Combine harvester and automatic driving operation path planning control method and device thereof

Abstract

A combine harvester and an automatic driving operation path planning control method and device thereof, the combine harvester comprising an automatic driving operation path planning control device, and adopting the following method for harvesting operation: a local positioning area covered by signals is built, and field position information of the combine harvester is acquired in real time in the area; based on the field position information, taking the first row harvested by the combine harvester as a reference row, a global operation path of the combine harvester based on a current row is generated based on the reference row and an operation area; after entering the row operation, the global operation path that has been planned is corrected according to an actual path of the planting row, and actual path information of the current row is recorded; after completing the operation of the current row, the path of the current row and the global operation path that has been planned are compared, whether the error is greater than a set value is judged, if yes, taking the current actual path as the reference row, a subsequent global operation path is regenerated, and the accumulated error of the parallel path junction row is eliminated; if not, the operation continues according to the global operation path that has been planned.

Description

Technical Field

[0001] This invention relates to agricultural machinery automation technology, and in particular to a combine harvester and its automatic driving operation path planning and control method and device based on uniform cotton planting row spacing. Background Technology

[0002] With the acceleration of agricultural modernization, the demand for automation and intelligence in agricultural machinery is becoming increasingly urgent. Precision operation, efficient collaboration, and low loss rate have become core evaluation indicators for modern agricultural machinery. Especially in the combined harvesting stage, the application of autonomous driving technology is directly related to harvest loss reduction and operational efficiency. However, different crop types, operating environments, and technical conditions pose differentiated challenges to autonomous driving devices, and existing technologies still have significant limitations. In the tillage and sowing stages, where the ground surface is unobstructed, GPS / BeiDou-based path planning technology is mature and can achieve straight-line tracking with an error of ≤2 cm, improving operational efficiency by more than 30%. In the plant protection and harvesting stages, rice / wheat does not require precise row alignment, but relying on high-precision RTK-GPS is costly and cannot be applied in large quantities. Corn / cotton harvesting requires strict alignment with crop rows, with a lateral deviation of ≤5 cm. Existing devices have problems such as insufficient inter-row recognition technology (e.g., weak resistance to light interference of visual sensors) and redundant turning trajectories (excessive turning radius damages unharvested crops).

[0003] Existing autonomous driving technologies suffer from information silos and data gaps. Crop row information during the sowing process is not standardized or shared, preventing harvesters from directly accessing sowing trajectories. Furthermore, the lack of deep integration between agricultural geographic information systems (GIS) and agricultural machinery control devices results in insufficient ability to avoid dynamic obstacles (such as temporary stockpiles and ditches). The dependence on and robustness of positioning signals are also insufficient; the lateral error of a single GPS navigation system can exceed 15 cm, failing to meet the high-precision row-based operation requirements of crops like corn and cotton. While current autonomous driving technology for combine harvesters has made progress in some scenarios, it is still limited by issues such as information fragmentation, environmental adaptability, and potential breakpoints throughout the entire process, preventing the formation of a universally applicable solution. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to address the above-mentioned deficiencies of the prior art by providing a combine harvester and its autonomous driving operation path planning and control method and device based on the integration of parallel path planning and touch-to-row technology, so as to solve the problems of high dependence on external information, low operation accuracy and limited application of the prior art, and realize precise autonomous driving operation of the combine harvester.

[0005] To achieve the above objectives, the present invention provides an autonomous driving operation path planning and control method, comprising the following steps:

[0006] The step of constructing a local positioning network involves building a local positioning area with signal coverage for the plot to be operated, and obtaining the field location information of the harvester in real time through the regional positioning unit within the local positioning area;

[0007] The steps for generating a baseline row and a global operation path are as follows: Based on the field location information, the first row harvested by the harvester is taken as the baseline row, and the path planning unit generates a global operation path for the harvester based on the current row based on the baseline row and the harvester's operating area.

[0008] The touch-based dynamic row alignment and path correction steps involve the automatic row alignment unit correcting the planned global operation path based on the actual path of the planting row after the harvester enters the row for operation, while integrating the actual path information of the current row recorded by the control unit.

[0009] The path iteration optimization and error elimination steps are as follows: After the harvester completes the current row operation, it compares the current row path with the planned global operation path to determine whether the error is greater than a set value. If so, the path planning unit regenerates the subsequent global operation path based on the current actual path as the baseline row to eliminate the accumulated error at the intersection of parallel paths; otherwise, it continues to work according to the planned global operation path.

[0010] In the above-mentioned autonomous driving operation path planning and control method, in the step of constructing a local positioning network, a drone carrying a signal transmitter is placed at the four corners and / or the surrounding area of ​​the plot to build a local positioning area with signal coverage. A signal receiver adapted to the signal transmitter is set on the harvester. The area positioning unit is connected to the signal receiver to obtain the field location information of the harvester in real time and construct a field operation map.

[0011] In the above-mentioned autonomous driving operation path planning and control method, in the step of generating the reference row and the global operation path, the first row of harvesting by the manually driven harvester is used as the reference row, the trajectory coordinates are recorded in real time, and the reference path is generated; the number of harvesting rows on the left and right sides of the reference row is input respectively to obtain the harvester's operation area and area size, and based on the reference row, the parallel path is expanded according to the standard row spacing of the ridge, and the turning path at the end of the field is planned using a progressive fishtail line turning algorithm.

[0012] In the aforementioned autonomous driving operation path planning and control method, the automatic alignment unit is touch-based in the touch-based dynamic alignment and path correction steps. Based on real-time touch information, a PID control algorithm is used to perform two-layer path correction in conjunction with autonomous path replanning, so as to enhance the robustness to adapt to complex environments.

[0013] The aforementioned autonomous driving operation path planning and control method also includes steps for real-time display and data storage of the operation path.

[0014] The aforementioned autonomous driving operation path planning and control method also includes a full-cycle adaptive loop step. After each reciprocating operation loop is completed, a positioning network health check is automatically performed, and an early warning is triggered when the signal strength and / or the drone's endurance status are abnormal.

[0015] To better achieve the above objectives, the present invention also provides an autonomous driving operation path planning and control device, wherein the autonomous driving operation path planning and control method described above includes:

[0016] The signal acquisition module includes a signal transmitter, a signal receiver, a touch sensor, a rear axle angle sensor, and a walking speed sensor. The signal transmitter is mounted on the drone, and the signal receiver is mounted on the harvester and adapted to the signal transmitter. The touch sensor, rear axle angle sensor, and walking speed sensor are respectively mounted on the harvester.

[0017] The planning and control module, connected to the harvester's walking device, includes a regional positioning unit, a path planning unit, an automatic alignment unit, and a fusion control unit. The regional positioning unit is connected to the signal receiver. The touch sensor, rear axle angle sensor, and walking speed sensor are respectively connected to the automatic alignment unit. The path planning unit is connected to the regional positioning unit, the automatic alignment unit, and the fusion control unit.

[0018] The aforementioned autonomous driving operation path planning and control device further includes: a display module, installed in the cab of the harvester and connected to the planning and control module, for real-time display and data storage of the operation path.

[0019] The aforementioned autonomous driving operation path planning and control device includes an automatic alignment unit comprising multiple touch-type signal acquisition components evenly distributed at the front end of the harvester. Each touch-type signal acquisition component comprises a first link, a second link, and a third link connected in sequence. The first and third links are symmetrically arranged, and the second link is connected to the front end of the harvester. Touch-type guide plates are symmetrically arranged at the ends of the first and third links, and touch sensors are respectively installed on the first and third links.

[0020] To better achieve the above objectives, the present invention also provides a combine harvester, which includes the above-described automatic driving operation path planning and control device and uses the above-described automatic driving operation path planning and control method to perform harvesting operations.

[0021] The technical advantages of this invention are as follows:

[0022] This invention uses regional positioning units to obtain the location information of cotton harvesters in the field, without relying on GPS / BeiDou sowing information and the strength of GPS / BeiDou positioning signals. It has a wide range of applications and solves the problems of information silos, weak signals, and insufficient coverage. It adopts a control method that integrates baseline row path planning and touch-based automatic row alignment to achieve precise dynamic correction of the operation path and improve operation accuracy. Through the fusion control of baseline row path planning and real-time perception of row information, as well as iterative optimization of the dynamic operation path, it realizes precise automatic driving operation of combine harvesters. It solves the problem that automatic driving technology relies on GPS / BeiDou sowing information and the strength and range of positioning signals, meets the requirements of automatic driving operation, and can be applied to cotton and corn harvesting equipment and other agricultural machinery.

[0023] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments, but this is not intended to limit the present invention. Attached Figure Description

[0024] Figure 1 This is a schematic diagram illustrating the working principle of an embodiment of the present invention;

[0025] Figure 2 This is a schematic diagram of a control method according to an embodiment of the present invention;

[0026] Figure 3 This is a schematic diagram illustrating the principle of dynamic correction of autonomous driving operation path according to an embodiment of the present invention;

[0027] Figure 4 This is a schematic diagram of the progressive fishtail turning algorithm path according to an embodiment of the present invention;

[0028] Figure 5 This is a structural block diagram of an autonomous driving operation path planning and control device according to an embodiment of the present invention;

[0029] Figure 6 This is a schematic diagram of the working principle of the automatic row alignment unit according to an embodiment of the present invention;

[0030] Figure 7 This is a schematic diagram of a touch-type signal acquisition device according to an embodiment of the present invention;

[0031] Figure 8 This is a schematic diagram illustrating the working principle of a touch-type signal acquisition device according to an embodiment of the present invention.

[0032] Among them, the attached figures are labeled

[0033] 1. Automated driving operation path planning and control device

[0034] 11 Signal Acquisition Module

[0035] 111 signal transmitter

[0036] 112 signal receiver

[0037] 113 Touch Sensor

[0038] 114 Rear Axle Angle Sensor

[0039] 115 Walking speed sensor

[0040] 12 Planning and Control Module

[0041] 121 Area Positioning Unit

[0042] 122 Path Planning Unit

[0043] 123 Automatic Alignment Unit

[0044] 1230 Touch-type signal acquisition device

[0045] 1231 Touch-sensitive guide plate

[0046] 1232 First Link

[0047] 1233 Second Link

[0048] 1234 Third Link

[0049] 124 Fusion Control Unit

[0050] 13 Display Modules

[0051] 2. Walking device

[0052] 3. Local positioning area

[0053] 4. Drones

[0054] 5 Plots of land awaiting construction Detailed Implementation

[0055] The structural and working principles of the present invention will be described in detail below with reference to the accompanying drawings:

[0056] The combine harvester of the present invention includes a walking device 2 and an automatic driving operation path planning and control device 1, and employs an automatic driving operation path planning and control method for harvesting operations. This combine harvester can be, for example, a cotton harvester or a corn harvester. The composition, structure, relative positions, connections, and functions of other parts of the combine harvester are all mature existing technologies, and therefore will not be described in detail here. The following will only provide a detailed description of the automatic driving operation path planning and control method and device of the present invention.

[0057] See Figures 1-4 , Figure 1 This is a schematic diagram illustrating the working principle of an embodiment of the present invention. Figure 2 This is a schematic diagram of a control method according to an embodiment of the present invention. Figure 3 This is a schematic diagram illustrating the principle of dynamic correction of autonomous driving operation path according to an embodiment of the present invention. Figure 4 This is a schematic diagram of a progressive fishtail turn algorithm according to an embodiment of the present invention. The autonomous driving operation path planning and control method of the present invention includes the following steps:

[0058] The step of constructing a local positioning network involves building a local positioning area 3 with signal coverage for the plot 5 to be operated on, and obtaining the field location information of the harvester in real time through the regional positioning unit 121 within the local positioning area 3.

[0059] The steps for generating a baseline row and a global operation path are as follows: Based on the field location information, the first row harvested by the harvester is taken as the baseline row, and the path planning unit 122 generates a global operation path for the harvester based on the current row based on the baseline row and the harvester's operating area.

[0060] The touch-based dynamic row alignment and path correction steps involve the automatic row alignment unit 123 correcting the planned global operation path based on the actual path of the planting row after the harvester enters the row for operation, while the control unit 124 records the actual path information of the current row.

[0061] The path iteration optimization and error elimination steps involve the harvester comparing the current row path with the planned global operation path after completing the current row operation. If the error exceeds a set value, the path planning unit 122 regenerates the subsequent global operation path based on the current actual path as the baseline to eliminate accumulated errors at parallel path junctions. If not, the harvester continues to operate according to the planned global operation path.

[0062] Repeat the steps of generating baseline rows and global job paths, followed by path iteration optimization and error elimination, until all jobs for plot 5 to be done are completed.

[0063] In the step of constructing a local positioning network in this embodiment, a drone 4 carrying a signal transmitter 111 is placed at the four corners and / or surrounding areas of the plot to establish a local positioning area 3 with signal coverage. A signal receiver 112 adapted to the signal transmitter 111 is set on the harvester. The area positioning unit 121 is connected to the signal receiver 112 to obtain the field position information of the harvester in real time and construct a field operation map. In the step of generating a baseline row and a global operation path, the first row harvested by the manually driven harvester is used as the baseline row. The trajectory coordinates are recorded in real time to generate a baseline path. The number of harvesting rows on the left and right sides of the baseline row are input to obtain the harvester's operation area and area size. Based on the baseline row, a parallel path is extended according to the standard row spacing of the ridges. A progressive fishtail turning algorithm is used to plan the turning path at the edge of the field. In the step of touch-based dynamic row alignment and path correction, the signal acquisition component of the automatic row alignment unit 123 is touch-based. Based on the real-time touch information, a PID control algorithm is used, combined with autonomous path replanning, to perform dual path correction to enhance the robustness to adapt to complex environments.

[0064] This embodiment may also include real-time display and data storage steps for the work path. It may also include a full-cycle adaptive loop step, automatically performing a positioning network health check after each reciprocating work loop is completed, and triggering an alert when signal strength and / or UAV 4 battery life are abnormal.

[0065] In this embodiment, the autonomous driving operation path planning and control method adopts a dual control architecture that integrates baseline path planning and touch-based automatic row alignment control, enabling precise dynamic correction of the operation path and enhancing robustness in adapting to complex environments. The following example illustrates the workflow of this control method during cotton harvester operation.

[0066] See Figure 2In this embodiment, firstly, a drone 4 equipped with a signal transmitter 111 is placed at the four corners of the cotton field. A signal receiver 112 is installed on the cotton harvester, and the area positioning unit 121 is activated to obtain the location information of the cotton harvester in the signal coverage area, preparing for path planning. Then, the first row harvested by the manually driven or automatically driven cotton harvester is used as the reference row. The number of harvesting rows on the left and right sides of the reference row is input into the device to determine the working area and size of the cotton harvester. The path planning unit 122 generates the working path of the cotton harvester based on the reference row. At the same time, the display module 13 displays the working path of the cotton harvester and stores the collected data. After the cotton harvester enters the row for operation, the touch-type automatic row alignment unit 1... 23 By collecting real-time data from the yaw detection touch sensor 113 and the rear axle angle sensor 114, the automatic row-alignment unit 123 dynamically adjusts the cotton harvester's direction of travel using a PID algorithm based on feedback information. It corrects the planned path according to the actual path of the cotton planting rows, while the fusion control unit 124 records the actual path of the current row. After the cotton harvester completes the current row's work, the fusion control unit 124 compares the current path with the planned path. If the comparison error is large, exceeding the error setting value, the path planning unit 122 regenerates the work path based on the corrected current row as the baseline; otherwise, it continues to work according to the planned path. After each row is completed, the Hausdorff distance between the actual path and the planned path is compared, with a threshold set at 3-5 cm; therefore, the preferred error setting value is 3-5 cm.

[0067] Each picking device of the cotton harvester is equipped with a touch-sensitive signal acquisition unit 1230, see [link / reference]. Figure 6 Touch-type guide plates 1231 are installed on the left and right sides. One end of the touch-type guide plate 1231 is connected to the touch sensor 113. When the other end of the touch-type guide plate 1231 contacts the cotton stalk and a certain displacement occurs, and the angle change value of the touch sensor 113 is greater than the set threshold, the path correction step is automatically performed on the row unit 123.

[0068] The path correction in this embodiment can be found in [reference]. Figure 3 After the automatic alignment unit 123 is executed, the angle value of the touch sensor 113, the rotation angle value of the rear wheel angle sensor and the current cotton harvester speed of the speed sensor are obtained in real time. The device determines whether the heading angle is greater than the threshold. If so, the steering wheel angle is adjusted through the PID control algorithm to correct the forward direction of the cotton harvester. When the heading angle is less than the threshold, the steering wheel returns to the center and the cotton harvester continues to work.

[0069] In one embodiment of the present invention, the autonomous driving operation path planning and control method is as follows:

[0070] Step S1: Construct a local localization network:

[0071] Deploy signal transmitter 111, and place RTK differential signal transmitter 111 on the four corners of the cotton field via UAV 4 to improve the efficiency of building long-distance positioning signal coverage area and transmit RTK differential signal to cover the working area (maximum support 500×500m²); install signal receiver 112 on cotton harvester to build virtual electronic map by transmitting and receiving signals, and obtain the real-time position of cotton harvester;

[0072] Signal synchronization is achieved by using a time synchronization protocol (PTP) to ensure that the UAV's four clock errors are less than 1μs, thus avoiding signal conflicts from multiple base stations.

[0073] Step S2: Generate the baseline row and global job path:

[0074] During the manual driving phase, the cotton harvester harvests along the first row at a constant speed. The regional positioning unit 121 records the trajectory points at a set frequency and fits them into a cubic B-spline curve as the reference path.

[0075] Global path expansion generates multiple parallel paths based on the baseline row, with the spacing between adjacent paths equal to the average distance between cotton rows multiplied by the number of working rows.

[0076] This further includes:

[0077] The definition and input steps for the baseline row are as follows: Manual teaching can be used, with the operator manually driving the cotton harvester along the first row of the cotton field. The trajectory of the first row is recorded in real time by the regional positioning unit 121, generating baseline row data including latitude and longitude sequences, heading angles, and speed curves. The number of working rows is set by inputting the number of rows to be harvested on the left and right sides of the baseline row (e.g., 20 rows on the left and 20 rows on the right) on the interactive interface. The total working width is calculated based on the average cotton row spacing (which can be manually entered or automatically measured). Area calculation is performed by automatically generating a rectangular working area based on the baseline row length and the total working width, and calculating the area using the following formula: Area = Baseline row length × (Number of left rows + Number of right rows) × Cotton row spacing. Exporting the data to a GeoJSON format farmland map is also supported.

[0078] Dynamic boundary protection steps: Embed electronic fence function in global path planning. If the cotton harvester approaches the boundary of the working area (distance ≤1m) due to accumulated error, an audible and visual alarm is triggered and a reverse turning path is automatically generated to prevent it from crossing the boundary and crushing the unworked area.

[0079] Step S3: Touch-based dynamic row and path correction. Multiple touch sensors 113 are evenly distributed at the front end of the cotton harvester to detect the cotton row position in real time; a rear axle angle sensor 114 is installed on the rear axle to monitor the rear axle angle; a PID algorithm is used to dynamically adjust the cotton harvester's operating path based on the information fed back from the touch sensors 113 and the rear axle angle sensor 114; further including:

[0080] The sensitivity grading process for the contact guide plate is divided into light contact and heavy contact. Light contact, i.e., when the signal of the rear axle angle sensor 114 is 1-10°, triggers lateral fine adjustment (±2-5cm) to maintain continuous operation; heavy contact, i.e. when the signal of the rear axle angle sensor 114 is >10°, triggers emergency pause, which is resumed after manual confirmation.

[0081] In the data fusion step, the fusion control unit 124 transmits the current path corrected by the touch-type automatic alignment unit 123 to the path planning unit 122 to generate a new work path. The guide plate contact signal and GNSS positioning data are weighted and fused, with a preferred weight ratio of 6:4 to prioritize alignment accuracy. That is, real-time pose estimation is output through feedback from the touch-type guide plate 1231, with real-time detection signals as the priority control signals.

[0082] Step S4, Path Iteration Optimization and Error Elimination:

[0083] Error assessment involves calculating the Fréchet distance (measuring curve similarity) between the actual path and the planned path, with a threshold set at 0.05m.

[0084] If the error exceeds the limit, the current actual path is used as the new baseline, and subsequent global paths are regenerated. The sliding window algorithm is used to limit the propagation of historical errors and improve the fit between the planned path and the actual cotton row.

[0085] Step S5, Full-cycle adaptive loop:

[0086] After each reciprocating operation loop is completed, the device automatically performs a health check on the positioning network (such as signal strength, drone battery life, etc.), and triggers an early warning if any abnormality is detected.

[0087] This embodiment also allows manual correction of the baseline row during operation, after which the device automatically updates the global path.

[0088] See Figure 5 , Figure 5 This is a structural block diagram of an autonomous driving operation path planning and control device 1 according to an embodiment of the present invention. The autonomous driving operation path planning and control device 1 of the present invention is used to implement the above-described autonomous driving operation path planning and control method, including:

[0089] The signal acquisition module 11 includes a signal transmitter 111, a signal receiver 112, a touch sensor 113, a rear axle angle sensor 114, and a travel speed sensor 115. The signal transmitter 111 is mounted on the UAV 4, and the signal receiver 112 is mounted on the harvester and adapted to the signal transmitter 111. The touch sensor 113, rear axle angle sensor 114, and travel speed sensor 115 are respectively mounted on the harvester. The touch sensor 113 is used to measure the harvester's yaw, the travel speed sensor 115 is used to measure the harvester's operating speed, and the rear axle angle sensor 114 is used for path correction.

[0090] The planning and control module 12, connected to the walking device 2 of the harvester, includes a region positioning unit 121, a path planning unit 122, an automatic alignment unit 123, and a fusion control unit 124. The region positioning unit is connected to the signal receiver 112. The touch sensor 113, the rear axle angle sensor 114, and the walking speed sensor 115 are respectively connected to the automatic alignment unit 123. The path planning unit 122 is connected to the region positioning unit 121, the automatic alignment unit 123, and the fusion control unit 124.

[0091] This embodiment also includes a display module 13, installed in the cab of the harvester and connected to the planning and control module 12, for real-time display and data storage of the work path, facilitating reference and analysis for subsequent operations. The work paths and related data information of the path planning unit 122, the automatic alignment unit 123, and the fusion control unit 124 are all displayed in real time on the display module 13.

[0092] See Figures 6-8 , Figure 6 This is a schematic diagram illustrating the working principle of the automatic row alignment unit 123 according to an embodiment of the present invention. Figure 7 This is a schematic diagram of the structure of a touch-type signal acquisition device 1230 according to an embodiment of the present invention. Figure 8This is a schematic diagram illustrating the working principle of a touch-type signal acquisition device 1230 according to an embodiment of the present invention. The automatic alignment unit 123 of this embodiment includes multiple touch-type signal acquisition devices 1230, evenly distributed at the front end of the harvester. Each touch-type signal acquisition device 1230 includes a first connecting rod 1232, a second connecting rod 1233, and a third connecting rod 1234 connected in sequence. The first connecting rod 1232 and the third connecting rod 1234 are symmetrically arranged, and the second connecting rod 1233 is connected to the front end of the harvester. Touch-type guide plates 1231 are symmetrically arranged at the ends of the first connecting rod 1232 and the third connecting rod 1234. Touch sensors 113 are respectively installed on the first connecting rod 1232 and the third connecting rod 1234. "Touch-type" refers to the fact that during operation, the touch-type guide plate 1231 swings left and right when it contacts the cotton stalk or other crops, causing the angle sensor angle to change via the connecting rod. When the touch-sensitive guide plate 1231 shifts due to changes in the cotton pattern, for example, when the right touch-sensitive guide plate shifts to the right, the first link 1232 rotates around hinge point A, which in turn drives the third link 1234 to rotate around hinge point B via the second link 1233, causing the left touch-sensitive guide plate to shift to the left. The same logic applies when the left touch-sensitive guide plate shifts to the left. The touch sensor 113 is an angle sensor installed at the hinge point. When the first link 1232 and / or the third link 1234 rotates, the angle sensor signal changes.

[0093] One or more drones 4 carrying area signal transmitters 111 are placed in multiple locations to construct a signal coverage area. The area signal transmitters 111 are mounted on the drones 4 and parked at multiple locations around the cotton field. Signal receivers 112 are installed on the cab of the cotton harvester to receive signals, forming an area positioning unit 121. In this embodiment, four signal transmitters 111 are respectively mounted on four drones 4, controlling the drones 4 to fly to four locations around the cotton field, forming an area that includes the cotton field. The signal receivers 112 are installed on the cab of the cotton harvester, constructing a virtual electronic map within the area formed by the four signal transmitters, allowing real-time acquisition of the cotton harvester's location. The touch-sensitive signal acquisition unit 1230 in the automatic alignment unit 123 is installed on the harvester head of the cotton harvester, a touch sensor 113 is installed on the touch-sensitive signal acquisition unit 1230, and a rear axle angle sensor 114 and a walking speed sensor 115 are installed on the rear axle of the cotton harvester. The path planning unit 122, the fusion control unit 124, and the display module 13 are installed on the right side of the cab and communicate with each other via CAN.

[0094] In this embodiment, the signal transmitter 111 is preferably an RTK differential signal transmitter 111, mounted on the UAV 4. The signal receiver 112 is preferably a dual-frequency GNSS receiver, compatible with GPS / BeiDou / GLONASS, and is mounted on the cotton harvester. The UAV 4 places the RTK differential signal transmitter 111 at the four corners of the cotton field and constructs a local high-precision positioning network through carrier phase real-time dynamic positioning (RTK) to track the real-time position of the cotton harvester. The path planning unit 122 completes the generation of the reference row and the global path. When the cotton harvester harvests along the first row, it records the trajectory coordinates in real time, generating a reference path including latitude and longitude, heading angle, speed curve, etc. Based on the reference row, the parallel path is extended according to the standard row spacing of the cotton ridges, and a progressive fishtail turning algorithm is used to plan the turning path at the edge of the field.

[0095] A touch-sensitive guide plate 1231 installed on the cotton picking head senses the position of the cotton row. The touch-sensitive guide plate 1231 can trigger an angle sensor through a connecting plate mechanism. The angle signal generated when the touch-sensitive guide plate 1231 contacts the cotton plant is transmitted to the automatic row alignment unit 123 via a CAN bus. When the signal of the rear axle angle sensor 114 is maintained within a threshold range of -1° to 1°, it is determined that no deviation has occurred. If the signal exceeds the set threshold, a lateral correction command is generated, and a fuzzy PID control algorithm is used to dynamically adjust the correction amplitude to avoid over-steering. The fusion control unit 124 performs data fusion and outputs a real-time pose estimate based on the feedback signal from the touch-sensitive guide plate 1231. After each row is completed, the Hausdorff distance (threshold set at 3-5cm) between the actual path and the planned path is compared, and path iterative optimization is performed. If the error exceeds the limit, the subsequent global path is regenerated based on the current actual path to eliminate accumulated errors. The display module 13 adopts an industrial-grade touch screen. The interactive interface displays the cotton harvester's position, path deviation, device status, and alarm information in real time. It can also store data, record the operation trajectory, correction logs, and energy consumption data, and supports USB export and cloud synchronization.

[0096] During operation, the cotton harvester operates on a virtual electronic map constructed by the regional positioning unit 121. Based on the field location information of the cotton harvester obtained by the regional positioning unit 121, the first row harvested by the manually driven cotton harvester is used as the reference row. Based on the reference row, the path planning unit 122 generates a global operation path for the cotton harvester in the current row. After the cotton harvester enters the row, the touch-based automatic row alignment unit 123 corrects the planned operation path according to the actual situation of the cotton planting row, while the fusion control unit 124 records the actual path information of the current row. After the cotton harvester completes the operation of the current row, the fusion control unit 124 compares the current row path with the planned path. If the comparison error is large, the path planning unit 122 regenerates the operation path based on the corrected path as the reference row to eliminate the accumulated error at the intersection of parallel paths; otherwise, it continues to work according to the planned path. After the cotton harvester finishes harvesting the current row, it turns around at the end of the field using the path planned by the progressive fishtail turning algorithm. Figure 4 The algorithm path for the progressive fishtail turn-around is shown.

[0097] This invention solves the problems of strong signal dependence, low row alignment accuracy, and uncontrollable cumulative error in the existing automatic cotton harvester by constructing a local positioning network, touch-based dynamic row alignment correction, and adaptive path iterative optimization.

[0098] Of course, the present invention may have other various embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and modifications according to the present invention, but these corresponding changes and modifications should all fall within the protection scope of the appended claims.

Claims

1. A method for planning and controlling the path of an automated driving operation, characterized in that, The steps include the following: The step of constructing a local positioning network involves building a local positioning area with signal coverage for the plot to be operated, and obtaining the field location information of the harvester in real time through the regional positioning unit within the local positioning area; The steps for generating a baseline row and a global operation path are as follows: Based on the field location information, the first row harvested by the harvester is taken as the baseline row, and the path planning unit generates a global operation path for the harvester based on the current row based on the baseline row and the harvester's operating area. The touch-based dynamic row alignment and path correction steps involve the automatic row alignment unit correcting the planned global operation path based on the actual path of the planting row after the harvester enters the row for operation, while integrating the actual path information of the current row recorded by the control unit. as well as The path iteration optimization and error elimination steps are as follows: After the harvester completes the current row operation, it compares the current row path with the planned global operation path to determine whether the error is greater than a set value. If so, the path planning unit regenerates the subsequent global operation path based on the current actual path as the baseline row to eliminate the accumulated error at the intersection of parallel paths; otherwise, it continues to work according to the planned global operation path. In the steps of generating the baseline row and global operation path, the first row of harvesting by the manually driven harvester is used as the baseline row, and the trajectory coordinates are recorded in real time to generate the baseline path. The number of harvesting rows on the left and right sides of the baseline row is input respectively to obtain the harvester's operation area and area size. Based on the baseline row, the parallel path is expanded according to the standard row spacing of the ridge, and the turning path at the end of the field is planned using the progressive fishtail line turning algorithm.

2. The autonomous driving operation path planning and control method according to claim 1, characterized in that, In the step of constructing a local positioning network, a drone carrying a signal transmitter is used to park at the four corners and / or surroundings of the plot to build a local positioning area with signal coverage. A signal receiver adapted to the signal transmitter is set on the harvester. The area positioning unit is connected to the signal receiver to obtain the field location information of the harvester in real time and to construct a field operation map.

3. The autonomous driving operation path planning and control method according to claim 1, characterized in that, In the touch-based dynamic alignment and path correction steps, the automatic alignment unit is touch-based and uses a PID control algorithm based on real-time touch information. It performs two-layer path correction in conjunction with autonomous path replanning to enhance robustness in adapting to complex environments.

4. The autonomous driving operation path planning and control method according to claim 1, characterized in that, It also includes real-time display of job paths and data storage steps.

5. The autonomous driving operation path planning and control method according to claim 1, characterized in that, It also includes a full-cycle adaptive loop step, which automatically performs a positioning network health check after each reciprocating operation loop is completed, and triggers an early warning when the signal strength and / or drone battery life are abnormal.

6. An automated driving operation path planning and control device, characterized in that, The method for implementing the autonomous driving operation path planning and control method according to any one of claims 1-5 includes: The signal acquisition module includes a signal transmitter, a signal receiver, a touch sensor, a rear axle angle sensor, and a walking speed sensor. The signal transmitter is mounted on the drone, and the signal receiver is mounted on the harvester and adapted to the signal transmitter. The touch sensor, rear axle angle sensor, and walking speed sensor are respectively mounted on the harvester. The planning and control module, connected to the harvester's walking device, includes a regional positioning unit, a path planning unit, an automatic alignment unit, and a fusion control unit. The regional positioning unit is connected to the signal receiver. The touch sensor, rear axle angle sensor, and walking speed sensor are respectively connected to the automatic alignment unit. The path planning unit is connected to the regional positioning unit, the automatic alignment unit, and the fusion control unit.

7. The automated driving operation path planning and control device according to claim 6, characterized in that, Also includes: The display module, installed in the cab of the harvester and connected to the planning and control module, is used for real-time display of the work path and data storage.

8. The automated driving operation path planning and control device according to claim 6, characterized in that, The automatic alignment unit includes multiple touch-type signal acquisition components, evenly distributed at the front end of the harvester. Each touch-type signal acquisition component includes a first link, a second link, and a third link connected in sequence. The first link and the third link are symmetrically arranged, and the second link is connected to the front end of the harvester. Touch-type guide plates are symmetrically arranged at the ends of the first link and the third link, and touch sensors are respectively installed on the first link and the third link.

9. A combine harvester, characterized in that, The device includes the autonomous driving operation path planning and control device according to any one of claims 6-8, and the autonomous driving operation path planning and control method according to any one of claims 1-5 is used for harvesting operations.

Images