Intelligent agricultural machinery navigation control system based on Beidou high-precision positioning

Abstract

The application discloses a kind of intelligent agricultural machinery navigation control systems based on Beidou high-precision positioning in the field of agricultural machinery navigation technology, including data acquisition module, for obtaining Beidou observation value, unmanned aerial vehicle observation value and pose estimation value;State prediction module is used to carry out state prediction according to pose estimation value;Positioning update module is used to carry out difference analysis with unmanned aerial vehicle observation value as reference, to adjust the current state and observation value of navigation control system and update positioning result;The state prediction module and positioning update module alternately execute state prediction and positioning result update.The application controls the positioning state of navigation control system according to unmanned aerial vehicle observation value, in the fusion positioning state, based on unmanned aerial vehicle observation value Beidou observation difference and pose estimation difference are obtained, and fusion weight is dynamically distributed, to update positioning result, improve the self-adaptive adjustment ability of system to actual environment change, ensure the continuity and precision of positioning result.

Description

Technical Field

[0001] This invention relates to the field of agricultural machinery navigation, and specifically to an intelligent agricultural machinery navigation and control system based on BeiDou high-precision positioning. Background Technology

[0002] With the development of precision agriculture, the demand for high-precision and high-reliability positioning and navigation in intelligent agricultural machinery is increasing. Currently, most agricultural machinery navigation systems use BeiDou satellite positioning technology to achieve autonomous navigation. However, in actual field operations, agricultural machinery often travels near ridges, trees, or buildings, and BeiDou signals are easily interfered with by factors such as obstruction and multipath effects, leading to a decrease in positioning accuracy or even signal loss in systems that rely solely on BeiDou positioning.

[0003] While existing agricultural machinery navigation and control systems have attempted to integrate data from inertial and visual sensors to compensate for the shortcomings of BeiDou positioning, these integration strategies are often fixed or have simple switching logic. For example, when BeiDou signal quality fluctuates, they lack dynamic adaptability to environmental changes. When the observation quality of a particular sensor repeatedly changes around a threshold, it can easily cause the navigation system to jump frequently between various positioning modes, leading to control command jitter and affecting operational accuracy and stability. Furthermore, when multiple sensors malfunction or the effective data frame rate is too low, there is a lack of effective safety shutdown mechanisms, posing a safety hazard. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent agricultural machinery navigation and control system based on BeiDou high-precision positioning, which solves the problem that existing agricultural machinery navigation and control systems are unable to improve the accuracy of positioning results by combining multi-source data.

[0005] The present invention achieves the above objectives through the following technical solutions: An intelligent agricultural machinery navigation and control system based on BeiDou high-precision positioning includes: The data acquisition module is used to obtain BeiDou observation values, UAV observation values, and pose estimation values ​​based on BeiDou positioning data, UAV image positioning data, and dead reckoning data. The state prediction module is used to predict the state based on the pose estimation value in order to obtain the prior state estimate and the prior state covariance matrix. The positioning update module is used to perform difference analysis based on UAV observations to adjust the current state and observations of the navigation control system, and to update the positioning results based on the prior state covariance matrix and prior state estimation. The state prediction module and the positioning update module alternately perform state prediction and positioning result update. The initial state of the navigation control system is BeiDou positioning state. When the BeiDou observation difference is less than the first threshold, the navigation control system maintains the BeiDou positioning state and updates the positioning result using the BeiDou observation value. When the BeiDou observation difference is not less than the first threshold, the navigation control system enters the fusion positioning state, weightedly fuses the BeiDou observation value and the pose estimation value to obtain the fusion observation value, and updates the positioning result using the fusion observation value.

[0006] As a further optimization of the invention, when the navigation control system is in a fusion positioning state, when the BeiDou observation difference is less than the second threshold, the navigation control system enters the BeiDou positioning state and updates the positioning result based on the BeiDou observation value; when the BeiDou observation difference is not less than the second threshold, the navigation control system maintains the fusion positioning state and updates the positioning result based on the fusion observation value, wherein the second threshold is less than the first threshold.

[0007] As a further optimization of the invention, it also includes: The BeiDou navigation module is used to acquire BeiDou positioning data; The drone positioning module is used to acquire drone image positioning data; The dead reckoning module is used to acquire dead reckoning data.

[0008] As a further optimization of the invention, the drone positioning module includes a drone body and a camera located at the bottom of the drone body. The agricultural machinery has a landing pad on its body, and an icon is placed on the landing pad. The drone body collects image data through the camera to identify the icon on the landing pad and the ridge boundary of the field where the agricultural machinery is located, and obtains the drone observation value based on the relative positional relationship between the icon and the ridge boundary.

[0009] As a further optimization of the invention, the positioning update module acquires UAV observations and BeiDou observations with the same timestamp within a time window, calculates the difference between the UAV observations and BeiDou observations at the same timestamp, and takes the average of all differences calculated within the time window as the BeiDou observation difference.

[0010] As a further optimization of the invention, in the fused positioning state, the positioning update module acquires pose estimation values ​​and BeiDou observation values ​​with the same timestamp within the time window, calculates the difference between the pose estimation values ​​and BeiDou observation values ​​at the same timestamp, and uses the average of all differences calculated within the time window as the pose estimation difference; the fused observation value is determined based on the BeiDou observation difference and the pose estimation difference, and the expression for the fused observation value is as follows: ;

[0011] ;

[0012] in, These are BeiDou observation values. For pose estimation value, These are the weights for the BeiDou observation values ​​and the pose estimation values, respectively. This is the difference in BeiDou observations. The difference is calculated for the pose.

[0013] As a further optimization of the invention, the positioning update module analyzes the validity of the UAV observations based on the following conditions before the difference analysis: ;

[0014] in, The window length is the number of cycles within the window. This represents the number of UAV observations within the current time window. This is the efficiency limit; If the conditions are met, the current cycle is determined to be a valid cycle, and difference analysis and positioning result updates are performed; if the conditions are not met, the current cycle is determined to be an invalid cycle, the positioning status is maintained, and the positioning result is updated according to the positioning result update method of the previous cycle; when the cumulative number of invalid cycles reaches a set value, the navigation control system controls the agricultural machinery to stop.

[0015] As a further optimization of the invention, in the fusion positioning state, when both the BeiDou observation difference and the pose estimation difference are less than the second threshold, the navigation control system controls the agricultural machinery to stop.

[0016] A method for intelligent agricultural machinery navigation and control based on BeiDou high-precision positioning includes the following steps: Step S1: The data acquisition module obtains BeiDou observation values, UAV observation values, and pose estimation values ​​based on BeiDou positioning data, UAV image positioning data, and dead reckoning data. Step S2: The state prediction module performs state prediction based on the pose estimation value to obtain the prior state estimate and the prior state covariance matrix. Step S3: The positioning update module performs difference analysis based on the UAV observations to adjust the current state and observations of the navigation control system, and updates the positioning results based on the prior state covariance matrix and prior state estimation. Repeat steps S2 and S3, alternately performing state prediction and positioning result update. The initial state of the navigation control system is BeiDou positioning state. When the BeiDou observation difference is less than the first threshold, the navigation control system maintains the BeiDou positioning state and updates the positioning result using the BeiDou observation value. When the BeiDou observation difference is not less than the first threshold, the navigation control system enters the fusion positioning state, weightedly fuses the BeiDou observation value and the pose estimation value to obtain the fusion observation value, and updates the positioning result using the fusion observation value.

[0017] A storage medium having computer program instructions stored thereon, which, when executed by a processor, implement the method.

[0018] The beneficial effects of this invention are as follows: This invention uses UAV observations as a benchmark and adjusts the positioning status of the navigation control system based on the difference between BeiDou observations and UAV observations. In the fusion positioning state, based on the difference between BeiDou observations and the difference in pose estimation, the fusion weights of BeiDou observations and pose estimations are dynamically allocated to generate fusion observations to update the positioning results. It can adaptively adjust the degree of trust in the observations of each sensor according to changes in the actual environment, and effectively fuse multi-source data in complex scenarios such as satellite signal obstruction, thereby improving the continuity and accuracy of the positioning results. This invention establishes a hysteresis comparison mechanism by setting a first threshold for triggering the fusion positioning state and a second threshold for exiting the fusion positioning state, with the second threshold being lower than the first threshold. This mechanism ensures that when the BeiDou observation difference fluctuates, the navigation control system only enters the fusion state when the difference exceeds the higher first threshold and only returns to the BeiDou positioning state when it falls below the lower second threshold. This avoids high-frequency jumps in the positioning state caused by minor fluctuations near a single threshold, thereby ensuring the stability and smoothness of system control and effectively preventing vibrations during agricultural machinery operation. In each update cycle under fusion positioning, this invention counts the number of valid UAV observations within a time window. When the ratio of valid data is lower than the validity limit, it is determined to be an invalid cycle, and the positioning update method of the previous cycle is maintained. When the number of invalid cycles accumulates to a set value, the agricultural machinery is controlled to stop. At the same time, under fusion positioning, when both the difference between Beidou observations and the difference between pose estimations are less than a second threshold, the agricultural machinery is controlled to stop. In extreme cases where multiple sensor data show abnormalities or their validity is severely reduced, the agricultural machinery operation can be stopped in time, improving the safety of the system's autonomous operation. Attached Figure Description

[0019] Figure 1 This is a system block diagram of the present invention; Figure 2 This is a schematic diagram of the overall structure of the present invention; Figure 3 This is a flowchart of the present invention; In the diagram: 1. Agricultural machinery; 2. Beidou navigation module; 3. UAV positioning module; 4. Dead reckoning module; 5. Data acquisition module; 6. Status prediction module; 7. Positioning update module. Detailed Implementation

[0020] The present application will now be described in further detail with reference to the accompanying drawings. It should be noted that the following specific embodiments are only used to further illustrate the present application and should not be construed as limiting the scope of protection of the present application. Those skilled in the art can make some non-essential improvements and adjustments to the present application based on the above application content.

[0021] First Embodiment like Figures 1 to 3 As shown, this embodiment relates to an intelligent agricultural machinery navigation control system based on BeiDou high-precision positioning. This system is used for positioning and navigation control of agricultural machinery 1. The navigation control system includes a BeiDou navigation module 2 for acquiring BeiDou positioning data, a UAV positioning module 3 for acquiring UAV image positioning data, a dead reckoning module 4 for acquiring dead reckoning data, and an industrial control computer. The industrial control computer includes a data acquisition module 5 for receiving BeiDou positioning data, UAV image positioning data, and dead reckoning data and preprocessing the aforementioned data. The agricultural machinery 1 has a brake pedal, an accelerator pedal, and a steering lever on its frame, with a steering wheel at the top of the steering lever. The agricultural machinery 1 has angle adjustment components for controlling the brake and accelerator pedals. These angle adjustment components are preferably two electric push rods hinged to the frame of the agricultural machinery 1, with the two electric push rods respectively hinged to the brake and accelerator pedals. The agricultural machinery 1 also has a rotary drive component for controlling the steering lever, preferably a servo motor.

[0022] This embodiment uses a rice transplanter as an example; the agricultural machine 1 is an automatic rice transplanter. The BeiDou navigation module 2 is installed on the automatic rice transplanter and includes a GNSS receiver and two antennas connected to the GNSS receiver. The two antennas are respectively installed on both sides of the automatic rice transplanter frame. The antennas receive satellite signals, and the GNSS receiver calculates the three-dimensional coordinate data of the antenna mounting points, i.e., the BeiDou positioning data. The GNSS receiver sends the BeiDou positioning data and the corresponding timestamp to the data acquisition module 5.

[0023] The UAV positioning module 3 consists of the UAV body and a camera located at the bottom of the UAV body. The agricultural machinery 1 has a landing pad H in the middle of its body, with an icon on the landing pad H. The UAV body collects image data through the camera to identify the icon on the landing pad H and the boundary of the ridges in the field where the agricultural machinery 1 is located. Based on the relative positional relationship between the icon and the ridge boundary, the UAV body obtains the UAV observation value. The UAV body then sends the UAV observation value and the corresponding timestamp to the data acquisition module 5. The timestamp corresponding to the UAV observation value is used to mark the data acquisition time of the corresponding image data.

[0024] The dead reckoning module 4 is preferably a SLAM system. It acquires point cloud data or image data around the agricultural machinery 1 through a visual sensor, builds a map model, estimates the position of the agricultural machinery 1, and obtains dead reckoning data. The SLAM system sends the dead reckoning data and the corresponding timestamp to the data acquisition module 5.

[0025] The data acquisition module 5 acquires positioning data in real time and preprocesses it. Specifically, after receiving BeiDou positioning data, UAV observations, and dead reckoning data in real time, it performs data preprocessing including coordinate system alignment, outlier removal, and linear interpolation. The preprocessed BeiDou positioning data forms BeiDou observation values. The UAV observations are preprocessed to form UAV observation data. Dead reckoning data is preprocessed to generate pose estimation values. Both the dead reckoning module 4 and the data acquisition module 5 are mounted on the frame of the automatic rice transplanter.

[0026] The industrial control computer also includes a state prediction module 6 and a positioning update module 7. The state prediction module 6 is used to calculate values ​​based on the pose. Perform SLAM state prediction and obtain prior state estimates. Furthermore, the prior state covariance matrix is ​​predicted to obtain the prior state covariance matrix. Prior state estimation The prior state covariance matrix is ​​calculated based on the posterior state estimate of the previous cycle and the pose change between the current cycle and the previous cycle. The coordinates are calculated based on the posterior state covariance matrix and process noise covariance matrix of the previous cycle, and will not be elaborated further. Positioning update module 7 is used to update the UAV observations. A difference analysis is performed based on the current state of the navigation control system, the observations corresponding to the current state, and the prior state covariance matrix. and prior state estimation Update the positioning results. Both the status prediction module 6 and the positioning update module 7 are installed on the frame of the automatic rice transplanter.

[0027] The navigation and control system initially operates under BeiDou positioning, based on BeiDou observations. Compared with drone observations The difference is obtained from the BeiDou observation difference. When the BeiDou observation difference When less than the first threshold The navigation and control system is in BeiDou positioning mode, based on BeiDou observation values. Update positioning results; when BeiDou observation difference Not less than the first threshold At that time, the navigation and control system enters the fusion positioning state, integrating BeiDou observation values. pose estimation value , to obtain fused observation values Based on fused observations Update the positioning results. The industrial control computer calculates the vehicle speed based on the changes in the positioning results, and controls the brake pedal and accelerator pedal through the angle adjustment component, and adjusts the steering lever through the rotation drive component, thereby regulating the driving status of agricultural machinery 1.

[0028] State prediction module 6 and positioning update module 7 form a Kalman filter. The Kalman filter alternately performs state prediction and positioning update to achieve multi-cycle positioning result updates. When the navigation control system enters the fused positioning state and updates the positioning result, it performs the next cycle of state prediction and positioning result update. During the next cycle of positioning result update, when the BeiDou observation difference... Less than the second threshold At that time, the navigation and control system returned to BeiDou positioning status, based on BeiDou observation values. Update positioning results; when BeiDou observation difference Not less than the second threshold At the same time, the navigation and control system maintains a fused positioning state, integrating BeiDou observation values. pose estimation value , to obtain fused observation values Based on fused observations Update the location results. Second threshold. Less than the first threshold .

[0029] The specific methods for obtaining the BeiDou observation difference are as follows: Within a fixed-length time window, acquire UAV observations with the same timestamp within that time window. and BeiDou observation values Calculate UAV observations at the same timestamp With BeiDou observation values The difference The difference calculated within this time window The average value is used as the difference in BeiDou observations. The navigation and control system acquires positioning data in real time.

[0030] During the fusion localization process, within the aforementioned time window, pose estimation values ​​with the same timestamp are acquired. and BeiDou observation values Calculate the pose estimation value under the same timestamp. With BeiDou observation values The difference The difference calculated within this time window The average value is used as the pose estimation difference. Based on BeiDou observation difference Difference between pose estimation and position estimation Set weights to calculate the fused observations using a weighted fusion method. , merge observations The calculation formula is as follows: ;

[0031] ;

[0032] in, These represent the weights for BeiDou observations and pose estimations, respectively. In the above data preprocessing, linear interpolation is used to ensure that UAV observations at different timestamps are consistent. Each has a corresponding BeiDou observation value. and pose estimation value .

[0033] In addition, before the positioning update, the positioning update module 7 analyzes whether the UAV observations are valid based on the following conditions: ;

[0034] in, The window length is the number of cycles within the window. This represents the number of UAV observations within the current time window. For efficiency, As an efficiency limit, in this embodiment , .

[0035] When the above conditions are met, the current period is determined to be a valid period, difference analysis is performed, and the positioning results are updated. When the above conditions are not met, the current cycle is determined to be an invalid cycle. The positioning status of the navigation control system is maintained, and the positioning update method of the previous cycle is used to update the positioning result. That is, if BeiDou positioning was used in the previous cycle, BeiDou positioning will still be used in the current cycle, and the positioning result will be updated based on BeiDou observations; if fusion positioning was used in the previous cycle, fusion positioning will still be used in the current cycle, fusing BeiDou observations and pose estimation values ​​to update the positioning result. Entering the next cycle, state prediction and positioning updates are repeated. When invalid cycles persist and the cumulative number of invalid cycles reaches a set value, the navigation control system controls agricultural machinery 1 to stop. Furthermore, in fusion positioning mode, when the BeiDou observation difference... Difference between pose estimation and position estimation All are less than the second threshold At the same time, the navigation control system also controls the agricultural machinery 1 to stop.

[0036] The location results are updated based on the following formula: ① Under BeiDou positioning status: The Kalman gain matrix under BeiDou positioning conditions is calculated based on the following formula: ;

[0037] ;

[0038] in, The Kalman gain matrix for the current period. Let be the prior state covariance matrix. The matrix from which the position components are extracted is the known observation matrix. , This is the transpose of the observation matrix. To observe the noise covariance matrix, This is the fixed noise matrix of the GNSS receiver. Based on the Kalman gain matrix, the posterior state estimate and posterior state covariance matrix are calculated using the following formulas: ;

[0039] ;

[0040] in, For prior state estimation, For posterior state estimation, Let be the posterior state covariance matrix. Let be the identity matrix. The above posterior state estimation... This is the location result updated in the current cycle.

[0041] ②In the fused positioning state: The Kalman gain matrix under BeiDou positioning conditions is calculated based on the following formula: ;

[0042] ;

[0043] in, This is the observation noise covariance matrix for SLAM pose estimation. Based on the Kalman gain matrix above, the posterior state estimate and posterior state covariance matrix are calculated using the following formulas: ;

[0044] ;

[0045] Similarly, this posterior state estimate This is the location result updated in the current cycle.

[0046] Second Embodiment This embodiment relates to an intelligent agricultural machinery navigation and control method based on BeiDou high-precision positioning. This method is applicable to the intelligent agricultural machinery navigation and control system of the previous embodiment and includes the following steps: Step S1: The data acquisition module obtains BeiDou observation values ​​based on BeiDou positioning data, UAV image positioning data, and dead reckoning data. UAV observations pose estimation value ; Step S2: The state prediction module calculates the value based on the pose. Perform state prediction to obtain prior state estimates. and prior state covariance matrix ; Step S3: The positioning update module performs difference analysis based on the UAV observations to adjust the current state and observations of the navigation control system, and based on the prior state covariance matrix. and prior state estimation Update location results; Repeat steps S2 and S3, alternately performing state prediction and positioning result updates. The initial state of the navigation control system is the BeiDou positioning state, based on BeiDou observations. Compared with drone observations The difference is obtained from the BeiDou observation difference. When the BeiDou observation difference When less than the first threshold The navigation and control system is in BeiDou positioning mode, based on BeiDou observation values. Update positioning results; when BeiDou observation difference Not less than the first threshold At that time, the navigation and control system enters the fusion positioning state, integrating BeiDou observation values. pose estimation value , to obtain fused observation values Based on fused observations Update location results.

[0047] After the navigation and control system enters the fusion positioning state and updates the positioning results, it performs the state prediction and positioning result update for the next cycle. During the positioning result update process of the next cycle, when the BeiDou observation difference... Less than the second threshold At that time, the navigation and control system returned to BeiDou positioning status, based on BeiDou observation values. Update positioning results; when BeiDou observation difference Not less than the second threshold At the same time, the navigation and control system maintains a fused positioning state, integrating BeiDou observation values. pose estimation value , to obtain fused observation values Based on fused observations Update location results.

[0048] Third Embodiment This embodiment relates to a storage medium storing computer program instructions, which, when executed by a processor, implement the intelligent agricultural machinery navigation control method of the second embodiment.

[0049] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. An intelligent agricultural machinery navigation and control system based on BeiDou high-precision positioning, characterized in that, include: The data acquisition module is used to obtain BeiDou observation values, UAV observation values, and pose estimation values ​​based on BeiDou positioning data, UAV image positioning data, and dead reckoning data. The state prediction module is used to predict the state based on the pose estimation value in order to obtain the prior state estimate and the prior state covariance matrix. The positioning update module is used to perform difference analysis based on UAV observations to adjust the current state and observations of the navigation control system, and to update the positioning results based on the prior state covariance matrix and prior state estimation. The state prediction module and the positioning update module alternately perform state prediction and positioning result update. The initial state of the navigation control system is BeiDou positioning state. When the BeiDou observation difference is less than the first threshold, the navigation control system maintains the BeiDou positioning state and updates the positioning result using the BeiDou observation value. When the BeiDou observation difference is not less than the first threshold, the navigation control system enters the fusion positioning state, weightedly fuses the BeiDou observation value and the pose estimation value to obtain the fusion observation value, and updates the positioning result using the fusion observation value.

2. The intelligent agricultural machinery navigation and control system based on BeiDou high-precision positioning according to claim 1, characterized in that: When the navigation control system is in fusion positioning state, when the difference between BeiDou observations is less than the second threshold, the navigation control system enters BeiDou positioning state and updates the positioning result based on the BeiDou observations; when the difference between BeiDou observations is not less than the second threshold, the navigation control system maintains fusion positioning state and updates the positioning result based on the fusion observations, where the second threshold is less than the first threshold.

3. The intelligent agricultural machinery navigation and control system based on BeiDou high-precision positioning according to claim 1, characterized in that, Also includes: The BeiDou navigation module is used to acquire BeiDou positioning data; The drone positioning module is used to acquire drone image positioning data; The dead reckoning module is used to acquire dead reckoning data.

4. The intelligent agricultural machinery navigation and control system based on BeiDou high-precision positioning according to claim 3, characterized in that: The drone positioning module includes the drone body and a camera located at the bottom of the drone body. The agricultural machinery has a landing pad on its body, and there is an icon on the landing pad. The drone body collects image data through the camera to identify the icon on the landing pad and the boundary of the ridge in the field where the agricultural machinery is located. Based on the relative positional relationship between the icon and the ridge boundary, the drone observation value is obtained.

5. The intelligent agricultural machinery navigation and control system based on BeiDou high-precision positioning according to claim 1, characterized in that: Within a time window, the positioning update module acquires UAV observations and BeiDou observations with the same timestamp, calculates the difference between the UAV observations and BeiDou observations at the same timestamp, and uses the average of all differences calculated within the time window as the BeiDou observation difference.

6. The intelligent agricultural machinery navigation and control system based on BeiDou high-precision positioning according to claim 5, characterized in that: In the fusion positioning state, the positioning update module obtains the pose estimation value and Beidou observation value with the same timestamp within the time window, calculates the difference between the pose estimation value and the Beidou observation value with the same timestamp, and takes the average of all the differences calculated within the time window as the pose estimation difference. The fused observation values ​​are determined based on the differences between BeiDou observations and pose estimations. The expression is as follows: in, These are BeiDou observation values. For pose estimation value, These are the weights for the BeiDou observation values ​​and the pose estimation values, respectively. This is the difference in BeiDou observations. The difference is calculated for the pose.

7. The intelligent agricultural machinery navigation and control system based on BeiDou high-precision positioning according to claim 6, characterized in that: Before the difference analysis, the positioning update module analyzes whether the UAV observations are valid based on the following conditions: in, The window length is the number of cycles within the window. This represents the number of UAV observations within the current time window. This is the efficiency limit; If the conditions are met, the current cycle is determined to be a valid cycle, and difference analysis and positioning result updates are performed; if the conditions are not met, the current cycle is determined to be an invalid cycle, the positioning status is maintained, and the positioning result is updated according to the positioning result update method of the previous cycle; when the cumulative number of invalid cycles reaches a set value, the navigation control system controls the agricultural machinery to stop.

8. The intelligent agricultural machinery navigation and control system based on BeiDou high-precision positioning according to claim 1, characterized in that: In the fusion positioning state, when both the BeiDou observation difference and the pose estimation difference are less than the second threshold, the navigation control system controls the agricultural machinery to stop.

9. A method for intelligent agricultural machinery navigation and control based on BeiDou high-precision positioning, comprising the intelligent agricultural machinery navigation and control system according to any one of claims 1-8, including the following steps: Step S1: The data acquisition module obtains BeiDou observation values, UAV observation values, and pose estimation values ​​based on BeiDou positioning data, UAV image positioning data, and dead reckoning data. Step S2: The state prediction module performs state prediction based on the pose estimation value to obtain the prior state estimate and the prior state covariance matrix. Step S3: The positioning update module performs difference analysis based on the UAV observations to adjust the current state and observations of the navigation control system, and updates the positioning results based on the prior state covariance matrix and prior state estimation. Repeat steps S2 and S3, alternately performing state prediction and positioning result update. The initial state of the navigation control system is BeiDou positioning state. When the BeiDou observation difference is less than the first threshold, the navigation control system maintains the BeiDou positioning state and updates the positioning result using the BeiDou observation value. When the BeiDou observation difference is not less than the first threshold, the navigation control system enters the fusion positioning state, weightedly fuses the BeiDou observation value and the pose estimation value to obtain the fusion observation value, and updates the positioning result using the fusion observation value.

10. A storage medium storing computer program instructions thereon, characterized in that, The method of claim 9 is implemented when the computer program instructions are executed by the processor.

Images