A device and method for detecting motion state of spherical particles

Abstract

The present invention relates to a spherical particle motion state detection device and method thereof, the device comprises a particle container containing a plurality of spherical filling particles, wherein a magnetic positioning device is arranged on the wall surface of the particle container, a movable spherical tracer particle detection device is also placed in the particle container, the magnetic positioning device and the spherical tracer particle detection device are respectively connected to a computer, the spherical tracer particle detection device is used to collect its own inertial motion data, the magnetic positioning device is used to collect the internal magnetic induction intensity data of the spherical tracer particle detection device, the computer performs attitude solution according to the inertial motion data, and performs position solution according to the magnetic induction intensity, and obtains the motion state of the spherical tracer particle detection device through data fusion. Compared with the prior art, the present invention combines the inertial navigation technology with the magnetic field positioning technology, and can accurately and conveniently detect the motion state information of the spherical particles.

Description

Technical Field

[0001] The present invention relates to the technical field of spherical particle measurement, and in particular to a spherical particle motion state detection device and method. Background Art

[0002] Spherical particles are widely present in nature, industry and production life. The motion measurement of spherical fuel elements in nuclear reactors is traditionally obtained through inertial devices based on inertial navigation technology. Inertial devices are subject to the influence of factors such as processing technology during the manufacturing process, resulting in errors such as white noise in the inertial devices themselves. When the inertial device is integrated to solve the displacement, the noise will be amplified. After the white noise is integrated, it is equivalent to adding a random white noise error on the basis of the current error. This makes it easy for the inertial device to accumulate errors when solving the displacement, and it is impossible to accurately calculate the displacement through inertial navigation technology. It can be seen that the sampling frequency of inertial navigation technology is high, the sampling time is short, and there is no need to deploy a large number of sensors. Its disadvantage is poor positioning accuracy.

[0003] For the positioning technology of obtaining displacement, the magnetic field positioning technology with higher accuracy is currently based on the magnetic field. It measures the size and direction of the magnetic field through magnetic induction intensity data, and the size of the magnetic induction intensity of the magnetic object is also related to factors such as the density and volume of the magnetic material. The positioning method constructed by the magnetic dipole model and using magnetic induction intensity data has been widely used in the positioning of magnetic materials, such as seabed energy surveys. In addition, the magnetic field positioning technology that locates magnetic materials through multiple magnetometer sensor arrays is mainly used in medical and other fields. For example, the wireless capsule endoscope locates small magnets through a magnetometer sensor array to achieve capsule tracking and positioning. Although the magnetic field positioning technology has high positioning accuracy, its shortcomings are low sampling frequency and long sampling time, requiring the deployment of a large number of sensors, and relatively complex circuits. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a spherical particle motion state detection device and method. By combining inertial navigation technology with magnetic field positioning technology, the motion state information of spherical particles can be accurately and conveniently detected.

[0005] The objective of the present invention can be achieved through the following technical scheme: a spherical particle motion state detection device, comprising a particle container containing a plurality of spherical filling particles, a magnetic positioning device is arranged on the wall surface of the particle container, a movable spherical tracer particle detection device is also placed in the particle container, the magnetic positioning device and the spherical tracer particle detection device are respectively connected to a computer, the spherical tracer particle detection device is used to collect its own inertial motion data, the magnetic positioning device is used to collect the internal magnetic induction intensity data of the spherical tracer particle detection device, the computer performs posture solution according to the inertial motion data, and performs position solution according to the magnetic induction intensity, and obtains the motion state of the spherical tracer particle detection device through data fusion.

[0006] Furthermore, the spherical tracer particle detection device includes a spherical shell, in which a first microcontroller and a magnet are installed, the first microcontroller is respectively connected to a first memory, an accelerometer and an angular velocity meter, the first microcontroller is connected to a power supply through a control switch, after the control switch is closed, the first microcontroller is connected to the power supply, and the spherical tracer particle detection device starts to move in the particle container, the accelerometer and the angular velocity meter are used to collect corresponding inertial motion data, and store the data through the first memory, and the first microcontroller is communicatively connected to a computer to transmit the inertial motion data to the computer.

[0007] Furthermore, the first microcontroller, the magnet, the first memory, the accelerometer, the angular velocity meter and the control switch are all integrated and mounted on the same PCB board.

[0008] Furthermore, the magnet is specifically a cylindrical magnet.

[0009] Furthermore, the power source is specifically a button battery.

[0010] Furthermore, the magnetic positioning device includes a magnetometer group, which is connected to a first microcontroller, the first microcontroller is connected to a second memory, the second microcontroller is communicatively connected to a computer, and the magnetometer group is used to collect magnetic induction intensity data of the magnet in the spherical tracer particle detection device, store the data through the second memory, and transmit the magnetic induction intensity data to the computer through the second microcontroller.

[0011] Furthermore, the magnetometer group includes a plurality of magnetometer sensors arranged at equal intervals on the wall surface of the particle container.

[0012] A method for detecting the motion state of spherical particles comprises the following steps:

[0013] S1, the spherical tracer particle detection device moves in the particle container and collects corresponding inertial motion data;

[0014] At the same time, the magnetic positioning device collects magnetic induction intensity data inside the spherical tracer particle detection device;

[0015] S2, the spherical tracer particle detection device transmits the collected inertial motion data to the computer; at the same time, the magnetic positioning device transmits the collected magnetic induction intensity data to the computer;

[0016] S3, the computer performs attitude calculation on the inertial motion data to obtain the initial attitude angle; and performs position calculation on the magnetic induction intensity data to obtain the initial magnet direction;

[0017] S4. According to the initial attitude angle and the initial magnet direction, the computer performs unified processing of the coordinate system and the time axis for the inertial navigation and magnetic field positioning;

[0018] Afterwards, data correspondence is performed with magnetic positioning as the core;

[0019] Then the attitude angle is used instead of the magnet direction to solve the position;

[0020] Finally, data correspondence is performed with inertial navigation as the core to obtain displacement, acceleration and attitude angle information, and the differential speed and integral speed are calculated. The differential speed and integral speed are fused to obtain the speed information.

[0021] Furthermore, the step S3 specifically performs attitude calculation on the inertial motion data by using the quaternion method, and performs position calculation on the magnetic induction intensity data by using the magnetic dipole model.

[0022] Furthermore, the step S4 specifically includes the following steps:

[0023] S41. Unify the coordinate systems of the two systems based on the principles of inertial navigation and magnetic field positioning:

[0024] After completing the solution of the two systems, the coordinate systems of the two systems are unified using the initial attitude angle calculated by the inertial motion data and the initial magnet direction calculated by the magnetic induction intensity data:

[0025]

[0026]

[0027] Where ψ is the yaw angle of inertial navigation, θ is the pitch angle of inertial navigation, is the angle between the magnet and the x-axis, is the angle between the magnet and the z-axis;

[0028] S42. Unify the timeline of the two systems:

[0029] Extract the start and end times of the two system movements respectively; match the movement times of the two systems and calculate the overall movement time to form a new timeline.

[0030] t s =t is -t ie

[0031] Where, t is is the time when the movement starts, t ie is the time when the movement ends, t s is the overall movement time;

[0032] S43, data correspondence based on magnetic positioning data:

[0033] First, determine the multiple relationship between the sampling rates of the magnetic positioning system and the inertial navigation system, and use it as the data interval distance, that is, each data distance of the inertial navigation system corresponds to a data point in the magnetic positioning system:

[0034]

[0035] In the formula, K i is the sampling rate of the inertial navigation system, K m is the sampling rate of the magnetic positioning system, round means rounding to the nearest integer, k means the data interval distance, and every k points of the inertial navigation system correspond to one point of the magnetic positioning system;

[0036] S44. Replace the orientation of the magnet with the attitude angle of the inertial navigation system:

[0037] m=sinψ*cosθ

[0038] n=sinψ*sinθ

[0039] p=cosψ

[0040] In the formula, m, n, and p are the direction parameters of the magnet, among which m represents the direction between the magnet and the x-axis, n represents the direction between the magnet and the y-axis, and p represents the direction between the magnet and the z-axis. By replacing the magnet direction with the attitude angle and performing position calculation again, the positioning accuracy can be improved.

[0041] S45, using inertial navigation data as the core to perform data correspondence and calculate speed:

[0042] Since the sampling rate of the inertial navigation system is higher than that of the magnetic field positioning, after the data correspondence, some of the information corresponding to the inertial navigation data will be blank, that is, the corresponding displacement is missing; for the missing displacement data between adjacent displacements, the adjacent displacements are averaged, the acceleration between adjacent displacements is integrated twice, and the two displacements are averaged to determine the missing displacement data X between adjacent displacements:

[0043]

[0044] V a =∫∫adt

[0045] V m =dx / dt

[0046] In the formula, is the displacement mean of two adjacent points, X l is the displacement after twice integrating the acceleration between adjacent displacements, V a is the velocity after the acceleration is integrated, V m is the velocity after displacement differentiation, a is acceleration, t is time, dx is the displacement difference at adjacent moments, and dt is the time difference at adjacent moments;

[0047] S46. Combine the speeds of the two systems:

[0048] Since there are certain errors in the speeds obtained by the two methods, the speeds are fused through Kalman filtering to obtain accurate speeds. Specifically, V a and V t Take the mean as the observed value:

[0049]

[0050] Then calculate V a and V t The covariance of is , and by comparing the minimum value of the covariance, the trusted speed is selected as the state variable:

[0051] Y (k|k) =Y (k|k-1) +K g(k) *[Z k -H*Y (k|k-1) ]

[0052] Where Y (k|k) is the optimal estimate of the system state variables at time k, Z k is the observed value of the system at time k, K g(k) The Kalman gain at time k can be calculated from the covariance, H T The transposed matrix of the system observation matrix is ​​the identity matrix, Y (k|k-1)The state variable at time k predicted at time k-1;

[0053] This achieves the fusion of the two systems, that is, the corresponding acceleration, angular velocity, displacement and speed information can be obtained at every moment.

[0054] Compared with the prior art, the present invention places a movable spherical tracer particle detection device in a particle container, arranges a magnetic positioning device on the wall of the particle container, and connects the magnetic positioning device and the spherical tracer particle detection device to a computer respectively, collects its own inertial motion data by using the spherical tracer particle detection device, collects the internal magnetic induction intensity data of the spherical tracer particle detection device by using the magnetic positioning device, and then uses a computer to perform attitude calculation based on the inertial motion data, and performs position calculation based on the magnetic induction intensity, and obtains the motion state of the spherical tracer particle detection device through data fusion. Thus, by fusing the inertial measurement data and the magnetic field positioning data, the rotation, speed and displacement of the spherical particles in motion can be accurately calculated; the inaccuracy caused by the cumulative error in inertial navigation when solving the displacement is effectively solved, and the number of sensors required for the magnetic positioning system is reduced under the same positioning accuracy.

[0055] When fusing inertial measurement data and magnetic field positioning data, the present invention can further ensure the reliability of data fusion and improve the accuracy of spherical particle motion state detection results by unifying the coordinate system and the time axis, taking magnetic positioning as the core for data correspondence, replacing the magnet direction with the attitude angle, and taking inertial navigation as the core for data correspondence. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic diagram of the structure of the device of the present invention;

[0057] Figure 2 It is a schematic diagram of the structure of the spherical tracer particle detection device;

[0058] Figure 3 It is a structural schematic diagram of a magnetic positioning device;

[0059] Figure 4 It is a schematic diagram of the method flow of the present invention;

[0060] Explanation of the markings in the figure: 1. magnetometer group, 2. particle container, 3. filling particles, 4. spherical tracer particle detection device, 41. power supply, 42. first microcontroller, 43. accelerometer, 44. angular velocity meter, 45. first memory, 46. magnet, 47. control switch, 52. second microcontroller, 53. second memory. DETAILED DESCRIPTION

[0061] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0062] Example

[0063] like Figure 1 As shown, a spherical particle motion state detection device includes a particle container 2 containing a plurality of spherical filling particles 3, a magnetic positioning device (including a magnetometer group 1 and a second microcontroller 52, wherein the magnetometer group 1 includes a plurality of magnetometer sensors arranged at equal intervals on the wall of the particle container 2, and 6 magnetometer sensors are arranged in this embodiment), a movable spherical tracer particle detection device 4 is also placed in the particle container 2, the magnetic positioning device and the spherical tracer particle detection device 4 are respectively connected to a computer, the spherical tracer particle detection device 4 is used to collect its own inertial motion data, and the magnetic positioning device is used to collect the internal magnetic induction intensity data of the spherical tracer particle detection device 4, the computer performs posture solution according to the inertial motion data, and performs position solution according to the magnetic induction intensity, and obtains the motion state of the spherical tracer particle detection device 4 through data fusion.

[0064] like Figure 2 As shown, the spherical tracer particle detection device 4 includes a spherical shell, in which a first microcontroller 42 and a magnet 46 are installed. The first microcontroller 42 is respectively connected to a first memory 45, an accelerometer 43 and an angular velocity meter 44. The first microcontroller 42 can be connected and disconnected with a power supply 41 through a control switch 47. After the control switch 47 is closed, the first microcontroller 42 is connected to the power supply 41, and the spherical tracer particle detection device 4 starts to move in the particle container 2. The accelerometer 43 and the angular velocity meter 44 are used to collect corresponding inertial motion data and store the data through the first memory 45. The first microcontroller 42 is connected to a computer for communication to transmit the inertial motion data to the computer.

[0065] In actual application, the first microcontroller 42, the magnet 46, the first memory 45, the accelerometer 43, the angular velocity meter 44 and the control switch 47 are all integrated and mounted on the same PCB board. In this embodiment, the magnet 46 is specifically a cylindrical magnet, and the power supply 41 is a button battery.

[0066] like Figure 3 As shown, the magnetic positioning device includes a magnetometer group 1, which is connected to a second microcontroller 52, and the second microcontroller 52 is connected to a second memory 53. The second microcontroller 52 is communicatively connected to a computer. The magnetometer group 1 is used to collect magnetic induction intensity data of the magnet in the spherical tracer particle detection device, and store the data through the second memory 53, and transmit the magnetic induction intensity data to the computer through the second microcontroller 52.

[0067] This embodiment constructs a spherical particle motion state detection device, including 6 magnetometer sensors, an accelerometer measurement chip, an angular velocity measurement chip, two memory chips, two microcontrollers, a button battery power supply device, a control switch, a cylindrical magnet, a computer, and a container containing spherical filling particles. Then, the accelerometer, the angular velocity meter, a memory chip, a control switch, a microcontroller, and a cylindrical magnet are integrated and installed on a PCB board to obtain a spherical measurement device and used as a tracer particle;

[0068] On the surface of a container containing spherical particles, six magnetometer sensors are placed at equal intervals, and a microcontroller, a memory chip and the six magnetometer sensors are combined into a magnetic positioning device.

[0069] Turn on the control switch of the spherical tracer particles. When the tracer particles move in the container, the inertial motion data collected by the accelerometer and the angular velocity meter are saved in the corresponding memory. The magnetometer sensor collects the magnetic induction intensity of the magnet inside the tracer particles and also stores it in the corresponding memory. Finally, the inertial motion data and magnetic induction intensity data are transmitted to the computer through the USART serial port of the microcontroller. The computer performs attitude calculation based on the inertial motion data and position calculation based on the magnetic induction intensity. Finally, the data of the two systems (inertial navigation and magnetic positioning) are integrated to obtain the motion state of the spherical tracer particles.

[0070] Based on the above device, a method for detecting the motion state of spherical particles is implemented, such as Figure 4 As shown, the following steps are included:

[0071] S1. The spherical tracer particle detection device is placed in the particle container, moves downward under the drive of gravity, and collects the corresponding inertial motion data;

[0072] At the same time, the magnetic positioning device collects magnetic induction intensity data inside the spherical tracer particle detection device;

[0073] S2, the spherical tracer particle detection device transmits the collected inertial motion data to the computer; at the same time, the magnetic positioning device transmits the collected magnetic induction intensity data to the computer;

[0074] S3, the computer performs attitude calculation on the inertial motion data to obtain the initial attitude angle; and performs position calculation on the magnetic induction intensity data to obtain the initial magnet direction;

[0075] S4. According to the initial attitude angle and the initial magnet direction, the computer performs unified processing of the coordinate system and the time axis for the inertial navigation and magnetic field positioning;

[0076] Afterwards, data correspondence is performed with magnetic positioning as the core;

[0077] Then the attitude angle is used instead of the magnet direction to solve the position;

[0078] Finally, data correspondence is performed with inertial navigation as the core to obtain displacement, acceleration and attitude angle information, and the differential speed and integral speed are calculated. The differential speed and integral speed are fused to obtain the speed information.

[0079] This embodiment applies the above technical solution, which mainly includes the following processes:

[0080] 1) An inertial measurement system is formed by using inertial elements (accelerometer and angular velocity meter) to measure the acceleration and angular velocity of the tracer particle movement and store them in the corresponding memory.

[0081] 2) A magnetic positioning system is formed by using a cylindrical magnet and six magnetometer sensors, in which the cylindrical magnet and the inertial element are integrated on a PCB board. The magnetometer sensor placed on the surface of the particle container detects the change in magnetic induction intensity generated by the magnet and stores it in the corresponding memory.

[0082] 3) The inertial data and magnetic field data are uploaded to the computer through the USART serial ports of the microcontroller of the inertial measurement system and the microcontroller of the magnetic field positioning system respectively.

[0083] 4) The measured inertial data is used to calculate the attitude through the quaternion method:

[0084]

[0085] Where q is the expression of quaternion, is the skew matrix of angular velocity, w x 、w y 、w z is the measurement value of the gyroscope, I is the identity matrix, and T is the sampling time interval.

[0086] 5) The measured magnetic field data is used to calculate the position using the magnetic dipole model:

[0087]

[0088] In the formula, B t is a constant related to the magnet itself; H0 is the magnetization direction of the magnet; R l Indicates the distance from the magnet to the sensor, X l is the distance vector from the magnet to the sensor. The least squares error equation is constructed with formula (2), and the position solution is realized through the optimization algorithm:

[0089]

[0090] 6) Unify the coordinate systems of the two systems based on the principles of inertial navigation and magnetic field positioning:

[0091] After completing the solution of the two systems, the coordinate systems of the two systems are unified using the initial attitude angle calculated by the inertial data and the initial direction of the magnet calculated by the magnetic induction intensity data:

[0092]

[0093]

[0094] Where ψ is the yaw angle of the inertial navigation system, θ is the pitch angle of the inertial navigation system, is the angle between the magnet and the x-axis, is the angle between the magnet and the z-axis.

[0095] 7) Unify the timeline of the two systems:

[0096] Extract the start and end times of the two system movements respectively; match the movement times of the two systems and calculate the overall movement time to form a new timeline.

[0097] t s =t is -t ie (6)

[0098] Where, t is is the time when the movement starts, t ie is the time when the movement ends, t s The overall exercise time.

[0099] 8) Data correspondence based on magnetic positioning data:

[0100] First, determine the multiple relationship between the sampling rates of the magnetic positioning system and the inertial navigation system, and use it as the data interval distance, that is, each data interval of the inertial navigation system corresponds to a data point in the magnetic positioning system:

[0101]

[0102] In the formula, K i is the sampling rate of the inertial navigation system, K m is the sampling rate of the magnetic positioning system, round means rounding to the nearest integer, and k means the data interval. Every k points of the inertial navigation system correspond to one point of the magnetic positioning system.

[0103] 9) Replace the orientation of the magnet with the attitude angle of the inertial navigation system:

[0104] m=sinψ*cosθ (8)

[0105] n=sinψ*sinθ (9)

[0106] p=cosψ (10)

[0107] In the formula, m, n, and p are the direction parameters of the magnet, where m represents the direction between the magnet and the x-axis, n represents the direction between the magnet and the y-axis, and p represents the direction between the magnet and the z-axis. By replacing the magnet direction with the attitude angle and substituting formulas (8) to (10) into formula (3) for position calculation, the positioning accuracy can be further improved.

[0108] 10) Use inertial navigation data as the core to perform data matching and calculate speed:

[0109] Since the sampling rate of the inertial navigation system is higher than that of the magnetic field positioning, after the data correspondence, some of the information corresponding to the inertial navigation data will be blank, that is, the corresponding displacement is missing; for the missing displacement data between adjacent displacements, the adjacent displacements are averaged, the acceleration between adjacent displacements is integrated twice, and the two displacements are averaged to determine the missing displacement data X between adjacent displacements:

[0110]

[0111] V a =∫∫adt (12)

[0112] V m =dx / dt (13)

[0113] In the formula, is the displacement mean of two adjacent points, X l is the displacement after twice integrating the acceleration between adjacent displacements, V a is the velocity after the acceleration is integrated, V m is the velocity after the displacement is differentiated, a is the acceleration, t is the time, dx is the displacement difference between adjacent moments, and dt is the time difference between adjacent moments.

[0114] 11) Combine the speeds of the two systems:

[0115] Since there are certain errors in the speeds obtained by both methods, the speeds are fused through Kalman filtering to obtain a more accurate speed. a and V t Take the mean as the observed value Then calculate V a and V t The covariance of is calculated, and the trusted speed is selected as the state variable by comparing the minimum value of the covariance.

[0116] Y (k|k) =Y (k|k-1) +K g(k)*[Z k -H*Y (k|k-1) ] (14)

[0117] Y (k|k) is the optimal estimate of the system state variables at time k, Z k is the observed value of the system at time k, K g(k) The Kalman gain at time k can be calculated from the covariance, H T The transposed matrix of the system observation matrix is ​​the identity matrix, Y (k|k-1) The state variable at time k predicted at time k-1.

[0118] 12) Thus, the two systems are integrated through formulas (4) to (14), that is, the corresponding motion information such as acceleration, angular velocity, displacement and speed can be obtained at every moment.

[0119] In summary, this technical solution combines inertial navigation technology with magnetic field positioning technology. By fusing inertial measurement data with magnetic field positioning data, the rotation, speed and displacement of spherical particles in motion can be accurately calculated. The inaccuracy caused by cumulative errors in inertial navigation when solving displacements is effectively solved, and the number of sensors used in the magnetic positioning system is reduced under the same positioning accuracy.

Claims

1. A method for detecting the motion state of spherical particles, applied to a device for detecting the motion state of spherical particles, characterized in that: The spherical particle motion state detection device comprises a particle container (2) containing a plurality of spherical filling particles (3), a magnetic positioning device is arranged on the wall surface of the particle container (2), a movable spherical tracer particle detection device (4) is also placed in the particle container (2), the magnetic positioning device and the spherical tracer particle detection device (4) are respectively connected to a computer, the spherical tracer particle detection device (4) is used to collect its own inertial motion data, the magnetic positioning device is used to collect the internal magnetic induction intensity data of the spherical tracer particle detection device (4), the computer performs posture calculation according to the inertial motion data, and performs position calculation according to the magnetic induction intensity, and obtains the motion state of the spherical tracer particle detection device (4) through data fusion; The spherical particle motion state detection method comprises the following steps: S1, the spherical tracer particle detection device moves in the particle container and collects corresponding inertial motion data; At the same time, the magnetic positioning device collects magnetic induction intensity data inside the spherical tracer particle detection device; S2, the spherical missing particle detection device transmits the collected inertial motion data to the computer; at the same time, the magnetic positioning device transmits the collected magnetic induction intensity data to the computer; S3, the computer performs attitude calculation on the inertial motion data to obtain the initial attitude angle; and performs position calculation on the magnetic induction intensity data to obtain the initial magnet direction; S4. According to the initial attitude angle and the initial magnet direction, the computer performs unified processing of the coordinate system and the time axis for the inertial navigation and magnetic field positioning; Afterwards, data correspondence is performed with magnetic positioning as the core; Then the attitude angle is used instead of the magnet direction to solve the position; Finally, the data is matched with the inertial navigation as the core to obtain the displacement, acceleration and attitude angle information, and the differential speed and integral speed are calculated. The differential speed and integral speed are fused to obtain the speed information. Wherein, step S4 specifically includes the following steps: S41. Unify the coordinate systems of the two systems based on the principles of inertial navigation and magnetic field positioning: After completing the solution of the two systems, the coordinate systems of the two systems are unified using the initial attitude angle calculated by the inertial motion data and the initial magnet direction calculated by the magnetic induction intensity data: Where ψ is the yaw angle of inertial navigation, θ is the pitch angle of inertial navigation, is the angle between the magnet and the x-axis, is the angle between the magnet and the z-axis; S42. Unify the timeline of the two systems: Extract the start and end times of the two system movements respectively; match the movement times of the two systems, and calculate the overall movement time to form a new timeline: t s =t is -t ie Where, t is is the time when the movement starts, t ie is the time when the movement ends, t s is the overall movement time; S43, data correspondence based on magnetic positioning data: First, determine the multiple relationship between the sampling rates of the magnetic positioning system and the inertial navigation system, and use it as the data interval distance, that is, each data interval distance of the inertial navigation system corresponds to a data point in the magnetic positioning system: In the formula, K i is the sampling rate of the inertial navigation system, K m is the sampling rate of the magnetic positioning system, round means rounding to the nearest integer, k means the data interval distance, and every k points of the inertial navigation system correspond to one point of the magnetic positioning system; S44. Replace the orientation of the magnet with the attitude angle of the inertial navigation system: m=sinψ*cosθ n=sinψ*sinθ p=cosψ In the formula, m, n, and p are the direction parameters of the magnet, among which m represents the direction between the magnet and the x-axis, n represents the direction between the magnet and the y-axis, and p represents the direction between the magnet and the z-axis. By replacing the magnet direction with the attitude angle and performing position calculation again, the positioning accuracy can be improved. S45, using inertial navigation data as the core to perform data correspondence and calculate speed: Since the sampling rate of the inertial navigation system is higher than that of the magnetic field positioning, after the data correspondence, some of the information corresponding to the inertial navigation data will be blank, that is, the corresponding displacement is missing; for the missing displacement data between adjacent displacements, the adjacent displacements are averaged, the acceleration between adjacent displacements is integrated twice, and the two displacements are averaged to determine the missing displacement data X between adjacent displacements: In a =∫adt V m =dx / dt In the formula, is the displacement mean of two adjacent points, X l is the displacement after twice integrating the acceleration between adjacent displacements, V a is the velocity after the acceleration is integrated, V m is the velocity after displacement differentiation, a is acceleration, t is time, dx is the displacement difference at adjacent moments, and dt is the time difference at adjacent moments; S46. Combine the speeds of the two systems: Since there are certain errors in the speeds obtained by the two methods, the speeds are fused through Kalman filtering to obtain accurate speeds. Specifically, V a and V m Take the mean as the observed value: Then calculate V a and V m The covariance of is , and by comparing the minimum value of the covariance, the trusted speed is selected as the state variable: Y (k|k) =Y (k|k-1) +K g(k) *[Z k -H*Y (k|k-1) ] Where Y (k|k) is the optimal estimate of the system state variables at time k, Z k is the observed value of the system at time k, K g(k) The Kalman gain at time k can be calculated from the covariance, H is the system observation matrix, here is the unit matrix, Y (k|k-1) The state variable at time k predicted at time k-1; This achieves the fusion of the two systems, that is, the corresponding acceleration, angular velocity, displacement and speed information can be obtained at every moment.

2. A method for detecting the motion state of spherical particles according to claim 1, characterized in that: The spherical tracer particle detection device (4) comprises a spherical shell, in which a first microcontroller (42) and a magnet (46) are installed, the first microcontroller (42) is respectively connected to a first memory (45), an accelerometer (43) and an angular velocity meter (44), the first microcontroller (42) is connected to a power source (41) via a control switch (47) in an on-off manner, after the control switch (47) is closed, the first microcontroller (42) is connected to the power source (41), the spherical tracer particle detection device (4) starts to move in the particle container (2), the accelerometer (43) and the angular velocity meter (44) are used to collect corresponding inertial motion data, and store the data via the first memory (45), the first microcontroller (42) is connected to a computer in communication to transmit the inertial motion data to the computer.

3. A method for detecting the motion state of spherical particles according to claim 2, characterized in that: The first microcontroller (42), the magnet (46), the first memory (45), the accelerometer (43), the angular velocity meter (44) and the control switch (47) are all integrated and mounted on the same PCB board.

4. A method for detecting the motion state of spherical particles according to any one of claims 2 to 3, characterized in that: The magnet (46) is specifically a cylindrical magnet.

5. A method for detecting the motion state of spherical particles according to claim 2, characterized in that: The power source (41) is specifically a button battery.

6. A method for detecting the motion state of spherical particles according to claim 2, characterized in that: The magnetic positioning device comprises a magnetometer group (1), wherein the magnetometer group (1) is connected to a second microcontroller (52), the second microcontroller (52) is connected to a second memory (53), the second microcontroller (52) is communicatively connected to a computer, and the magnetometer group (1) is used to collect magnetic induction intensity data of a magnet (46) in a spherical tracer particle detection device (4), store the data through the second memory (53), and transmit the magnetic induction intensity data to the computer through the second microcontroller (52).

7. A method for detecting the motion state of spherical particles according to claim 6, characterized in that: The magnetometer group (1) comprises a plurality of magnetometer sensors arranged at equal intervals on the wall surface of the particle container (2).

8. A method for detecting the motion state of spherical particles according to claim 1, characterized in that: The step S3 specifically performs attitude calculation on the inertial motion data by using the quaternion method, and performs position calculation on the magnetic induction intensity data by using the magnetic dipole model.

Images