Precise seeding machine speed measurement method and system based on radar Beidou inertial navigation multi-source data fusion

Abstract

The invention belongs to the technical field of accurate measurement seeding, and discloses a seeding machine accurate speed measurement method and system based on radar Beidou inertial navigation multi-source data fusion. The speed measurement system comprises a seed sowing device driven by a motor to operate, a speed monitoring module, a display screen and a control unit for implementing the accurate speed measurement method. The speed monitoring module comprises a ground speed radar, an inertial measurement element and a Beidou receiver; the display screen is used for selecting a speed measurement mode and displaying a seeding state; and the control unit calculates the advancing speed of the precise seeder based on the speed related data obtained by the speed monitoring module and the selected speed measurement mode, and then adjusts the rotating speed of the motor according to the advancing speed of the precise seeder. According to the invention, multiple speed measurement modes, namely Beidou speed measurement, radar speed measurement and inertial navigation speed measurement, are integrated and combined pairwise, fusion speed measurement of different modes can be selected and switched according to environmental conditions, the limitation of a single mode can be effectively overcome, and a complex environment can be effectively dealt with.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent agricultural machinery equipment, and relates to a precise speed measurement method and system for a seeder that fuses multi-source data of radar, Beidou, and inertial navigation. Background Technique

[0002] With the development of seeding machines towards intelligence, the electric drive seeding technology is the core technology of intelligent seeding machines. Its working principle mainly includes three major links: speed measurement, electric drive, and speed following control. The speed measurement module monitors the traveling speed of the seeding machine in real time. The control system calculates the target seeding speed and the corresponding motor speed according to the real-time speed and agronomic requirements, and controls the seeding speed by adjusting the motor speed, ultimately achieving the speed following control of the seeding speed. The electric drive seeding technology can effectively improve the seeding uniformity and increase the crop yield.

[0003] Precisely obtaining the real-time speed of the seeding machine is the key to realizing electric drive intelligent seeding. Currently, the methods for measuring the speed of seeding machines are mainly divided into two categories. One is to measure the rotational speed of the wheels, such as Hall sensors, encoders, etc.; the other is to measure the vehicle body speed, such as speed measurement by a Beidou satellite positioning module, radar speed measurement, etc. Among them, the speed measurement by Hall sensors and encoders has high real-time performance but is prone to slipping during operation and affects the measurement accuracy; satellite positioning obtains the position and speed of the tractor by parsing message information, but is susceptible to satellite signals and has a certain lag; radar speed measurement has relatively high real-time performance, but its measurement accuracy is very sensitive to the installation angle of the radar.

[0004] A single speed measurement method no longer meets the requirements for high-precision speed measurement of seeding machines. Multiple speed measurement sensors are gradually used to measure the traveling speed of seeding machines. Different speed measurement sensors can complement each other's advantages and disadvantages, improving the speed measurement accuracy and anti-interference ability. However, the current development is not sufficient. Mostly, different speed measurement sensors are set to be switched and used for speed measurement under different speed conditions, or the speed measurement results of different speed measurement sensors are simply weighted and averaged.

[0005] For example, in the Chinese patent "An Electric Drive Seeding Control System for Precision Seeding" with the publication number CN111813032A, a ground wheel speed measurement sensor is used as the first test speed and a positioning and transmission module is used as the second test speed. According to different working states and the different magnitude relationships of the two speed measurement results, one of the test speeds is taken as the speed measurement result. This invention distinguishes multiple situations and uses different speed measurement sensors, making full use of the advantages of different speed measurement sensors at different speed stages. However, essentially, it is still a single speed measurement sensor for speed measurement, and the accuracy of the speed data measured by the speed measurement sensor is relatively low.

[0006] For example, in the Chinese patent "High-speed seeding machine control system, method, electronic device and storage medium" with the publication number CN117369350A, the invention comprehensively analyzes and judges the first vehicle speed measured by the ground wheel speed sensor, the second vehicle speed measured by the GNSS speed measurement module, and the third vehicle speed measured by the radar speed sensor. Specifically: when the first vehicle speed is not greater than the first speed threshold, the target vehicle speed is determined to be zero; when the first vehicle speed is greater than the first speed threshold but not greater than the second speed threshold, and the third vehicle speed is greater than zero, the first vehicle speed is determined to be the target vehicle speed of the seeding machine; when it is determined that the first vehicle speed is greater than the second speed threshold, according to the signal strength information of the GNSS speed measurement module, the weighted average of the first vehicle speed, the second vehicle speed and the third vehicle speed is used as the target vehicle speed. The judgment logic of this invention is still relatively simple and direct, and the selection of fixed thresholds and weighting coefficients cannot well adapt to the complex agricultural operation environment.

[0007] For example, in the Chinese patent "An agricultural implement speed measurement method and speed measurement device based on ground wheel and satellite positioning" with the publication number CN118549672A, the invention uses an encoder and a satellite positioning module to collect speed data respectively, and the speed measurement result is the weighted average of the two. The weight is calculated according to the relationship between the speed measurement result of the speed sensor and the predetermined speed threshold, realizing the dynamic adjustment of the weight coefficient according to the speed measurement situation. However, there is also the problem that the selection of fixed thresholds cannot well adapt to the complex agricultural operation environment, and at the same time, the correlation between the current speed and the speed at the previous moment is not utilized, and there is no filtering process.

[0008] For example, in the Chinese patent "A vehicle speed prediction method based on adaptive Kalman filter" with the publication number CN107909815A, the invention uses a Kalman filter to achieve accurate prediction of the vehicle speed based on the speed at the previous moment and the speed measurement result of the speed sensor. However, only one speed sensor is used, and the prediction accuracy can be further improved. Summary of the Invention

[0009] Aiming at the problems existing in the prior art, the present invention provides an accurate speed measurement method and system for a seeding machine with radar, Beidou and inertial navigation multi-source data fusion.

[0010] An accurate speed measurement method and system for a seeding machine with radar, Beidou and inertial navigation multi-source data fusion. The speed measurement method is to simultaneously use radar, Beidou and inertial navigation to measure the speed of the seeding machine, obtain multi-source speed measurement results, and then use a Kalman filter for data fusion by pairwise combination to obtain the high-precision traveling speed of the seeding machine.

[0011] A method and system for accurately measuring the speed of a seed drill by fusion of radar, Beidou and inertial navigation multi-source data. The speed measuring system comprises a seed meter driven by a motor, a speed monitoring module, a display screen, and a control unit for implementing the accurate speed measuring method; the speed monitoring module comprises a ground speed radar, an inertial measurement element and a Beidou receiver; the display screen selects a speed measuring mode and displays a seeding status; the control unit calculates the traveling speed of the accurate seed drill based on the speed-related data obtained by the speed monitoring module and the selected speed measuring mode, and then adjusts the rotation speed of the motor according to the traveling speed of the accurate seed drill.

[0012] Furthermore, the speed measurement modes include: a fusion speed measurement mode of radar speed measurement and Beidou speed measurement, a fusion speed measurement mode of radar speed measurement and inertial navigation speed measurement, and a fusion speed measurement mode of Beidou speed measurement and inertial navigation speed measurement;

[0013] In the fusion mode of radar speed measurement and Beidou speed measurement: radar speed measurement can quickly and accurately capture speed changes. In an environment with weak satellite signals and low speed, radar speed measurement can fill the gap of Beidou speed measurement. At the same time, Beidou speed measurement provides a stable and accurate reference for the speed measurement results.

[0014] In the fusion mode of radar speed measurement and inertial navigation speed measurement: inertial navigation speed measurement can provide accurate speed estimation in real time, while radar speed measurement can correct the drift error of inertial navigation speed measurement during long-term operation. Radar speed measurement can provide stable and accurate speed data. The fusion of the two further improves the reliability of speed measurement;

[0015] In the fusion mode of Beidou speed measurement and inertial navigation speed measurement: when the satellite signal is interfered or weak, inertial navigation speed measurement can fill the gap of Beidou speed measurement and maintain the continuity of speed measurement. When speed measurement is required for a long time, Beidou speed measurement can eliminate the accumulated error of inertial navigation speed measurement and ensure the accuracy of the final speed calculation.

[0016] Further, the speed monitoring module includes a ground speed radar, an inertial measurement unit and a Beidou receiver;

[0017] The ground speed radar is installed at the tail of the seeder to ensure that the signal path of the ground speed radar is not blocked by the structure or equipment of the seeder to prevent the ground speed radar from failing to detect the working area. The antenna of the Beidou receiver should be installed on the top or a higher position of the seeder. The installation position should ensure that the antenna is not blocked and can receive sufficient satellite signals to ensure positioning accuracy. The inertial measurement element should be installed at the center of the seeder. Its installation position should avoid excessive mechanical interference or severe vibration as much as possible. It needs to be placed horizontally and aligned with the reference coordinate system of the seeder to ensure that the angle and acceleration measurement results of the inertial measurement element are the pitch angle and acceleration of the seeder in the front and rear directions to ensure the accuracy of the attitude data.

[0018] Further, the radar speed measurement principle is based on the Doppler effect. According to the Doppler effect, when electromagnetic waves encounter a moving target, the frequency of the reflected wave will change. This frequency change is called the Doppler frequency shift, which is proportional to the speed of the target object. The Doppler frequency shift formula is as follows:

[0019]

[0020] where: f d is the Doppler frequency shift; v is the speed of the moving object; f 0 is the frequency of the transmitted signal; c is the speed of light. Thus, by installing a radar on the seeder, the radar continuously emits electromagnetic waves of a certain frequency to the ground and receives its reflected signal, measures the frequency change, and can calculate its traveling speed according to the formula. Radar speed measurement does not rely on wheel rotation and satellite signals and can perform speed measurement independently, which makes it superior to encoders and satellite speed measurement and is not affected by wheel slip and satellite signal problems. However, the speed measurement result is affected by the change of the radar's angle to the ground, and there are significant differences in the effects of different radars;

[0021] The core components of the inertial measurement element are the gyroscope and the accelerometer. The gyroscope can measure the angular velocity magnitudes in three axes, and the accelerometer can measure the acceleration magnitudes in three axes. In addition to providing the 16-bit ADC signal acquisition function of the three-axis gyroscope and three-axis accelerometer sensors, the three-axis accelerometer can output the components of the current vehicle acceleration in the three axes in real time at a high frequency. After subtracting the influence of the gravity acceleration component, the speed and attitude of the vehicle at this time can be obtained through integral operation based on the measurement interval time and the speed at the previous moment. The inertial measurement element has high real-time performance and high instantaneous measurement accuracy and can perform speed measurement independently. However, the cumulative error of the integral operation is large;

[0022] The Beidou speed measurement is realized through a Beidou receiver. The Beidou receiver automatically searches for satellites and can provide accurate speed measurement in various environments. The Beidou receiver outputs message information, including speed information, and the control unit can obtain the speed information by parsing the message information. The advantage of Beidou speed measurement is its high precision and all-weather operation, and it is not affected by wheel slip and vehicle body vibration. However, the disadvantage is that it depends on satellite signals. When affected by obstacles such as buildings and trees, signal loss or error increase may occur, and there is a certain lag, and the speed measurement error is large at low speeds.

[0023] Further, the Kalman filter is mainly divided into two parts: prediction and correction. In the prediction stage, the filter uses the estimate of the previous state to make an estimate of the current state. In the correction stage, the filter uses the measured value of the current state to correct the predicted value obtained in the prediction stage to obtain a more accurate new estimate value.

[0024] Further, in the prediction stage, the Kalman filter predicts the predicted value at the current moment based on the state at the previous moment and the state transition matrix A, i.e., the state equation. This predicted value is an estimated value and does not consider the measurement value at the current moment; Calculate the predicted value:

[0025]

[0026]

[0027] At the same time, calculate the predicted error covariance matrix which is calculated from the error covariance matrix at the previous moment and the process noise covariance matrix Q; the covariance matrix Q of the process noise represents the dynamic noise of the system itself;

[0028] Calculate the predicted error covariance matrix:

[0029]

[0030] Further, in the correction stage, the Kalman filter corrects the state estimate at the current moment according to the difference between the predicted state and the measurement value. This estimated value is a more accurate estimated value and takes into account the measurement value at the current moment; the core of the correction step is to calculate the Kalman gain K k , weight the predicted value and the measurement value through the Kalman gain, and update the error covariance matrix for use at the next moment;

[0031] Calculate the Kalman gain:

[0032]

[0033] H is the measurement matrix, and R is the covariance matrix of the measurement noise, representing the noise of the measurement data;

[0034] Correct the state estimate:

[0035]

[0036] where is the corrected state estimate, Z k is the measurement value, is the residual between the measurement value and the predicted value, representing the difference between the measurement data and the predicted data;

[0037] Update the error covariance:

[0038]

[0039] where P k is the updated error covariance, and I is the identity matrix;

[0040] The parameter settings of the Kalman filter include the following core elements: state transition matrix A, measurement matrix H, process noise covariance Q, measurement noise covariance R, and Kalman gain K k , state estimation error covariance matrix P k ; among them, the state transition matrix A and the measurement matrix H are determined by the system model; among them, the Kalman gain K k , state estimation error covariance matrix P k is the calculation result of the Kalman filter; adjusting the filtering effect of the Kalman filter is achieved by adjusting the process noise covariance Q and the measurement noise covariance R;

[0041] The process noise covariance Q represents the uncertainty or noise in the system state transition process; increasing Q indicates a higher uncertainty in the system state change, that is, assuming the system is easily affected by noise interference, and decreasing Q assumes that the system output change is relatively stable; if Q is too small, the filter may overly rely on previous predictions, resulting in insensitivity to measurement data; if Q is too large, the filter may be overly sensitive to noise, resulting in unstable estimation;

[0042] The measurement noise covariance R describes the noise characteristics in the measurement data, that is, the error generated by the measuring instrument or external interference, and R is used to represent the magnitude of the measurement noise of each sensor; if the measurement noise is small, that is, the sensor is very accurate, then R should be reduced, so that the Kalman filter will rely more on the measured values; if the measurement noise is large, then increase R, so that the filter will rely more on the model prediction.

[0043] Another object of the present invention is to provide an information data processing terminal, and the information data processing terminal is used to implement the precise speed measurement method and system of the seeder for radar Beidou inertial multi-source data fusion.

[0044] Combined with the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by the present invention are:

[0045] Existing field operation speed measurement systems usually rely on a single speed measurement technology, or simply set to switch to different speed measurement sensors for speed measurement under different speed conditions, such as Beidou, encoder, radar, etc. Different speed measurement sensors have different advantages and disadvantages and cannot fully cope with various challenges in complex environments. The present invention integrates multiple speed measurement methods: Beidou speed measurement, radar speed measurement, inertial navigation speed measurement, and two-by-two combinations. It can choose to switch different modes of fusion speed measurement according to environmental conditions, which can effectively overcome the limitations of a single mode and effectively cope with complex environments. For example, when the satellite signal is weak, the fusion speed measurement data of radar speed measurement and inertial navigation speed measurement can be used first; in severe shaking, the fusion speed measurement data of Beidou speed measurement and inertial navigation speed measurement can be used first; in flat terrain, the fusion speed measurement data of radar speed measurement and Beidou speed measurement can be used first. The speed measurement system can switch to other speed measurement modes when a certain sensor fails or the signal is unstable, and continuously provide accurate speed data.

[0046] A single sensor has inherent defects in a specific environment, such as low accuracy and weak anti-interference ability. Due to changes in the environment and operating conditions, it is difficult to maintain stable and accurate speed measurement during seeding operations. The present invention uses the Kalman filter data fusion algorithm to fuse various sensor data in real time and utilize the speed information at the previous moment, thereby improving the accuracy, stability and anti-interference ability of the speed measurement system. Multi-sensor fusion can effectively reduce the errors of each sensor, complement each other's advantages, and make the final speed measurement results more accurate. Kalman filtering can dynamically adjust the data weights of the measured values ​​and predicted values, smooth out the noise and errors, effectively reduce the fluctuations caused by the noise or errors of a single sensor, and provide stable speed measurement results.

[0047] The present invention realizes accurate, stable and anti-interference real-time speed monitoring by constructing a field operation speed measurement system based on a multi-sensor fusion algorithm. After the technology is transformed, it is applied to large-scale sowing agricultural machinery to provide accurate speed information support for sowing operations, which is conducive to precision sowing operations, avoiding over-sowing or under-sowing, improving crop growth conditions and increasing yields. At the same time, the invention can be extended to variable fertilization, spraying and other operations of ground machinery to reduce the amount of fertilizers and pesticides used and the impact on the environment. The invention enhances the intelligence level of agricultural machinery, improves the competitiveness of the agricultural machinery market, provides innovative products for enterprises, and opens up the market for intelligent agricultural equipment.

[0048] At present, the speed measurement technology for field operations mainly relies on a single sensor, which has problems such as insufficient measurement accuracy, poor environmental adaptability, and easy signal interruption. Through the innovative fusion of data from radar, Beidou positioning, and inertial navigation sensors and the introduction of the Kalman filtering algorithm, the present invention realizes precise speed measurement in complex field environments and establishes a high-precision, low-fluctuation, and highly robust speed measurement system for dealing with complex field environments. By switching the speed measurement mode and data fusion, the limitations of a single sensor are effectively solved, and a multi-mode high-precision speed measurement scheme for complex field operation environments is constructed. It can achieve single-sensor speed measurement in two forms of radar and Beidou, as well as dual-sensor combined speed measurement in three forms of radar-Beidou, radar-inertial navigation, and Beidou-inertial navigation, which can effectively improve the reliability of the speed measurement hardware unit and the robustness of the data results, and will fill the technical gap of high-precision multi-mode speed measurement in the field of intelligent field operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 FIG. is a structural diagram of an accurate speed measurement method and system for a seeder with multi-source data fusion of radar, Beidou, and inertial navigation provided by an embodiment of the present invention;

[0050] Figure 2 FIG. is an effect diagram of data fusion of radar speed measurement and Beidou speed measurement provided by an embodiment of the present invention;

[0051] Figure 3 FIG. is an effect diagram of data fusion of radar speed measurement and inertial navigation speed measurement provided by an embodiment of the present invention;

[0052] Figure 4 FIG. is an effect diagram of data fusion of Beidou speed measurement and inertial navigation speed measurement provided by an embodiment of the present invention;

[0053] In the figure: 1, inertial measurement element; 2, Beidou receiver; 3, ground speed radar; 4, display screen; 5, control unit; 6, metering motor. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0055] An embodiment of the present invention provides an accurate speed measurement method and system for a seeder with multi-source data fusion of radar, Beidou, and inertial navigation. The speed measurement system includes a metering device driven by a metering motor 6, a speed monitoring module, and a control unit 5 and a display screen 4 for implementing the accurate speed measurement method;

[0056] The control unit 5 calculates the accurate traveling speed of the seeder based on the speed-related data obtained by the speed monitoring module and the selected speed measurement mode, and then adjusts the rotation speed of the motor according to the traveling speed of the seeder.

[0057] The speed monitoring module includes a ground speed radar 3, an inertial measurement element 1, and a Beidou receiver 2;

[0058] The ground speed radar 3 is installed at the tail of the seeder to ensure that the signal path of the ground speed radar 3 is not blocked by the structure or equipment of the seeder, so as to avoid the ground speed radar 3 being unable to detect the operation area; the antenna of the Beidou receiver 2 should be installed on the top or a higher position of the seeder, and the installation position should ensure that the antenna is unobstructed and can receive sufficient satellite signals to ensure the positioning accuracy; the inertial measurement element 1 should be installed at the center position of the seeder, and its installation position should avoid excessive mechanical interference or severe vibration as much as possible. At the same time, it needs to be placed horizontally and aligned with the reference coordinate system of the seeder to ensure that the angle and acceleration measurement results of the inertial measurement element 1 are the pitch angle and the acceleration in the front and rear directions of the seeder, so as to ensure the accuracy of the attitude data.

[0059] Furthermore, the speed measurement modes include: the integrated speed measurement mode of radar speed measurement and Beidou speed measurement, the integrated speed measurement mode of radar speed measurement and inertial navigation speed measurement, and the integrated speed measurement mode of Beidou speed measurement and inertial navigation speed measurement, which are switched through the display screen;

[0060] In the integrated mode of radar speed measurement and Beidou speed measurement: Radar speed measurement can quickly and accurately capture speed changes. In an environment with weak satellite signals and low speeds, radar speed measurement can fill the gap of Beidou speed measurement. At the same time, Beidou speed measurement provides a stable and accurate reference for the speed measurement result;

[0061] In the integrated mode of radar speed measurement and inertial navigation speed measurement: Inertial navigation speed measurement can provide accurate speed estimation in real time, while radar speed measurement can correct the drift error of inertial navigation speed measurement during long-term operation. The radar can provide stable and accurate speed data, and the integration of the two further improves the reliability of speed measurement;

[0062] In the integrated mode of Beidou speed measurement and inertial navigation speed measurement: In an environment where satellite signals are interfered or weak, inertial navigation speed measurement can fill the gap of Beidou speed measurement and maintain the continuity of speed measurement. When speed measurement needs to be carried out for a long time, Beidou speed measurement can eliminate the cumulative error of inertial navigation speed measurement and ensure the accuracy of the final speed calculation.

[0063] Furthermore, the principle of the radar speed measurement is based on the Doppler effect. According to the Doppler effect, when electromagnetic waves encounter a moving target, the frequency of the reflected wave will change. This frequency change is called the Doppler frequency shift, which is proportional to the speed of the target object. The Doppler frequency shift formula is as follows:

[0064]

[0065] Where: f d Doppler frequency shift; v is the speed of the moving object; f 0$f$ is the frequency of the transmitted signal; $c$ is the speed of light; thus, by installing a radar on the seeder, the radar continuously emits electromagnetic waves of a certain frequency towards the ground and receives the reflected signals, measures the frequency change, and can calculate its traveling speed according to the formula; radar speed measurement does not rely on wheel rotation and satellite signals and can perform speed measurement independently, which makes it superior to encoder and satellite speed measurement and is not affected by wheel slip and satellite signal problems. However, the speed measurement result is affected by the change of the radar's angle to the ground, and there are significant differences in the effects of different radars;

[0066] The core components of the inertial measurement element are gyroscopes and accelerometers. The gyroscope can measure the angular velocity magnitudes in three axes, and the accelerometer can measure the acceleration magnitudes in three axes; in addition to providing the 16-bit ADC signal acquisition function of the three-axis gyroscope and three-axis accelerometer sensors, the three-axis accelerometer can output the components of the current vehicle acceleration in the three axes in real time at high frequency. After subtracting the influence of the gravity acceleration component, the vehicle's speed and attitude at this time can be obtained through integral operation based on the measurement interval time and the speed at the previous moment; the inertial measurement element has high real-time performance and high instantaneous measurement accuracy and can perform speed measurement independently. However, the cumulative error of the integral operation is large;

[0067] The Beidou speed measurement is realized through a Beidou receiver. The Beidou receiver automatically searches for satellites and can provide accurate speed measurement in various environments; the Beidou receiver outputs message information, including speed information, and the control unit can obtain the speed information by parsing the message information; the advantages of Beidou speed measurement are its high precision and all-weather, and it is not affected by wheel slip and vehicle body vibration. However, the disadvantage is that it depends on satellite signals. When affected by obstacles such as buildings and trees, signal loss or error increase may occur, and there is a certain lag, and the speed measurement error is large at low speeds.

[0068] Furthermore, the Kalman filter is mainly divided into two parts: prediction and correction. In the prediction stage, the filter uses the estimate of the previous state to make an estimate of the current state; in the correction stage, the filter uses the measured value of the current state to correct the predicted value obtained in the prediction stage to obtain a more accurate new estimated value.

[0069] Furthermore, in the prediction stage, the Kalman filter predicts the predicted value of the current moment according to the state at the previous moment and the state transition matrix $A$ that is, the state equation. This predicted value is an estimated value and does not consider the measured value at the current moment;

[0070] Calculate the predicted value:

[0071]

[0072] At the same time, calculate the predicted error covariance matrix Calculated from the error covariance matrix at the previous moment and the process noise covariance matrix Q; the covariance matrix Q of the process noise represents the dynamic noise of the system itself;

[0073] Calculate the predicted error covariance matrix:

[0074]

[0075] Furthermore, in the correction stage, the Kalman filter corrects the state estimate at the current moment according to the difference between the predicted state and the measured value, and this estimated value is a more accurate estimated value that takes into account the measured value at the current moment; the core of the correction step is to calculate the Kalman gain K k , weight the predicted value and the measured value through the Kalman gain, and update the error covariance matrix P k Prepare for use at the next moment;

[0076] Calculate the Kalman gain:

[0077]

[0078] H is the measurement matrix, and R is the covariance matrix of the measurement noise, representing the noise of the measurement data;

[0079] Correct the state estimate:

[0080]

[0081] Among them, is the corrected state estimate, Z k is the measured value, is the residual between the measured value and the predicted value, representing the difference between the measurement data and the predicted data;

[0082] Update the error covariance:

[0083]

[0084] Among them, P k is the updated error covariance, and I is the identity matrix;

[0085] The parameter settings of the Kalman filter include the following core elements: state transition matrix A, measurement matrix H, process noise covariance Q, measurement noise covariance R, Kalman gain K k , state estimate error covariance matrix P k ; among them, the state transition matrix A and the measurement matrix H are determined by the system model; among them, the Kalman gain K k , state estimate error covariance matrix Pk is the calculation result of the Kalman filter; the filtering effect of the Kalman filter is adjusted by adjusting the process noise covariance Q and the measurement noise covariance R;

[0086] The process noise covariance Q represents the uncertainty or noise in the system state transition process; increasing Q indicates a higher uncertainty in the system state change, that is, it is assumed that the system is more vulnerable to noise interference, and decreasing Q assumes that the system output change is relatively stable; if Q is too small, the filter may overly rely on previous predictions, resulting in insensitivity to measurement data; if Q is too large, the filter may be overly sensitive to noise, resulting in unstable estimation;

[0087] The measurement noise covariance R describes the noise characteristics in the measurement data, that is, the error generated by the measuring instrument or external interference. R is used to represent the magnitude of the measurement noise of each sensor; if the measurement noise is small, that is, the sensor is very precise, then R should be reduced, so that the Kalman filter will rely more on the measured values; if the measurement noise is large, then increase R, so that the filter will rely more on the model prediction.

[0088] The Kalman filter is a powerful data fusion tool, a recursive algorithm based on a linear system, widely used to estimate the state of a dynamic system and obtain an optimal estimate from noisy measurements. It improves the estimation accuracy of the system state by weighted combining the predicted value and the measured value. The core idea of the Kalman filter is to minimize the mean square error of the estimation error and can handle the noise in the system and measurements. In practical applications, the Kalman filter can eliminate the influence of noise and provide an optimal estimate in multi-source data fusion.

[0089] Advantages of the Kalman filter:

[0090] Under the assumptions of a linear system and Gaussian noise conditions, the Kalman filter can provide an optimal state estimate, that is, minimize the mean square error. The Kalman filter is recursive and does not require storing a large amount of historical data. It only needs the state at the current moment and the estimate at the previous moment, with low computational requirements and is suitable for real-time processing. The Kalman filter is a low-pass filter that can effectively filter out the noise in the system, provide a smooth estimation result, and can dynamically adjust the weighting coefficients for prediction and measurement according to the actual situation of the system, so as to adapt to different noise levels.

[0091] I. Specific application fields or related products of the present invention

[0092] The embodiment of the present invention provides an information data processing terminal, and the information data processing terminal is used to implement the precise speed measurement method and system of the seeder for radar Beidou inertial navigation multi-source data fusion.

[0093] The present invention is mainly applicable to speed-controlled seeders, which can provide accurate speed data for the seeders in complex working environments, so as to support the seeders to dynamically adjust their working states according to real-time speed information, ensure operation accuracy, and improve the uniformity of sowing, the accuracy of fertilization, and the overall quality of other agricultural operations. In addition, the invention can also be widely applied to other agricultural machinery and equipment, such as transplanters, fertilizer applicators, sprayers, etc., providing efficient solutions for scenarios that require accurate motion speed information.

[0094] Common filtering algorithms usually focus on smoothing sensor data through mathematical methods, such as mean filtering, median filtering, etc., reducing the influence of noise, and improving the smoothness and accuracy of the data. These methods can improve the smoothness of the measured data and slightly improve the accuracy to a certain extent, but they do not fundamentally solve the systematic errors and measurement deviations existing in the sensors themselves. For example, when the zero drift of the sensor occurs and it is in an abnormal working state, the measurement error of the sensor is large and remains for a long time, and a single filtering algorithm cannot effectively improve the measurement accuracy.

[0095] Single sensors have inherent defects in specific environments, with low accuracy and weak anti-interference ability. Due to changes in environmental and operating conditions, it is difficult to maintain stable and accurate speed measurement during sowing operations. The present invention uses the Kalman filter data fusion algorithm to fuse various types of sensor data in real time and utilize the speed information at the previous moment. The Kalman filter can dynamically adjust the data weights of the measured values and predicted values, smooth the noise and errors, effectively reduce the fluctuations caused by the noise or errors of a single sensor, and provide a stable speed measurement result. Multi-sensor fusion can utilize the characteristics of different sensors, enhance the adaptability of the system to complex environments or dynamic changes, effectively reduce the errors of each sensor, complement the advantages, and make the final speed measurement result more accurate. For example, the real-time performance of radar speed measurement is used to make up for the time delay and signal instability of Beidou speed measurement, while Beidou speed measurement is used to make up for the influence of vehicle body vibration and terrain changes on radar speed measurement, and radar speed measurement or Beidou speed measurement is used to solve the drift problem of inertial navigation speed measurement.

[0096] Embodiments of the present invention integrate multiple speed measurement methods, including: Beidou speed measurement, radar speed measurement, and inertial navigation speed measurement, and combine them to perform fusion speed measurement by switching different modes according to environmental conditions. This method overcomes the limitations of a single mode, can effectively cope with complex environments, and improve the accuracy and robustness of speed measurement.

[0097] In the present invention, the Kalman filtering technology is adopted for data fusion. The Kalman filter can provide an optimal state estimate and minimize the mean square error under the assumptions of a linear system and Gaussian noise. As a recursive filter, the Kalman filter does not need to store a large amount of historical data, only relying on the state estimate at the current moment and the prediction result at the previous moment, with relatively low computational requirements and being suitable for real-time processing. It can effectively filter out the noise in the system, provide a smooth estimation result, and can dynamically adjust the weighting coefficients for the predicted value and the measured value according to the actual system situation, thereby reducing the influence of noise.

[0098] In this embodiment, multiple groups of speed measurement experiments were conducted. The relative error of the original radar speed measurement result was 4.03%, and the relative error of the original Beidou speed measurement result was 4.50%. The accuracy was improved in different fusion speed measurement modes, such as Figure 2 、 Figure 3 、 Figure 4 . Due to the cumulative error problem, the inertial navigation cannot measure speed alone, and only its acceleration measurement results are used in the fusion speed measurement. Therefore, the inertial navigation speed measurement result curve is not plotted in the figure.

[0099] In the fusion speed measurement mode of radar speed measurement and Beidou speed measurement, when the specific parameters of the Kalman filter are: Q = 0.00015, R 1 = 0.345, R 2 = 0.582, the average relative error of each group of the obtained fusion speed measurement results is 1.94%. In an actual case with a reference speed of 3.60 km / h, the average relative error of the fusion result is 1.74%, and the maximum relative error is 4.18%, such as Figure 2 .

[0100] In the fusion speed measurement mode of radar speed measurement and inertial navigation speed measurement, when the specific parameters of the Kalman filter are: Q 1 = 0.00038, Q 2 = 0.00010, R 1 = 0.130, R 2 = 0.450, the average relative error of each group of the obtained fusion speed measurement results is 3.60%. In an actual case with a reference speed of 3.60 km / h, the average relative error of the fusion result is 1.78%, and the maximum relative error is 6.34%, such as Figure 3 .

[0101] In the fusion speed measurement mode of Beidou speed measurement and inertial navigation speed measurement, when the specific parameters of the Kalman filter are: Q 1 = 0.00028; Q 2 = 0.00010; R 1 = 0.120; R 2= 0.450, the average relative error of each group of the obtained integrated speed measurement results is 3.25%. In the actual case where the reference speed is 7.20 km / h, the average relative error of the integrated result is 1.90%, and the maximum relative error is 4.71%, as Figure 4 .

[0102] Integration of Radar Speed Measurement and Beidou Speed Measurement in Embodiment 1

[0103] In this embodiment, a ground speed radar and a Beidou receiver are selected as the speed measurement modules to integrate radar speed measurement and Beidou speed measurement; an STM32 single-chip microcomputer is selected as the control unit to complete the control and data processing of the system; a serial port screen is selected as the display screen, which supports touch control to achieve human-computer interaction, curve display of speed changes, selection of speed measurement modes and seeding settings, etc. Install according to the installation method in the implementation manner and complete the corresponding connection with the data cable.

[0104] The measurement objects of radar speed measurement and Beidou speed measurement are the same, both are the actual traveling speed of the seeder. Assume that the measured value of radar speed measurement is v 1 , the measurement noise covariance is R 1 , the measured value of radar speed measurement is v 2 , the observation noise covariance is R 2 . First, according to the observation noise covariance of the two, perform weighted averaging according to the following formula to achieve data fusion, and the fusion result is used as the measurement value Z of the Kalman filter.

[0105]

[0106] System Model Construction:

[0107] Before performing Kalman filtering, it is first necessary to clarify the state and observation models of the system, and determine the state transition matrix A and the measurement matrix H. In the integration of radar speed measurement and Beidou speed measurement, the measurement value of the Kalman filter is speed, which is a one-dimensional Kalman filter.

[0108]

[0109] The speed data measured by radar and Beidou have different precisions and noise characteristics. Different weights can be assigned to different speed measurement sensors by adjusting their respective noise covariance matrices. The specific values are given according to the measurement error description of the sensors or actual experiments. For example, if the measurement accuracy of the radar is higher, then it is smaller, and the measurement result of the radar will have a greater weight. The noise covariance matrix R of the Kalman filter is:

[0110]

[0111] When the seeder is operating normally, it should move forward at a stable speed. Therefore, assume that the speed of the seeder is in uniform linear motion, that is, the speed of the seeder at this moment should be the same as the speed of the seeder at the previous moment. Therefore, the state equation A in it is 1. Since the speed magnitude is directly measured, the measurement matrix H is 1.

[0112] Initialize the Kalman filter. The seeder starts from a standstill, and set the initial speed estimate value Initialize the error covariance matrix

[0113] The working process of the integrated speed measurement of radar speed measurement and Beidou speed measurement:

[0114] The seeder speed measurement system is powered on, the system is initialized and system parameters are set. Select the integrated speed measurement mode of radar speed measurement and Beidou speed measurement, enable the timer interrupt, and set the integrated speed measurement working frequency;

[0115] After power-on, the ground speed radar and inertial measurement components automatically maintain high-frequency data measurement;

[0116] The single-chip microcomputer sends speed measurement commands to the ground speed radar through the serial port in a loop. After the ground speed radar receives the command, the ground speed radar sends a data packet containing speed measurement data, and analyzes the data packet to obtain the radar speed measurement result v 1,k ;

[0117] The single-chip microcomputer receives the data packet from the Beidou receiver, and obtains the Beidou speed measurement result v after parsing 2,k ;

[0118] After the timer interrupt is triggered, the Kalman filter starts filtering:

[0119] Data fusion calculation. Perform data fusion on the speed measurement results of radar speed measurement and Beidou speed measurement;

[0120]

[0121] Prediction step. Predict the speed at this moment according to the speed at the previous moment. In this embodiment, A = 1, H = 1, so the predicted speed value at this moment is equal to the speed estimate value at the previous moment; at the same time, calculate the predicted error covariance matrix

[0122]

[0123] k , correct the state estimate Update the error covariance

[0124]

[0125] The corrected state estimate output by the final Kalman filter is the fusion result of radar speed measurement and Beidou speed measurement, and is an accurate estimate of the traveling speed of the seeder.

[0126] Subsequently, according to the traveling speed of the seeder, the setting of the seeding plant spacing, and the calculated seeding plant spacing, the target rotational speed and motor PWM of the metering unit motor are set;

[0127] Appropriate PWM is output to the metering unit motor to adjust the motor speed, and at the same time, information such as the traveling speed and seeding status of the seeder is written into the display register, and the display performs data visualization display;

[0128] Meanwhile, when the single-chip microcomputer monitors the debugging signal of the seeding state from the display, the speed measurement system is correspondingly adjusted after the interruption is triggered. The user can input the required operation parameters through the display, such as starting the operation, theoretical plant spacing, speed measurement setting, etc., and at the same time observe the seeding working state through the display.

[0129] In this embodiment, according to the selected device and working environment, the parameter Q = 0.00015; R 1 = 0.34500; R 2 = 0.582, multiple groups of speed measurements are carried out, and the average value of the average relative error of each group of speed measurement fusion results is: 1.94%. Next Figure 2 In an actual case, the reference speed is 3.6 km / h, the average relative error of the fusion result is: 1.74%, and the maximum relative error is: 4.18%.

[0130] Embodiment 2 Fusion of radar speed measurement and inertial navigation speed measurement

[0131] In this embodiment, a ground speed radar and an inertial measurement element are selected as the speed measurement module to carry out the fusion of radar speed measurement and inertial navigation speed measurement; an STM32 single-chip microcomputer is selected as the control unit to complete the control and data processing of the system; a serial port screen is selected as the display, which supports touch control to realize human-computer interaction, curve display of speed changes, selection of speed measurement mode and seeding setting, etc. Install according to the installation method in the implementation manner, and complete the corresponding connection with the data cable.

[0132] The measurement object of radar speed measurement is the actual traveling speed of the seeder. Assume that the measured value of radar speed measurement is v k ,; the measurement object of inertial navigation speed measurement is the acceleration of the seeder, and the measured value of inertial navigation speed measurement is a k .

[0133] System model construction

[0134] Before performing Kalman filtering, it is first necessary to clarify the system state and observation model, and determine the state transition matrix A and measurement matrix H. In the fusion of radar speed measurement and inertial navigation speed measurement, the Kalman filter is a two-dimensional Kalman filter.

[0135]

[0136] According to the relationship between speed and acceleration, assuming that the acceleration of the seeder at this moment is equal to that of the seeder at the previous moment, the state equation of

[0137]

[0138] Since the magnitudes of speed and acceleration are directly measured, the measurement matrix

[0139]

[0140] Initialize the Kalman filter. The seeder starts from rest, and set the initial speed estimate Initialize the error covariance matrix

[0141] Set the process noise covariance matrix Q

[0142] The process noise covariance matrix Q represents the error magnitude of the system's prediction of speed and acceleration. A noise variance needs to be set for each state of speed and acceleration, indicating the uncertainty during prediction. The process noise Q of speed 1 , and the process noise Q of acceleration 2 . If the change in speed is slow, then the process noise Q of speed 1 can be set relatively small, indicating that the error in speed prediction is relatively small; if the change in acceleration is large, the process noise Q of acceleration 2 should be set relatively large, and the specific value needs to be adjusted comprehensively according to the sensor and the operating scenario. For the process noise covariance matrix Q of the Kalman filter, since speed and acceleration are independent in the model, the covariance between them is zero.

[0143]

[0144] Set the measurement noise covariance matrix R

[0145] The observation noise covariance matrix R represents the error of the observed data, and the observation noises of radar speed measurement and inertial navigation speed measurement need to be set separately. For radar, there may be certain errors in radar speed measurement due to ranging and angle errors. For inertial navigation, drift will be introduced during the integration process, so the observation noise of the inertial navigation system may be relatively large. According to the magnitude of the observation error, the radar noise covariance R 1 and the inertial navigation measurement noise covariance R 2 can be adjusted.

[0146]

[0147] The working process of integrated speed measurement of radar speed measurement and inertial navigation speed measurement

[0148] The speed measurement system of the seeder is powered on, the system is initialized and system parameters are set. The integrated speed measurement mode of radar speed measurement and inertial navigation speed measurement is selected, the timer interrupt is enabled, and the working frequency of integrated speed measurement is set;

[0149] After power-on, the ground speed radar and inertial measurement components automatically maintain high-frequency data measurement;

[0150] The single-chip microcomputer sends speed measurement commands to the ground speed radar through the serial port in a loop. After the ground speed radar receives the commands, the ground speed radar sends a data packet containing speed measurement data, and the radar speed measurement result v is obtained by parsing the data packet 1 ;

[0151] The single-chip microcomputer reads the calculation result quaternion of the built-in digital motion processor of the inertial measurement component from the inertial measurement component register through IIC communication in a loop; the single-chip microcomputer calculates the quaternion to obtain the tilt angle and acceleration of the seeder.

[0152] After the timer interrupt is triggered, the Kalman filter starts filtering:

[0153] Prediction step. Predict the speed and acceleration at this moment according to the speed and acceleration at the previous moment. In this embodiment Therefore, the predicted value of the speed at this moment is equal to the estimated value of the speed at the previous moment; at the same time, the predicted error covariance matrix is calculated

[0154]

[0155] Correction step. Calculate the Kalman gain K k , correct the state estimate Update the error covariance

[0156]

[0157] Finally, the corrected state estimate output by the Kalman filter The v in k is the integrated result of radar speed measurement and inertial navigation speed measurement, which is an accurate estimate of the traveling speed of the seeder.

[0158] Subsequently, according to the traveling speed of the seeder, the setting of the seeding plant spacing, and the seeding plant spacing, the target speed and motor PWM of the metering unit motor are calculated;

[0159] Output an appropriate PWM to the metering unit motor to adjust the motor speed, and at the same time write information such as the traveling speed and seeding situation of the seeder into the display register, and the display performs data visualization display;

[0160] Meanwhile, when the single-chip microcomputer monitors the sowing status debugging signal from the display screen, it makes corresponding adjustments to the speed measurement system after the interruption is triggered. The user can input the required operating parameters through the display screen, such as starting the operation, theoretical plant spacing, speed measurement settings, etc., and at the same time observe the sowing working status through the display screen.

[0161] In this embodiment, according to the selected equipment and working environment, parameter Q is selected 1 = 0.00038; Q 2 = 0.00010; R 1 = 0.130; R 2 = 0.450, and multiple groups of speed measurements are carried out. The average value of the average relative error of the speed measurement fusion result is: 3.60%. Next Figure 3 Actual case, the reference speed is 3.60 km / h, the average relative error of the fusion result is: 1.78%, and the maximum relative error is: 6.34%.

[0162] Example 3 Fusion of Beidou speed measurement and inertial navigation speed measurement

[0163] The implementation process of the fusion of Beidou speed measurement and inertial navigation speed measurement in Example 3 is the same as that of the fusion of Beidou speed measurement and inertial navigation speed measurement in Example 2, and only the relevant parameters need to be adjusted.

[0164] In this embodiment, according to the selected equipment and working environment, parameter Q is selected 1 = 0.00028; Q 2 = 0.00010; R 1 = 0.120; R 2 = 0.450, and multiple groups of speed measurements are carried out. The average value of the average relative error of the speed measurement fusion result is: 3.25%. Next Figure 4 Actual case, the reference speed is 7.20 km / h, the average relative error of the fusion result is: 1.90%, and the maximum relative error is: 4.71%.

[0165] Example 3

[0166] In Example 3, the ground speed radar is fixedly installed at the tail of the seeder to ensure that there is no obstruction in its signal transmission path; the antenna of the Beidou receiver is installed on the top of the seeder to ensure that the antenna is unobstructed; the inertial measurement element is fixed by using the installation grooves and holes reserved at the central position of the seeder and is fixed to the shockproof mounting frame by screws; each sensor is fixed by using a special mounting frame, and the mounting frame is firmly connected to the seeder structure by bolts and nuts to ensure that each sensor maintains a fixed angle and position consistent with the system reference coordinate system during installation.

[0167] In Embodiment 3, the control unit receives speed measurement data from the ground speed radar, angular velocity and acceleration data collected by the inertial measurement element, and speed data output by the Beidou receiver. Among them, the ground speed radar uses a speed measurement method based on the Doppler effect, that is, multiplying the speed of the moving object by two and then multiplying it by the frequency of the transmitted signal, and then dividing the obtained product by the speed of light to obtain the Doppler frequency shift. After deducting the gravitational acceleration, the inertial measurement element accumulates and integrates the product of the collected acceleration and the sampling time interval with the speed at the previous moment to obtain the current speed. The control unit uses the Kalman filter algorithm. In its prediction stage, it calculates the predicted state by multiplying the previous state by the state transition matrix, and calculates the predicted error covariance by combining the previous moment error covariance and the process noise covariance. In the correction stage, it determines the Kalman gain according to the measurement matrix and the measurement noise covariance, and then performs a weighted average on the predicted state and the current measurement value to update the state estimate and the error covariance, so as to calculate the accurate traveling speed of the seeder by fusing multi-source data.

[0168] Embodiment 4

[0169] In Embodiment 4, the installation structures of the sensors are designed for vibration resistance. In addition to being fixed to the tail of the seeder, a vibration isolation device is provided in front of the ground speed radar. The Beidou receiver antenna is installed in the high-position area on the top of the seeder and fixed with a waterproof and dustproof cover. The inertial measurement element is installed in the center of the seeder and fastened to the mounting frame through a shockproof bracket. The adopted mounting frame design reserves a shockproof isolation layer, special mounting holes and shockproof screws to ensure that in a high mechanical vibration environment, the installation angles, positions and orientations of the sensors all meet the system requirements.

[0170] In Embodiment 4, after the control unit collects the output data of each sensor, it fuses multi-source data through the same data processing flow as in Embodiment 1. Specifically, it includes: calculating the speed measurement data using the Doppler effect of the ground speed radar, obtaining the speed from the acceleration data of the inertial measurement element through integral operation, and directly parsing the Beidou receiver message to obtain the speed; then using the Kalman filter algorithm. First, in the prediction stage, it calculates the predicted state according to the previous moment state and the state transition matrix, and at the same time calculates the predicted error covariance by combining the process noise covariance; subsequently, in the correction stage, it determines the Kalman gain according to the measurement matrix and the measurement noise covariance, and obtains the corrected state estimate by weighted averaging the predicted state and the current measurement value, and updates the error covariance, so as to still obtain a stable traveling speed of the seeder in a high vibration environment.

[0171] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, any modifications, equivalent replacements and improvements made within the spirit and principle of the present invention shall all be covered within the protection scope of the present invention.

Claims

1. A radar Beidou inertial navigation multi-source data fusion planter precise speed measurement system, characterized in that: The system comprises: a seed meter driven by an electric motor; A speed monitoring module, the speed monitoring module comprising a ground speed radar, an inertial measurement unit and a Beidou receiver; A display screen; A control unit, whose circuit is connected to the speed monitoring module, the display screen and the motor of the seed meter, and the control unit calculates the precise travel speed of the seed meter based on the speed data obtained by the speed monitoring module and the selected speed measurement mode, wherein the speed measurement mode includes a fusion speed measurement mode of radar speed measurement and Beidou speed measurement, a fusion speed measurement mode of radar speed measurement and inertial navigation speed measurement, and a fusion speed measurement mode of Beidou speed measurement and inertial navigation speed measurement, and adjusts the speed of the seed meter motor according to the precise travel speed.

2. The system according to claim 1, characterized in that In the speed monitoring module: The ground speed radar is installed at the tail of the seed drill to ensure that the signal path of the ground speed radar is not blocked by the structure or equipment of the seed drill; The antenna of the Beidou receiver is installed on the top of the planter or at a higher position to ensure that the antenna is not blocked; The inertial measurement unit is installed at the center of the planter, placed horizontally and aligned with the reference coordinate system of the planter.

3. The system according to claim 1, characterized in that The ground speed radar measures speed based on the Doppler effect, wherein the Doppler frequency shift is described in words as follows: the speed of a moving object is doubled, then multiplied by the frequency of the transmitted signal, and the value obtained by dividing the product by the speed of light is the Doppler frequency shift.

4. The system according to claim 1, characterized in that The inertial measurement element includes a gyroscope and an accelerometer, wherein the gyroscope is used to collect angular velocities in three axes, and the accelerometer is used to collect accelerations in three axes; the inertial measurement element is also equipped with a 16-bit analog-to-digital converter, which is used to collect signals from the gyroscope and the accelerometer, wherein after deducting the gravity acceleration component, the current speed is obtained by integration operation, and the integration operation is described in words as follows: the speed at the current moment is equal to the speed at the previous moment plus the acceleration at this moment after removing the influence of gravity acceleration multiplied by the measurement time interval.

5. The system according to claim 1, wherein: The system is provided with a Kalman filter module for fusing speed data from a ground speed radar, a Beidou receiver and an inertial measurement element; the Kalman filter module is divided into a prediction stage and a correction stage, the prediction stage is described in words as: the predicted state at the current moment is calculated by multiplying the state at the previous moment by the state transfer matrix, and the predicted error covariance is calculated by combining the error covariance matrix at the previous moment with the process noise covariance; the correction stage is described in words as: according to the difference between the predicted state and the current measured value, the Kalman gain is determined by the measurement matrix and the measurement noise covariance, and then the predicted state and the measured value are weighted averaged by the Kalman gain to obtain a corrected state estimate, and the error covariance matrix is ​​updated.

6. The system according to claim 1, characterized in that The display screen is connected to the control unit, and the display screen is provided with a data input part for inputting sowing operation parameters and a data output part for displaying speed data and sowing status. The control unit calculates the precise travel speed of the seeder based on the speed data obtained by the speed monitoring module and the selected speed measurement mode, and adjusts the speed of the seed metering device motor accordingly.

7. A method for accurately measuring the speed of a seed drill by fusion of radar and Beidou inertial navigation multi-source data, characterized in that: The method comprises the following steps: (a) Collect velocity data, inertial data and positioning data from ground speed radar, inertial measurement unit and Beidou receiver respectively; (b) selecting one of the fusion speed measurement mode of radar speed measurement and Beidou speed measurement, the fusion speed measurement mode of radar speed measurement and inertial navigation speed measurement, or the fusion speed measurement mode of Beidou speed measurement and inertial navigation speed measurement; (c) processing the data based on the collected data and the selected speed measurement mode to calculate the precise travel speed of the planter; (d) adjusting the rotation speed of the seed metering device motor according to the precise travel speed.

8. The method according to claim 7, characterized in that The data collection in step (a) comprises: The speed data is collected from a ground speed radar installed at the rear of the planter with an unobstructed signal path; Inertial data is collected from an inertial measurement unit mounted at the center of the planter and aligned with the reference coordinate system; Positioning data is collected from a BeiDou receiver mounted on top of the planter without obstruction.

9. The method according to claim 7, characterized in that Calculating the precise speed in step (c) further includes fusing the data collected by the ground speed radar, Beidou receiver and inertial measurement element using a Kalman filter algorithm, wherein the Kalman filter algorithm includes a prediction phase and a correction phase, wherein: In the prediction stage, the predicted state at the current moment is obtained by multiplying the state at the previous moment by the state transfer matrix, and the prediction error covariance is calculated by combining the error covariance matrix at the previous moment with the process noise covariance, and the process noise covariance is described in words as representing the uncertainty in the system state transfer process; In the correction stage, the Kalman gain is calculated by the measurement matrix and the measurement noise covariance, and the predicted state and the current measurement value are weighted averaged using the Kalman gain to obtain the corrected state estimate, and the error covariance matrix is ​​updated at the same time.

10. The method according to claim 7, characterized in that Calculating the precise travel speed in step (c) further comprises: For the speed data collected by the ground speed radar, the speed is measured based on the Doppler effect. The Doppler frequency shift is described in words as follows: the speed of the moving object is doubled, and then multiplied by the frequency of the transmitted signal. The value obtained by dividing the product by the speed of light is the Doppler frequency shift; For the inertial data collected by the inertial measurement element, after deducting the gravity acceleration component, the product of the acceleration at the current moment and the time interval is added to the velocity value at the previous moment according to the measurement time interval, and the velocity at the current moment is obtained by integration operation.

Images