: Among them, mean_f x ,mean_f y ,mean_f z are the current three-axis measured acceleration averages.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] Given the periodicity of pedestrian gait, this method uses the pedestrian's single-step motion cycle as a premise. The method uses the non-collinearity between the mean value of the acceleration projection in the navigation system and the gravity acceleration within a period greater than the single-step motion cycle to compensate for errors. Compared with traditional foot-worn solutions, this method avoids the posture divergence problem caused by the high dynamic angular velocity of the foot contacting the ground during walking by wearing a badge. It also suppresses posture errors through an improved complementary filtering method, thereby improving navigation accuracy.

[0033] Compared with the traditional complementary filtering method, the improved complementary filtering method proposed here effectively filters out the random noise of the specific force itself based on the periodic motion law of pedestrians. At the same time, it utilizes the characteristic that the acceleration changes periodically within a time greater than one step, and observes the projection error of the measured acceleration in the navigation system caused by the attitude error through the gravity acceleration observation, that is, the degree of non-collinearity of the projection of the gravity acceleration and the measured acceleration in the navigation system to compensate for the attitude error, thereby improving the navigation positioning accuracy of pedestrians.

[0034] Additional aspects and advantages of embodiments according to the present invention will become apparent in the following description or may be learned through practice of embodiments according to the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments with reference to the accompanying drawings, in which:

[0036] Figure 1 This is a flow chart of the pedestrian navigation method of the present invention. DETAILED DESCRIPTION

[0037] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0038] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0039] See also Figure 1 , the following describes some embodiments of the present invention's badge-type consumer-grade IMU pedestrian navigation method.

[0040] The embodiment of the first aspect of the present invention proposes a badge-type consumer-grade IMU pedestrian navigation method. In some embodiments of the present invention, such as Figure 1 As shown, a badge-type consumer-grade IMU pedestrian navigation method is provided, and the badge-type consumer-grade IMU pedestrian navigation method includes:

[0041] Pedestrian navigation methods are mainly divided into two parts: static and dynamic. The dynamic and static judgments are based on the current state of the object. For example, in a stationary state, the combined acceleration signal should be the acceleration of the earth's gravity, and the angular velocity signal should be set according to the actual noise parameters of the device. The specific parameter design can be referred to as follows. The static part mainly functions as initial alignment and error correction in inertial navigation. The dynamic part calculates the corresponding attitude angle change, velocity, and position change based on the actual angular velocity and acceleration data measured by the MEMS gyroscope and MEMS accelerometer during motion.

[0042] First, it is necessary to determine whether the pedestrian is in a static state. The static and dynamic judgment conditions are: select the average measurement value of the MEMS gyroscope and MEMS accelerometer for 100ms, and remove a certain amount of random noise interference. The specific parameter setting examples are as follows.

[0043]

[0044] Where f is the current accelerometer measurement value, ω is the gyroscope measurement value, and ε acc To determine whether it is a static acceleration condition, ε gyro The angular velocity condition for determining whether it is static.

[0045] Static part: In the present invention, the input of the static part is the three-axis angular velocity and acceleration average of the gyroscope and accelerometer during the static process.

[0046]

[0047] where mean_f x ,mean_f y ,mean_f z is the current three-axis measured acceleration mean, mean_ω x ,mean_ω y ,mean_ω z The current average angular velocity of the three axes.

[0048] Step 1-1: Calculate the three-axis gyroscope bias. This involves calculating the average of the three-axis angular velocities measured by the gyroscope during static conditions. This average is the gyroscope's three-axis constant bias. This bias reflects the fact that an object maintains a certain angular velocity while in motion, even when stationary. This undesirable non-actual motion is therefore undesirable. Therefore, this bias is calculated during static conditions to remove this interference during dynamic conditions.

[0049]

[0050] Where gB is used to calculate the zero bias of the three-axis gyroscope.

[0051] Step 1-2: Calculate the accelerometer scale. Due to scaling errors within the accelerometer, there's a scaling error between the measured acceleration and the true acceleration. However, since an object is only affected by gravity when it's static, the error caused by scaling can be eliminated by comparing the three-axis acceleration measured by the accelerometer at static conditions with the acceleration due to gravity.

[0052]

[0053] where BD_f is the calculated accelerometer scale.

[0054] Steps 1-3: Calculate the horizontal attitude angle. Because the gyroscopes in consumer-grade MEMS inertial devices are not sensitive to the Earth's rotational angular velocity, other methods are needed to calculate the heading angle. The horizontal attitude angle can be calculated using the three-axis acceleration measured by the accelerometer. This is because the three-axis acceleration measured by the accelerometer is projected from the acceleration of gravity. Therefore, the horizontal attitude angle can be calculated using the projection relationship. The calculation formula is as follows:

[0055]

[0056] where mean_f x ,mean_f y ,mean_f z is the current three-axis measured acceleration mean, is the attitude matrix, θ is the pitch angle, and γ is the roll angle.

[0057] Step 1-4: Calculate the initial quaternion. When obtaining the current horizontal attitude angle, assume that the heading angle is The current quaternion can be calculated based on the attitude angle information for subsequent corresponding navigation solutions.

[0058]

[0059] Where Q is a quaternion, q x Represents the x-th element in a quaternion.

[0060] Steps 1-5: Calculate the initial attitude matrix. After obtaining the quaternion, calculate the attitude matrix containing attitude information for subsequent corresponding navigation solutions.

[0061]

[0062] Among them, T xy is the number in the xth row and yth column of the attitude matrix.

[0063] Step 1-6: Keep the current position unchanged and reset the velocity to zero. Since the error in inertial navigation will diverge over time during dynamic solution, the position is kept unchanged and the velocity error is reset to zero during the static state of pedestrian navigation to suppress error divergence.

[0064] P i n =P i-1 n , V i n =0;

[0065] Where P is the position, V is the velocity, the superscript n indicates the description in the navigation system, the subscript i indicates the position and velocity in the current state, and the subscript i-1 indicates the position and velocity at the previous moment.

[0066] The dynamic part includes:

[0067] Step 2-1: Calculate Angular Velocity. The key to ensuring accurate positioning and navigation during motion lies in accurate attitude angles, and the core of attitude angles is the accurate input angular velocity. Therefore, calculating angular velocity is extremely important. The three-axis angular velocity measured by the gyroscope contains errors caused by the device's inherent bias, also known as zero offset. Furthermore, since errors accumulate during motion, these errors can be compensated for using an improved complementary filtering method to eliminate some of the accumulated errors, thereby improving positioning and navigation accuracy.

[0068]

[0069] Wherein, gB is the three-axis gyro bias obtained in a static state, ω is the measured angular velocity, α is the equivalent bias value corresponding to the feedback compensation of the present invention, and b and n represent the carrier system and the navigation system, respectively.

[0070] Step 2-2: Update the quaternion and attitude matrix. After determining the angular velocity, the quaternion is updated by integrating the actual angular velocity change. The attitude matrix is ​​then updated based on the updated quaternion. This quaternion and attitude matrix contain the angular transformation relationship between the carrier system and the navigation system.

[0071]

[0072] in is the antisymmetric matrix formed by the three-axis angular velocity, and dt is the navigation time.

[0073] Step 2-3: Comparative force projection. Since the acceleration measured by the accelerometer is the actual acceleration of the object under the load system, and the carrier coordinate system is in motion and cannot be used as a reference for position changes, it is necessary to project the acceleration measured by the accelerometer into the fixed navigation coordinate system through the attitude matrix. In this navigation coordinate system, the influence of gravity acceleration is removed to obtain the actual object motion acceleration.

[0074]

[0075] where f n is the acceleration in the navigation system, g n is the acceleration due to gravity, is the posture matrix.

[0076] Step 2-4: Velocity update: After obtaining the object's actual acceleration in the navigation system, the object's velocity can be calculated by integrating the acceleration.

[0077]

[0078] Step 2-5: Position calculation: After obtaining the object's true velocity, the displacement of the object can be calculated by velocity integration.

[0079]

[0080] Step 2-6: Attitude Calculation: After obtaining the attitude matrix, the object's previous attitude information can be calculated using the following formula: that is, the rotation transformation relationship between the carrier coordinate system where the object is located and the fixed navigation coordinate system.

[0081]

[0082] Step 2-7: Compensate for errors using an improved complementary filtering method. Because inertial navigation errors accumulate and diverge over time under long-term conditions, and because acceleration fluctuates significantly during the acceleration phase at the beginning of a movement and the deceleration phase before the end, acceleration errors dominate. Therefore, an improvement to the traditional complementary filtering method is proposed to address this issue. The projected mean value of the accelerometer acceleration information measured within 500ms in the navigation frame is compared with the gravity acceleration in the navigation frame. The degree of non-collinearity between these two vectors reflects the error caused by time-accumulated navigation errors and acceleration fluctuations. This error is primarily caused by attitude inaccuracy. Therefore, the angular error caused by this non-collinearity is incorporated into the solution as a zero-bias compensation loop using a feedback coefficient. This feedback coefficient reflects the effects of the errors caused by the acceleration and deceleration phases, suppressing the acceleration and deceleration errors. This achieves error compensation in pedestrian navigation and improves navigation accuracy. The specific implementation method is as follows.

[0083] ① In the navigation system, cross-multiply the mean value of the specific force projection by the gravitational acceleration;

[0084]

[0085] in is the projection mean of the comparative force in the navigation system within 500ms, e n is the non-collinearity error.

[0086] ② Project the cross product result (i.e. error compensation angle) onto the load system;

[0087]

[0088] ③The gyro's zero bias compensation value is as follows

[0089] The non-collinear error between the projection mean of the specific force in the navigation system and the gravitational acceleration needs to be closed into the navigation through a feedback coefficient. The feedback coefficient mainly reflects the idea of ​​pattern recognition. The error size under the acceleration and deceleration state is judged through the optimization scheme, and the feedback coefficient size is designed accordingly to achieve the best compensation effect.

[0090]

[0091] Where α is the equivalent zero bias value corresponding to the feedback compensation of the present invention, The feedback coefficient can be calculated as K = [0.00001 0.00001 0.00001].

[0092] In the description of the present invention, it should be understood that the terms "longitudinal", "transverse", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention, 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 cannot be understood as a limitation on the present invention.

[0093] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by persons skilled in the art should fall within the scope of protection defined by the claims of the present invention.

Claims

1. A badge-type consumer-grade IMU pedestrian navigation method, characterized by: The angular velocity and acceleration of the pedestrian are obtained by a gyroscope and an accelerometer for navigation. The navigation method includes the following steps: S1: Obtain the pedestrian's current angular velocity and acceleration, and determine whether the pedestrian is in a static state based on a pre-set judgment threshold. If so, proceed to S2; otherwise, proceed to S3. S2: Obtain the motion parameters of the pedestrian at the current moment, correct the gyroscope and accelerometer according to the angular velocity and acceleration of the pedestrian in a static state, reset the pedestrian's speed and proceed to S5; S3: Obtaining the corrected angular velocity measured by the gyroscope, and projecting the corrected acceleration measured by the accelerometer to the navigation system according to the angular velocity, and updating the motion parameter of the pedestrian at the current moment according to the acceleration in the navigation system; S4: calculating a deflection angle between the gravitational acceleration and an average value of the acceleration in the navigation system within a period of time greater than one step period of the pedestrian, and compensating and updating the gyroscope according to the deflection angle; S5: Output motion parameters, perform navigation according to the motion parameters, and return to S1; Before clearing the pedestrian's speed, S2 further includes: Calculate the horizontal attitude angle based on the acquired acceleration; Calculate the pedestrian's quaternion at the current moment using the horizontal posture angle; Calculating a posture matrix including posture information according to the quaternion; The step of projecting the acceleration measured by the corrected accelerometer to the navigation system according to the angular velocity specifically includes: The angular velocity of human body motion is calculated by the angular velocity measured by the gyroscope, specifically using the following formula: ； The pedestrian's quaternion at the current moment is updated using the angular velocity of the human body's movement, specifically using the following formula: ； Calculate and update the attitude matrix using the updated quaternion ; The acceleration is projected into the navigation system using the updated attitude matrix as the projection coefficient, specifically as follows: ； in, is the angular velocity of human motion, is the three-axis gyro zero bias obtained in static state, To measure angular velocity, is the corrected compensation update value, For the carrier system, For navigation system, is the antisymmetric matrix formed by the three-axis angular velocity, For the i Quaternion of time, For navigation time, is the acceleration of the navigation system, is the acceleration due to gravity, is the attitude matrix, and , represents the acceleration of the load system, A vector representation of the gravitational acceleration in the navigation frame.

2. The badge-type consumer-grade IMU pedestrian navigation method according to claim 1, characterized in that: The motion parameters include: speed, displacement and motion posture of the pedestrian.

3. The badge-type consumer-grade IMU pedestrian navigation method according to claim 1, characterized in that: The step of updating the motion parameters of the pedestrian at the current moment according to the acceleration in the navigation system includes: Obtain the pedestrian's current speed through acceleration calculation under the navigation system; Obtain the pedestrian's displacement based on the current speed calculation; The motion posture at the current moment is obtained based on the updated posture matrix and is calculated using the following formula: ； in, is the heading angle, is the pitch angle, is the roll angle, is the principal value of the heading angle calculated by the attitude matrix, is the principal value of the pitch angle calculated by the attitude matrix, is the main value of the roll angle calculated by the attitude matrix, and the motion attitude includes: 、 and .

4. The badge-type consumer-grade IMU pedestrian navigation method according to claim 1, characterized in that: The steps of S4 include: In the navigation system, the mean value of the specific force projection is cross-multiplied with the gravity acceleration to obtain the error compensation of the pedestrian in the non-static state, specifically: ； The error compensation is projected onto the carrier system through the current attitude matrix, specifically: ； Obtain the corrected compensation update value of the gyroscope through error compensation calculation under the load system , specifically: ； in, s 1 is the number of data in 500ms, is the compensation vector under the load system, is the pitch angle compensation value, is the roll angle compensation value, is the heading angle compensation value, To compare the average value of the projection within 500ms in the navigation system, is the non-collinearity error, is the feedback coefficient and its calculation example is .

5. The badge-type consumer-grade IMU pedestrian navigation method according to claim 1, characterized in that: The judgment threshold includes an angular velocity threshold and an acceleration threshold, and is judged using the following rules: ； in, j belongs to [1, ]、 The data volume is 100ms, is the current accelerometer measurement value, is the gyro measurement value, The acceleration condition for determining whether it is static is: 、 The angular velocity condition for determining whether it is static is: .

6. The badge-type consumer-grade IMU pedestrian navigation method according to claim 1, characterized in that: The correction of the gyroscope is: When the pedestrian is static, the mean of the three-axis angular velocity measured by the gyroscope is converted into deviations in the three-axis directions; The deviations in the three-axis directions are aggregated as the three-axis gyroscope zero bias of the pedestrian in the current static state, and the three-axis gyroscope zero bias is eliminated and corrected in the next non-static angular velocity measurement of the pedestrian; and the three-axis gyroscope zero bias is calculated using the following formula: ； The correction for the accelerometer is: When the pedestrian is static, obtain the three-axis acceleration measured by the accelerometer; Compare the three-axis acceleration with the gravity acceleration to obtain the deviation on each axis; The accelerometer scale is corrected based on the deviation on each axis and is calculated using the following formula: ； in, To calculate the accelerometer scale, is the total amount of data at static state, j belongs to [1, ]、 The data volume is 100ms. The current accelerometer measurement value.

7. The badge-type consumer-grade IMU pedestrian navigation method according to claim 1, characterized in that: The horizontal attitude angle is calculated based on the acquired acceleration using the following formula: ； in, are the mean accelerations of the three axes at the current moment.

Images