Agricultural machinery automatic driving method based on Beidou satellite and inertial navigation fusion

Abstract

The invention relates to the technical field of Beidou satellites, in particular to an agricultural machinery automatic driving method based on Beidou satellite and inertial navigation fusion. The method comprises the following steps: determining a farmland area needing to be operated by an agricultural machine, and obtaining starting point and terminal point positions; according to the method, weighted summation is carried out by using the diagonal matrix according to the weight coefficient to obtain the weight matrix, the position increment of the reference point is calculated through combination of the matrix, the weight matrix and the carrier phase difference value of the Beidou satellite, and high-precision positioning of the final reference point position is carried out through combination of historical data and the position increment of the reference point. Errors can be reduced through the position increment of the reference point, so that the precision of the position calculation of the reference point is improved, meanwhile, the high-precision information can be more effectively utilized through reasonable weighting of the weight matrix, the accuracy of the position increment calculation is further improved, and the overall positioning precision is further improved.

Description

Technical Field

[0001] The present invention relates to the field of Beidou satellite technology, and in particular to an automatic driving method for agricultural machinery based on the fusion of Beidou satellite and inertial navigation. Background Art

[0002] After years of construction and development, the BeiDou satellite navigation system has continuously improved its positioning accuracy. Currently, the BeiDou system's positioning accuracy has reached the meter level globally, and even sub-meter accuracy in the Asia-Pacific region, providing relatively accurate location information for autonomous driving of agricultural machinery.

[0003] The Beidou satellite system has global coverage. Whether in vast plains or remote mountainous areas, agricultural machinery can receive Beidou satellite signals, achieving global positioning and navigation, and providing guarantees for agricultural machinery operations in different geographical environments. Although Beidou satellite positioning technology has high positioning accuracy, it is easily affected by the external environment, such as signal obstruction. For example, when agricultural machinery uses the centimeter-level positioning accuracy of the Beidou satellite system for automatic driving, when the agricultural machinery enters a straight operating path surrounded by dense vegetation, the dense vegetation will intermittently block the satellite signal, causing the Beidou satellite signal to be temporarily lost, and the agricultural machinery cannot quickly adapt to the signal changes, thereby reducing the centimeter-level positioning accuracy of the agricultural machinery, resulting in deviations in the operating accuracy of the agricultural machinery around dense vegetation areas, greatly reducing the operating accuracy, flexibility, and efficiency of the agricultural machinery around dense vegetation areas. Therefore, we provide an agricultural machinery automatic driving method based on the fusion of Beidou satellites and inertial navigation. Summary of the Invention

[0004] The purpose of the present invention is to provide an agricultural machinery automatic driving method based on the fusion of Beidou satellite and inertial navigation to solve the problems raised in the above background technology.

[0005] To achieve the above objectives, the present invention provides an automatic driving method for agricultural machinery based on the fusion of Beidou satellite and inertial navigation, comprising the following method steps:

[0006] S1. Obtain a farmland area and set the starting and ending points. Draw an AB straight line operation path through the starting and ending points and segment it into multiple 3D points. Then segment the agricultural machinery into multiple 3D points. When the agricultural machinery automatically drives to the starting point, find the 3D point where the segmented agricultural machinery 3D point overlaps with the starting point. Use the overlapping 3D point as the agricultural machinery reference point and calculate the geometric distance between the reference point and the Beidou satellite.

[0007] S2. Obtain historical data and calculate the carrier phase observation value of the Beidou satellite received by the reference point based on the geometric distance between the reference point and the Beidou satellite. Then calculate the carrier phase difference value of the Beidou satellite to obtain the position increment of the reference point, and perform high-precision positioning of the final reference point position.

[0008] S3. Calculate the operating accuracy distance of the straight-line path through the final reference point position, use the operating accuracy distance of the straight-line path to determine whether the operating accuracy of the agricultural machinery has deviated, and then detect whether there is a dense vegetation area around the segmented three-dimensional point straight-line operating path. If so, before the agricultural machinery enters the segmented three-dimensional point straight-line operating path blocked by the dense vegetation area, record the three-dimensional point position and speed of the agricultural machinery to calculate the current reference point position, and then use the current reference point position to determine whether the operating accuracy of the agricultural machinery in the straight-line path around the dense vegetation area has deviated.

[0009] As a further improvement of the present technical solution, S2.2 in S2 inputs the carrier phase observation value of the Beidou satellite received at the reference point as an input value into the machine learning model. The machine learning model learns the input data and outputs a weight coefficient. The weight matrix is ​​obtained by weighted summing the diagonal matrix according to the weight coefficient. The position increment of the reference point is calculated by combining the matrix, the weight matrix and the carrier phase difference value of the Beidou satellite. The high-precision positioning of the final reference point position is performed by combining the historical data with the position increment of the reference point. The weight matrix obtained by using the diagonal matrix and the weight coefficient can be reasonably weighted for each data source (such as Beidou satellite signals, inertial navigation data) according to the reliability and accuracy of different data sources (such as Beidou satellite signals, inertial navigation data) in different environments and working conditions. For example, in open areas, the Beidou satellite signal has high accuracy and can be given a higher weight; in areas with dense vegetation, the inertial navigation data is relatively more stable and its weight can be increased. This can give full play to the advantages of each data source, reduce errors, and thus improve the accuracy of the reference point position calculation.

[0010] As a further improvement of the present technical solution, S3.1 in S3 calculates the operating accuracy distance of the straight path through the final reference point position and the coordinate position of the three-dimensional point of the straight path, and uses the operating accuracy distance of the straight path to determine whether the operating accuracy of the agricultural machinery deviates. When the operating accuracy distance of the straight path exceeds the set operating accuracy distance range value, it is determined that the operating accuracy of the agricultural machinery deviates, and automatically adjusted through the square disk of the agricultural machinery, and the sensor installed on the agricultural machinery is used to detect in real time whether there is a dense vegetation area around the segmented three-dimensional point straight line operating path.

[0011] As a further improvement of the present technical solution, in S3.2 of the above S3, the three-dimensional position and speed of the agricultural machinery are recorded before the agricultural machinery enters the segmented three-dimensional point straight line operation path blocked by the dense vegetation area. When the agricultural machinery enters the segmented three-dimensional point straight line operation path blocked by the dense vegetation area, it is detected that the satellite signal is temporarily lost, the three-dimensional position of the agricultural machinery is used as the reference point position before the blockage, and the speed of the agricultural machinery is used as the previous speed of the agricultural machinery, and the acceleration of the agricultural machinery is provided by inertial navigation. The speed of the agricultural machinery is calculated by integrating and fusing the acceleration and speed of the agricultural machinery, and then the speed of the agricultural machinery is calculated by the speed of the agricultural machinery. The displacement of the agricultural machinery is calculated by integration, and the current reference point position is calculated using the displacement of the agricultural machinery and the position of the reference point before occlusion. When vegetation blocks the satellite signal, Beidou satellite positioning may not work properly, resulting in loss of positioning information. Inertial navigation can be independent of satellite signals and continuously provide acceleration information of the agricultural machinery. The speed is calculated by integrating the acceleration, and the displacement is obtained by integrating the speed. This can continuously calculate the position change of the agricultural machinery during the interruption of the satellite signal, fill the positioning gap when the satellite signal is missing, and enable the automatic driving of the agricultural machinery to always grasp the approximate position of the agricultural machinery, ensuring the continuity of positioning.

[0012] Compared with the prior art, the present invention has the following beneficial effects:

[0013] 1. This agricultural machinery automatic driving method based on the fusion of Beidou satellite and inertial navigation inputs the carrier phase observation value of the Beidou satellite received at the reference point as the input value into the machine learning model. The machine learning model learns the input data and outputs the weight coefficient. The weight matrix is ​​obtained by weighted summation based on the diagonal matrix according to the weight coefficient. The position increment of the reference point is calculated by combining the matrix, the weight matrix and the carrier phase difference value of the Beidou satellite. The final reference point position is positioned with high precision by combining the historical data with the position increment of the reference point. The position increment of the reference point can reduce the error, thereby improving the accuracy of the reference point position calculation. At the same time, the reasonable weighting of the weight matrix can more effectively utilize this high-precision information, further improve the accuracy of the position increment calculation, and thus improve the overall positioning accuracy.

[0014] 2. This agricultural machinery automatic driving method based on the fusion of Beidou satellite and inertial navigation records the three-dimensional position and speed of the agricultural machinery before the agricultural machinery enters the segmented three-dimensional point straight line operation path blocked by dense vegetation. When the agricultural machinery enters the segmented three-dimensional point straight line operation path blocked by dense vegetation, it detects that the satellite signal is temporarily lost, and the three-dimensional position of the agricultural machinery is used as the reference point position before the blockage, and the agricultural machinery speed is used as the previous agricultural machinery speed. The agricultural machinery acceleration is provided by inertial navigation, and the agricultural machinery acceleration and agricultural machinery speed are integrated and integrated to calculate the agricultural machinery speed. The agricultural machinery displacement is then calculated by integrating the agricultural machinery speed. The current reference point position is calculated using the agricultural machinery displacement and the reference point position before the blockage. Through the fusion of inertial navigation and satellite positioning, the agricultural machinery can quickly adapt to signal changes and quickly switch positioning methods when the satellite signal is lost and restored, ensuring that the agricultural machinery can flexibly operate in different environments, thereby improving the flexibility and efficiency of agricultural machinery operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flowchart of the overall steps of the present invention. DETAILED DESCRIPTION

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0017] Example 1

[0018] This invention provides an agricultural machinery automatic driving method based on the fusion of Beidou satellite and inertial navigation, please refer to Figure 1 , comprising the following method steps:

[0019] S1. Determine the farmland area where the agricultural machinery needs to operate and obtain the starting and ending points. Draw an AB straight line operation path through the starting and ending points and segment it into multiple 3D points. Then segment the agricultural machinery into multiple 3D points. When the agricultural machinery automatically drives to the starting point, find the 3D point where the segmented agricultural machinery 3D points overlap with the starting point. Use the overlapping 3D point as the agricultural machinery reference point and calculate the geometric distance between the reference point and the Beidou satellite.

[0020] S1 has the following specific steps:

[0021] S1.1. Determine the farmland area where the agricultural machinery needs to operate using map data, and set point A (starting point) and point B (end point) in the farmland area where the operation is required. Then, obtain the longitude and latitude coordinates of point A and point B using the mushroom-shaped satellite antenna installed on the autonomous agricultural machinery. Convert the longitude and latitude coordinates of point A and point B into the positions of point A and point B. Draw an AB straight line operation path based on the positions of point A and point B. Then, divide the farmland area where the operation is required into multiple straight line operation paths parallel to the AB straight line operation path. Divide all the straight line operation paths into multiple three-dimensional points, and record the coordinate positions (x n ,y n , z n ), then segment the agricultural machine into multiple three-dimensional points, and the agricultural machine automatically drives to point A. Then find the three-dimensional point where the segmented three-dimensional points of the agricultural machine coincide with the starting point, and use the three-dimensional point as the reference point of the agricultural machine;

[0022] When it is known that the agricultural machinery starts to operate automatically along the divided three-dimensional point straight line operation path, S1.2 obtains the Beidou satellite position (x i ,y i , z i ), base station position (x base,b ,y base,b , z base,b ) and the reference point position (x mobile,m ,y mobile,m , z mobile,m ) (usually refers to the real-time location of agricultural machinery), where x i Refers to the x-axis coordinate position of the Beidou satellite, y i Refers to the y-axis coordinate position of the Beidou satellite, z i Refers to the z-axis coordinate position of the Beidou satellite, x base,b Refers to the bth x-axis coordinate position of the base station, y base,b Refers to the bth y-axis coordinate position of the base station, z base,b Refers to the bth z-axis coordinate position of the base station, x base The subscript base refers to the base station, mobile,m Refers to the mth x-axis coordinate position of the reference point, y mobile,m Refers to the mth y-axis coordinate position of the reference point, z mobile,m Refers to the mth z-axis coordinate position of the reference point, x mobile The subscript mobile refers to the reference point.

[0023] Calculate the geometric distance between the reference station and the Beidou satellite based on the Beidou satellite position and the reference station position Among them, the geometric distance ρ between the reference station and the satellite b,iThe unit is meter, and then the geometric distance between the reference point and the Beidou satellite is calculated based on the Beidou satellite position and the reference point position. Among them, the geometric distance ρ between the reference point and the satellite m,i The unit is meter;

[0024] S2. Obtain historical data and calculate the carrier phase observation value of the Beidou satellite received by the reference point based on the geometric distance between the reference point and the Beidou satellite. Then calculate the carrier phase difference value of the Beidou satellite to obtain the position increment of the reference point, and perform high-precision positioning of the final reference point position.

[0025] S2 has the following specific steps:

[0026] S2.1. Obtain historical data and extract BeiDou satellite signal wavelength λ from the historical data i , integer ambiguity N between the reference station and BeiDou satellite b，i , speed of light c, receiver clock error δt at the reference station base , the noise and error in the reference station observations ∈ b，i , through the carrier phase observation equation in RTK high-precision satellite positioning technology, according to the Beidou satellite signal wavelength, integer ambiguity between the reference station and the Beidou satellite, the speed of light, the receiver clock error of the reference station, the noise and error in the reference station observation value, and the geometric distance ρ between the reference station and the Beidou satellite b，i Combined with the calculated carrier phase observation value φ of the Beidou satellite received by the base station b，i =ρ b，i +λ i ×N b，i +c×δt base +∈ b，i , where φ b，i The unit is cycle, N b，i Unitless, λ i The unit of is meter, the unit of light speed is meter / second, the speed of light is a constant, about 299 792458m / s, δt base The unit is seconds, ∈ b，i The unit is cycle;

[0027] Similarly, the integer ambiguity N between the reference point and the BeiDou satellite is extracted from the historical data. m，i , receiver clock error δt at the reference point mobile , the noise and error in the observation values ​​of the moving station ∈ m，iThe carrier phase observation equation in RTK high-precision satellite positioning technology is used to calculate the carrier phase observation value φ of the Beidou satellite received by the reference point based on the Beidou satellite signal wavelength, the integer ambiguity between the reference point and the Beidou satellite, the speed of light, the receiver clock error of the reference point, the noise and error in the dynamic station observation value, and the geometric distance ρm,i between the reference point and the Beidou satellite. m，i =ρ m,i +λ i ×N m，i +c×δt mobile +∈ m，i , where φ m，i The unit is cycle, N m,i Unitless, δt mobile The unit is seconds, ∈ m，i The unit is cycle;

[0028] Historical data includes BeiDou satellite signal wavelength λ i (referring to the wavelength of the i-th satellite signal), the integer ambiguity N between the reference point and the BeiDou satellite m，i , integer ambiguity N between the reference station and BeiDou satellite b，i , speed of light c, receiver clock error δt at the reference station base , receiver clock error δt at the reference point mobile , the noise and error in the reference point observations ∈ m,i , the noise and error in the reference station observations ∈ b，i , the last reference point position S mobile,m-1 =(x mobile,m-1 ,y mobile,m-1 , z mobile,m-1 );

[0029] S2.2. Calculate the carrier phase difference of the Beidou satellite by using the carrier phase observation value of the Beidou satellite received by the reference station in S2.1 and the carrier phase observation value of the Beidou satellite received by the reference point, Δφ=φ b，i -φ m，i , where the unit of the BeiDou satellite carrier phase difference is cycle. This formula shows that the carrier phase observations at the reference point and the base station are subtracted to eliminate common errors (such as satellite clock error and atmospheric delay). The BeiDou satellite carrier phase difference is used to eliminate common errors and improve positioning accuracy.

[0030] By using the BeiDou satellite position (x i ,y i , z i ), reference point position (x mobile,m ,y mobile,m , z mobile,m ), the geometric distance ρ between the reference point and the BeiDou satellitem,i Design the matrix H and input the carrier phase observation value of the Beidou satellite received at the reference point into the machine learning model. The machine learning model learns the input data and outputs the weight coefficient ω e , the weight matrix W is obtained by weighted summation of the diagonal matrix according to the weight coefficient, and the position increment of the reference point ΔP is calculated by combining the matrix, the weight matrix and the carrier phase difference value of the Beidou satellite. T WH) -1 H T WΔφ, where the unit of the reference point position increment is meter. The position increment is the correction value of the reference point relative to the previous position, which is used to update the position of the reference point. The previous reference point position S is then extracted from the historical data of S2.1. mobile,m-1 The final reference point position S is obtained by combining the previous reference point position with the reference point position increment to perform high-precision positioning of the final reference point position. mobile =S mobile,m-1 +ΔP;

[0031] Design matrix specific algorithm formula:

[0032]

[0033] Among them, x q Refers to the qth x-axis coordinate of the Beidou satellite, y q Refers to the qth y-axis coordinate of the Beidou satellite, z q Refers to the qth z-axis coordinate of the Beidou satellite, ρ m，q Refers to the qth geometric distance between the reference point and the BeiDou satellite. This formula is used to design the matrix. Refers to the unit vector component in the X direction between the reference point and the Beidou satellite. Refers to the unit vector component in the y direction between the reference point and the Beidou satellite. It refers to the unit vector component between the reference point and the Beidou satellite in the z direction, reflecting the impact of satellite geometric distribution on positioning accuracy;

[0034] The specific algorithm formula of the weight matrix:

[0035]

[0036] in, ω 1 refers to the first weight coefficient, ω 2 refers to the second weight coefficient, ω e refers to the e-th weight coefficient;

[0037] S3. Calculate the operating accuracy distance of the straight path using the final reference point position, use the operating accuracy distance of the straight path to determine whether the agricultural machinery's operating accuracy has deviated, then detect whether there is an area of ​​dense vegetation around the segmented three-dimensional point straight line operating path. If so, before the agricultural machinery enters the segmented three-dimensional point straight line operating path blocked by the dense vegetation area, record the three-dimensional point position and speed of the agricultural machinery to calculate the current reference point position, and then use the current reference point position to determine whether the agricultural machinery's operating accuracy has deviated from the straight path around the dense vegetation area.

[0038] S3 specific steps are as follows:

[0039] S3.1. Calculate the operating accuracy distance j1 of the straight path using the final reference point position in S2.2 and the coordinate position of the three-dimensional point on the straight path in S1.1. Use the operating accuracy distance of the straight path and the set operating accuracy distance range value to determine whether the agricultural machinery operating accuracy has deviated. The set operating accuracy distance range value is [0, 2.5] centimeters. When the operating accuracy distance of the straight path exceeds the set operating accuracy distance range value, it is determined that the agricultural machinery operating accuracy has deviated, and the square plate of the agricultural machinery is automatically adjusted until the operating accuracy distance of the straight path is within the set operating accuracy distance range value. The agricultural machinery continues to automatically operate along the segmented three-dimensional point straight line operating path. Since dense vegetation areas can intermittently block satellite signals, causing temporary loss of satellite signals, when the agricultural machinery is operating automatically, the sensors installed on the agricultural machinery are used to detect in real time whether there are dense vegetation areas around the segmented three-dimensional point straight line operating path.

[0040] When a dense vegetation area is detected, S3.2, before the agricultural machine enters the segmented three-dimensional point straight line operation path blocked by the dense vegetation area, the three-dimensional point position and speed of the agricultural machine are recorded. When the agricultural machine enters the segmented three-dimensional point straight line operation path blocked by the dense vegetation area, it is detected that the satellite signal is temporarily lost, and the three-dimensional point position of the agricultural machine is used as the reference point position ZDS before the blockage, and the agricultural machine speed is used as the previous agricultural machine speed. v 0, and provide the agricultural machinery acceleration a at time T through inertial navigation, through the agricultural machinery acceleration a and agricultural machinery speed v The speed of the agricultural machine at time T is calculated by integrating the speed v(τ) = v0 + ∫α(τ)dτ at time τ, and the displacement of the agricultural machine is calculated by integrating the speed v(τ) at time τ to obtain the displacement ΔS = ∫v(τ)dτ. The displacement of the agricultural machine and the position of the reference point before occlusion are used to calculate the current reference point position DS = ΔS + ZDS.

[0041] The operating accuracy distance j12 of the straight path is calculated again using the current reference point position and the coordinate position of the three-dimensional point of the straight path in S1.1 (it must be the coordinate position of the three-dimensional point with the shortest distance to the current reference point position). The operating accuracy distance j12 of the straight path is compared with the set operating accuracy distance range value to determine whether the operating accuracy of the straight path of the agricultural machinery around the dense vegetation area deviates. When the operating accuracy distance j12 of the straight path exceeds the set operating accuracy distance range value, the operating accuracy of the straight path of the agricultural machinery around the dense vegetation area deviates, and the square disk of the agricultural machinery is automatically adjusted to stop when the operating accuracy distance j12 of the straight path is within the set operating accuracy distance range value.

[0042] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and improvements fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. An autonomous driving method for agricultural machinery based on the integration of Beidou satellite navigation and inertial navigation, characterized by: The method comprises the following steps: S1. Determine the farmland area where the agricultural machinery needs to operate and obtain the starting and ending points. Draw an AB straight line operation path through the starting and ending points and segment it into multiple 3D points. Then segment the agricultural machinery into multiple 3D points. When the agricultural machinery automatically drives to the starting point, find the 3D point where the segmented agricultural machinery 3D points overlap with the starting point. Use the overlapping 3D point as the agricultural machinery reference point and calculate the geometric distance between the reference point and the Beidou satellite. S2. Obtain historical data and calculate the carrier phase observation value of the Beidou satellite received by the reference point based on the geometric distance between the reference point and the Beidou satellite. Then calculate the carrier phase difference value of the Beidou satellite to obtain the position increment of the reference point, and perform high-precision positioning of the final reference point position. S3. Calculate the operating accuracy distance of the straight-line path through the final reference point position, use the operating accuracy distance of the straight-line path to determine whether the operating accuracy of the agricultural machinery has deviated, and then detect whether there is a dense vegetation area around the segmented three-dimensional point straight-line operating path. If so, before the agricultural machinery enters the segmented three-dimensional point straight-line operating path blocked by the dense vegetation area, record the three-dimensional point position and speed of the agricultural machinery to calculate the current reference point position, and then use the current reference point position to determine whether the operating accuracy of the agricultural machinery in the straight-line path around the dense vegetation area has deviated.

2. The method for automatic driving of agricultural machinery based on the fusion of Beidou satellite and inertial navigation according to claim 1 is characterized in that: The S1 method steps are as follows: S1.

1. Determine the farmland area where the agricultural machinery needs to operate and set the starting and end points. Then obtain the starting and end points. Draw an AB straight line operation path through the starting and end points and divide it into multiple three-dimensional points. Record the coordinate positions of the three-dimensional points on the straight line path. Then divide the agricultural machinery into multiple three-dimensional points. The agricultural machinery automatically drives to the starting point and finds the three-dimensional point where the divided three-dimensional points of the agricultural machinery overlap with the starting point. This three-dimensional point is used as the reference point of the agricultural machinery.

3. The method for automatic driving of agricultural machinery based on the fusion of Beidou satellite and inertial navigation according to claim 2 is characterized in that: In S1, S1.2 obtains the Beidou satellite position, the reference station position and the reference point position, calculates the geometric distance between the reference station and the Beidou satellite based on the Beidou satellite position and the reference station position, and then calculates the geometric distance between the reference point and the Beidou satellite based on the Beidou satellite position and the reference point position.

4. The method for automatic driving of agricultural machinery based on the fusion of Beidou satellite and inertial navigation according to claim 3 is characterized in that: The S2 specifically includes the following steps: S2.

1. Obtain historical data and calculate the carrier phase observation value of the Beidou satellite received by the base station based on the carrier phase observation equation in RTK high-precision satellite positioning technology, based on the historical data and the geometric distance between the base station and the Beidou satellite.

5. The method for automatic driving of agricultural machinery based on the fusion of Beidou satellite and inertial navigation according to claim 4 is characterized in that: In S2.1, the carrier phase observation value of the Beidou satellite received by the reference point is calculated based on historical data and the geometric distance between the reference point and the Beidou satellite using the carrier phase observation equation in the RTK high-precision satellite positioning technology.

6. The method for automatic driving of agricultural machinery based on the fusion of Beidou satellite and inertial navigation according to claim 5 is characterized in that: S2.2 in S2, calculating the carrier phase difference value of the Beidou satellite through the carrier phase observation value of the Beidou satellite received by the reference station and the carrier phase observation value of the Beidou satellite received by the reference point; The matrix is ​​designed based on the Beidou satellite position, reference point position, and the geometric distance between the reference point and the Beidou satellite.

7. The method for automatic driving of agricultural machinery based on the fusion of Beidou satellite and inertial navigation according to claim 6 is characterized in that: In S2.2 of the above S2, the carrier phase observation value of the Beidou satellite received at the reference point is input as an input value into the machine learning model. The machine learning model learns the input data and outputs a weight coefficient. The weight matrix is ​​obtained by weighted summation based on the diagonal matrix according to the weight coefficient. The position increment of the reference point is calculated by combining the matrix, the weight matrix and the carrier phase difference value of the Beidou satellite. The final reference point position is positioned with high precision by combining the historical data with the position increment of the reference point to obtain the final reference point position.

8. The method for automatic driving of agricultural machinery based on the fusion of Beidou satellite and inertial navigation according to claim 7 is characterized in that: The S3 specifically includes the following steps: S3.

1. Calculate the operating accuracy distance of the straight path through the final reference point position and the coordinate position of the three-dimensional points on the straight path. Use the operating accuracy distance of the straight path to determine whether the operating accuracy of the agricultural machinery has deviated. When the operating accuracy distance of the straight path exceeds the set operating accuracy distance range value, it is determined that the operating accuracy of the agricultural machinery has deviated, and the square disk of the agricultural machinery is automatically adjusted. The sensors installed on the agricultural machinery are used to detect in real time whether there is a dense vegetation area around the segmented three-dimensional point straight line operating path.

9. The method for automatic driving of agricultural machinery based on the fusion of Beidou satellite and inertial navigation according to claim 8 is characterized in that: In S3.2 of the above S3, before the agricultural machinery enters the segmented three-dimensional point straight line operation path blocked by the dense vegetation area, the three-dimensional point position and the speed of the agricultural machinery are recorded. When the agricultural machinery enters the segmented three-dimensional point straight line operation path blocked by the dense vegetation area, it is detected that the satellite signal is temporarily lost, the three-dimensional point position of the agricultural machinery is used as the reference point position before the blockage, and the speed of the agricultural machinery is used as the previous agricultural machinery speed. The agricultural machinery acceleration is provided by inertial navigation, and the speed of the agricultural machinery is calculated by integrating and fusing the acceleration and speed of the agricultural machinery. The displacement of the agricultural machinery is calculated by integrating the speed of the agricultural machinery, and the current reference point position is calculated using the displacement of the agricultural machinery and the reference point position before the blockage.

10. The method for automatic driving of agricultural machinery based on the fusion of Beidou satellite and inertial navigation according to claim 9, characterized in that: In S3, S3.2 uses the current reference point position and the coordinate position of the three-dimensional point of the straight path to recalculate the operation accuracy distance j12 of the straight path, and uses the operation accuracy distance j12 of the straight path to determine whether the operation accuracy of the straight path of the agricultural machinery deviates around the dense vegetation area.

Images