Agricultural machinery operation path intelligent planning method based on Beidou positioning

Abstract

The invention relates to the technical field of agricultural machinery automation control, and discloses an agricultural machinery operation path intelligent planning method based on Beidou positioning. The method comprises the following steps: firstly, acquiring Beidou positioning data and a farmland geographic data set, and eliminating positioning noise by using a dynamic filtering algorithm to obtain a calibrated agricultural machine track sequence; performing rasterization processing on the farmland geographic data to generate a digital elevation model, and dividing a job priority region; generating an initial operation path by adopting a hybrid optimization algorithm, and dynamically adjusting the path based on real-time operation state information; and finally, a path conflict detection module is combined with obstacle coordinates to calculate a collision probability, and a final agricultural machine operation path is output. According to the method, various factors are comprehensively considered, real-time dynamic adjustment of the path can be realized, path conflicts are effectively avoided, the operation precision and efficiency of agricultural machinery are improved, energy consumption and equipment loss are reduced, and agricultural mechanization and intelligentization development is promoted.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural machinery automation control, and specifically to an intelligent planning method for agricultural machinery operation paths based on Beidou positioning. Background Art

[0002] In the development of modern agriculture, the intelligent and efficient operation of agricultural machinery has become a key pursuit, and the limitations of traditional agricultural machinery operation path planning methods are becoming increasingly prominent. In the early days, agricultural machinery operations mostly relied on manual experience to plan routes. Operators relied on their familiarity with the farmland and past operating experience to determine the driving route. However, this method is significantly influenced by human factors, and the paths planned by different operators vary significantly, making it difficult to ensure standardized and efficient operations. For example, when operating on large areas of farmland, manual planning is prone to duplication of operations or omission of areas, resulting in low efficiency and increased energy consumption and wear of agricultural machinery. Moreover, it is difficult for humans to accurately consider the impact of complex factors such as terrain and soil on operations, making it impossible to achieve optimal resource allocation.

[0003] With technological advancements, some simple path planning algorithms have begun to be applied to agricultural machinery operations. However, these algorithms often rely solely on a single factor, such as planning a linear operation path based solely on the shape of the field boundary, without comprehensively considering multiple factors such as terrain elevation distribution, soil moisture parameters, and obstacles. In complex farmland, ignoring the slope factor can lead to insufficient power or excessive speed during climbing or descending, affecting operation quality and even posing safety risks. In areas with large variations in soil moisture, failing to prioritize operations can result in areas with water shortages or excess moisture not being addressed promptly and effectively, impacting crop growth.

[0004] Furthermore, existing agricultural machinery path planning methods fail to fully utilize real-time data. During operation, agricultural machinery's status (e.g., fuel consumption, operating speed) and operating environment (e.g., the appearance of new obstacles) constantly change. However, most traditional planning methods fix the path once the operation begins and are unable to dynamically adjust to these real-time changes. This makes it difficult for agricultural machinery to flexibly adapt to complex and changing operating scenarios, further reducing operational efficiency and resource utilization.

[0005] At the same time, existing path planning technologies also have shortcomings in handling path conflicts. For obstacles in farmland, such as trees and irrigation facilities, traditional methods either fail to accurately detect collision risks or, after detecting risks, lack effective local path replanning strategies. This can cause agricultural machinery to collide with obstacles, causing equipment damage and operational interruptions, resulting in serious losses to agricultural production. In summary, it is urgent to develop an intelligent path planning method for agricultural machinery operations that can comprehensively consider multiple factors, achieve real-time dynamic adjustments, and effectively avoid path conflicts, in order to meet the needs of modern agriculture for efficient and precise operations. Summary of the Invention

[0006] The purpose of the present invention is to provide an intelligent planning method for agricultural machinery operation paths based on Beidou positioning to solve the problems raised in the above background technology.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for intelligently planning agricultural machinery operation paths based on Beidou positioning, the method comprising: Obtaining Beidou positioning data and farmland geographic data sets; the Beidou positioning data includes real-time coordinates, movement speed, and operation status information of agricultural machinery; the farmland geographic data includes terrain elevation distribution, soil moisture parameters, and obstacle coordinates; Based on the Beidou positioning data, a dynamic filtering algorithm is used to eliminate positioning noise and generate a calibrated agricultural machinery trajectory sequence; Based on the farmland geographic data, a digital elevation model of the farmland is generated through rasterization processing, and operation priority areas are divided in combination with soil moisture parameters; A hybrid optimization algorithm is used to fuse the agricultural machinery trajectory sequence and the farmland digital elevation model to generate an initial operation path; the hybrid optimization algorithm includes a heuristic search and a constraint satisfaction mechanism; Based on real-time operation status information, the node spacing and steering angle of the initial operation path are dynamically adjusted through the feedback adjustment model to generate an optimized path; The optimized path is input into the path conflict detection module, and the collision probability is calculated in combination with the obstacle coordinates to output the final agricultural machinery operation path.

[0008] Preferably, eliminating positioning noise by a dynamic filtering algorithm includes: Perform outlier detection on Beidou positioning data to remove signal transition points and multipath interference data; The sliding window mean filter is used to smooth the real-time coordinates of the agricultural machinery and generate a denoised trajectory point set. The denoised trajectory point set is aligned with the farmland geographic coordinate system through the coordinate calibration algorithm, and the calibrated agricultural machinery trajectory sequence is output.

[0009] Preferably, the generating of the farmland digital elevation model by rasterization processing includes: Divide the grid cells according to the terrain elevation distribution data and calculate the slope and aspect characteristics of each cell; Based on soil moisture parameters and preset soil moisture thresholds, the priority weights of low-moisture and high-moisture areas are marked; The slope and aspect characteristics are integrated with priority weights to generate a multi-level farmland digital elevation model.

[0010] Preferably, the hybrid optimization algorithm includes: Initialize the starting and ending points of the heuristic search and set path constraints based on the farmland digital elevation model; Through the constraint satisfaction mechanism, feasible path nodes are screened to generate a set of candidate paths; The energy consumption and time consumption of the candidate path set are evaluated, and the path with the highest comprehensive score is selected as the initial operation path.

[0011] Preferably, the dynamically adjusting the initial operation path through the feedback regulation model includes: Real-time collection of agricultural machinery operation status information including fuel consumption, operation speed and steering delay parameters; Calculate the remaining operating mileage based on fuel consumption and operating speed, and dynamically compress or expand node spacing; The smoothness threshold of the path steering angle is adjusted based on the steering delay parameter to generate an optimized path.

[0012] Preferably, the path conflict detection module includes: Map the obstacle coordinates to the grid coordinate system of the optimized path and calculate the minimum distance between the path node and the obstacle; Based on the minimum distance and the moving speed of the agricultural machinery, the collision risk value is predicted through the probability density function; If the collision risk value exceeds the preset threshold, the local path is replanned to bypass the obstacle and the final agricultural machinery operation path is output.

[0013] Preferably, the method for determining the soil moisture threshold includes: Collect historical soil moisture data and the correlation between crop growth indicators, and fit the soil moisture impact function; Set dynamic soil moisture threshold intervals based on crop type and growth cycle; Based on real-time meteorological data, the dynamic soil moisture threshold is adaptively modified and the priority weight is updated.

[0014] Preferably, the parameter optimization method of the constraint satisfaction mechanism includes: According to the grid unit size of the farmland digital elevation model, the minimum step length and maximum turning angle of the path node are set; Through Monte Carlo simulation, we traverse different constraint parameter combinations and select the parameters with the best path coverage and energy consumption balance. Dynamically adjust constraints based on iteration results.

[0015] Preferably, the method for adjusting the smoothness threshold includes: Establish a mapping relationship table between steering angle and response time of agricultural machinery steering mechanism; Set the initial smoothness threshold based on the response time and calculate the threshold offset based on the real-time steering delay parameter; The smoothness threshold is dynamically compensated through linear interpolation to generate an optimized path that adapts to the current operation status.

[0016] Preferably, the local path replanning method includes: Taking the collision risk node as the center, extract the geographic data of the sub-area within the preset radius; A greedy algorithm is used to quickly generate a detour path in the sub-area and smoothly connect it with the global path; The feasibility of the detour path is verified through the energy consumption evaluation model, and the final agricultural machinery operation path is updated.

[0017] Compared with the prior art, the present invention has the following beneficial effects: The intelligent agricultural machinery operation path planning method based on Beidou positioning proposed in this invention has brought significant positive impacts in many aspects to agricultural production, effectively solved many problems existing in traditional agricultural machinery operation path planning, and strongly promoted the development of agricultural mechanization and intelligence.

[0018] In terms of data processing and path calibration, by acquiring Beidou positioning data and farmland geographic data sets, and using a dynamic filtering algorithm to eliminate positioning noise, a calibrated agricultural machinery trajectory sequence is generated. This process effectively improves the accuracy of the data and avoids operational deviations caused by positioning errors. For example, in complex electromagnetic environments or mountainous areas where signals are easily interfered with, traditional positioning methods may experience positioning drift, causing agricultural machinery operations to deviate from the predetermined route, resulting in uneven operations or omissions. The dynamic filtering algorithm of the present invention can accurately eliminate outliers, and through sliding window mean filtering and smoothing processing, it ensures the reliability of agricultural machinery trajectories, provides a solid data foundation for subsequent path planning, greatly improves the accuracy of agricultural machinery operations, reduces resource waste, and ensures comprehensive coverage and precise execution of farmland operations.

[0019] Operation strategies have been further optimized through rasterization of farmland geographic data and prioritization of operations. The generated digital elevation model of farmland incorporates information such as terrain elevation distribution and soil moisture parameters. Prioritized operation areas are divided based on soil moisture, allowing agricultural machinery to prioritize areas with a greater impact on crop growth. For example, in arid regions, areas with low soil moisture pose a significant constraint on crop growth. Prioritizing irrigation or other operations can effectively improve crop survival rates and yields. Furthermore, planning routes based on terrain slope and aspect can prevent agricultural machinery from operating on unsuitable terrain, reducing equipment wear and tear, extending the service life of agricultural machinery, improving operational safety, and reducing operational risks caused by terrain factors, such as agricultural machinery rollovers.

[0020] The application of a hybrid optimization algorithm significantly improves the scientific nature of path planning. By integrating agricultural machinery trajectory sequences with a digital elevation model of farmland, an initial operation path is generated through a heuristic search and constraint satisfaction mechanism. The candidate paths are then evaluated for energy and time consumption, and the path with the highest overall score is selected. This allows the planned path to achieve an optimal balance between energy consumption and time while satisfying the actual constraints of agricultural machinery operations. Compared to traditional paths that are arbitrarily planned or that only consider a single factor, the paths planned by this invention can significantly reduce agricultural machinery energy consumption and shorten operation time.

[0021] A real-time feedback adjustment mechanism ensures dynamic path optimization. Based on the real-time operating status of the agricultural machinery, such as fuel consumption, operating speed, and steering delay parameters, the node spacing and steering angle of the initial operating path are dynamically adjusted. During agricultural machinery operations, when fuel consumption is rapid or the remaining fuel is insufficient, the path node spacing can be adjusted in a timely manner to reduce unnecessary driving distance and ensure that the agricultural machinery completes operations in critical areas before the fuel runs out. The steering angle smoothness threshold is adjusted based on the steering delay parameters, making the agricultural machinery's steering smoother, reducing mechanical wear, and improving operational comfort and stability, further ensuring the quality and efficiency of agricultural machinery operations.

[0022] The path conflict detection module provides a strong guarantee for the safety of agricultural machinery operations. It calculates collision probabilities based on obstacle coordinates. If a collision risk is detected, the local path is promptly replanned to bypass the obstacle. This effectively prevents agricultural machinery from colliding with obstacles during operation, reduces equipment damage and repair costs, and mitigates economic losses caused by operational interruptions. Even in the presence of obstacles such as irrigation pipes and utility poles in farmland, the present invention can quickly and accurately detect risks and plan detours, ensuring the continuity and safety of agricultural machinery operations and the smooth progress of agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a working principle diagram of the intelligent planning method for agricultural machinery operation paths based on Beidou positioning according to the present invention; Figure 2 Flowchart for generating digital elevation models of farmland for rasterization processing; Figure 3 Flowchart for dynamically adjusting the initial operation path for the feedback regulation model; Figure 4 Flowchart of a method for local path replanning. DETAILED DESCRIPTION

[0024] 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 creative efforts are within the scope of protection of the present invention.

[0025] See also Figures 1-4 The present invention relates to an intelligent planning method for agricultural machinery operation paths based on Beidou positioning, and its specific implementation method is elaborated in detail below.

[0026] Obtain Beidou positioning data and farmland geographic data sets. Beidou positioning data includes the real-time coordinates, movement speed, and operating status of agricultural machinery. This data is obtained using Beidou positioning equipment installed on the machinery, which collects and transmits real-time information such as the position, speed, and operating status of the machinery during operation. Farmland geographic data includes terrain elevation distribution, soil moisture parameters, and obstacle coordinates. Terrain elevation data can be obtained using professional topographic surveying equipment, soil moisture parameters can be collected using soil moisture monitoring instruments, and obstacle coordinates can be determined through field surveys combined with satellite imagery.

[0027] Based on Beidou positioning data, a dynamic filtering algorithm eliminates positioning noise and generates a calibrated agricultural machinery trajectory sequence. In real-world operating environments, Beidou positioning data is susceptible to various interferences, which can lead to noise and affect the accuracy of agricultural machinery trajectories. By processing positioning data with a specific dynamic filtering algorithm, noise can be effectively eliminated, providing accurate trajectory data for subsequent operation path planning.

[0028] Based on farmland geographic data, a digital elevation model of farmland is generated through rasterization, and soil moisture parameters are combined to prioritize operation areas. Rasterization of farmland geographic data can transform complex terrain and soil information into a digital model that is easy for computers to process. This, combined with soil moisture parameters, can improve the efficiency of agricultural machinery operations and prioritize areas with the greatest impact on crop growth.

[0029] A hybrid optimization algorithm was used to fuse the agricultural machinery trajectory sequence with the farmland digital elevation model to generate an initial operation path. This hybrid optimization algorithm combines heuristic search with a constraint satisfaction mechanism, fully considering the actual constraints of farmland topography and agricultural machinery operations. The generated initial operation path is more consistent with actual operation requirements.

[0030] Based on real-time operating status information, a feedback adjustment model dynamically adjusts the node spacing and steering angles of the initial operating path to generate an optimized path. During operation, agricultural machinery's operating status, such as fuel consumption and operating speed, constantly changes. The feedback adjustment model dynamically adjusts the initial operating path based on real-time operating status information, optimizing the path and improving operational efficiency and quality.

[0031] The optimized path is input into the Path Conflict Detection module, which calculates the collision probability based on obstacle coordinates and outputs the final agricultural machinery operation path. Taking into account the potential for obstacles in the field, the Path Conflict Detection module calculates the collision probability between the path and the obstacle, adjusts paths with collision risks, ensures the safety of agricultural machinery during operation, and outputs the final safe and feasible operation path.

[0032] The technical solution of the present invention is further described below through specific embodiments: Example 1: In this embodiment, the specific process of eliminating positioning noise by using a dynamic filtering algorithm is described in detail.

[0033] Outlier detection is performed on Beidou positioning data. During actual data collection, signal transition points and multipath interference data may occur due to factors such as signal interference. These outliers are identified by setting reasonable threshold ranges and rules for determining data change trends. For example, if the real-time coordinates of agricultural machinery experience a significant sudden change within a short period of time, and this change is inconsistent with the normal movement speed and direction of the machinery, then this coordinate point may be identified as a signal transition point. Multipath interference data can be identified by analyzing characteristics such as signal strength and propagation time. Once an outlier is detected, it is removed to ensure data accuracy.

[0034] A sliding window mean filter is used to smooth the real-time coordinates of agricultural machinery. A sliding window of appropriate size is set, containing a certain number of consecutive coordinate points. The average value of the coordinate points within the window is calculated and used to replace the coordinate point at the center of the window. As the window slides across the coordinate sequence, this smoothing process is repeated for each coordinate point, generating a denoised trajectory point set. For example, if the sliding window size is set to 5, for the third point in the coordinate sequence, the average value of the coordinates of points 1-5 is calculated and used as the denoised coordinate of the third point.

[0035] A coordinate calibration algorithm aligns the denoised trajectory point set with the farmland's geographic coordinate system. Calibration is required because the coordinate data collected by agricultural machinery positioning equipment may deviate from the farmland's geographic coordinate system. Based on known farmland geographic reference points, the coordinate calibration algorithm uses a coordinate transformation formula to convert the coordinates of the denoised trajectory point set to the farmland's geographic coordinate system. The resulting calibrated agricultural machinery trajectory sequence provides accurate foundational data for subsequent path planning.

[0036] Example 2: This embodiment describes in detail the specific steps of generating a farmland digital elevation model through rasterization processing.

[0037] Grid cells are divided based on terrain elevation distribution data. The appropriate grid size is determined based on the actual size of the farmland and the required accuracy. For example, for large farmland with relatively flat terrain, a larger grid size, such as 10 meters by 10 meters, can be selected; for areas with complex terrain and high accuracy, a smaller grid size, such as 1 meter by 1 meter, can be selected. After the grid cells are divided, the slope and aspect characteristics of each cell are calculated. The slope can be calculated by the elevation difference and horizontal distance between adjacent grid cells using the following formula: in Indicates the slope, and Represent the elevations of two adjacent grid cells, Represents the horizontal distance between adjacent grid cells. The slope aspect can be determined by calculating the direction of elevation change.

[0038] Based on soil moisture parameters and preset soil moisture thresholds, priority weights are assigned to areas of low and high moisture. Soil moisture parameters reflect soil moisture content, while preset soil moisture thresholds are determined based on the growth requirements of different crops and empirical data. For example, for a particular crop, when soil moisture falls below a certain threshold, the area is marked as low moisture and given a higher priority for operations. When soil moisture exceeds a certain threshold, the area is marked as high moisture and given a lower priority for operations.

[0039] Slope and aspect characteristics are integrated with priority weights to generate a multi-level farmland digital elevation model. In this generated model, each grid cell not only contains terrain elevation information but also incorporates information such as slope, aspect, and operation priority weights. This multi-level representation more comprehensively reflects the geographical characteristics of the farmland, providing richer data support for subsequent path planning.

[0040] Example 3: This embodiment deeply explores the specific implementation process of the hybrid optimization algorithm.

[0041] Initialize the starting and ending points of the heuristic search. The starting point can be set to the coordinates of the current location of the agricultural machinery, while the ending point is determined based on the task objectives, such as the boundaries of the farmland or areas requiring focused work. Path constraints are set based on the farmland digital elevation model. These constraints include terrain slope limits and agricultural machinery turning radius limits. For example, if the maximum gradeability of the agricultural machinery is 20°, the slope of the selected path nodes during path planning cannot exceed this limit. Furthermore, a maximum steering angle limit is set based on the actual turning performance of the agricultural machinery to ensure path feasibility.

[0042] A constraint satisfaction mechanism is used to screen feasible path nodes and generate a set of candidate paths. During the search process, each possible path node is evaluated based on the set constraints, and only nodes that meet all constraints are retained. For example, if the slope of a path node exceeds the maximum climbing capacity of the agricultural machinery, or the steering angle of the node exceeds the maximum steering angle limit, the node is discarded. Through continuous screening, multiple candidate paths that meet the requirements are generated.

[0043] The energy consumption and time consumption of the candidate path set are evaluated, and the path with the highest comprehensive score is selected as the initial operation path. The energy consumption evaluation can be calculated based on the energy consumption model of agricultural machinery under different terrains and operating speeds, and the time consumption is estimated based on the path length and the operating speed of the agricultural machinery. For example, the energy consumption calculation formula can be: in represents the total energy consumption, Indicates that agricultural machinery Energy consumption per unit distance on the segment path, Indicates the By comprehensively considering energy consumption and time consumption, each candidate path is scored and the path with the highest score is selected as the initial operation path.

[0044] Example 4: This embodiment introduces in detail a specific method for dynamically adjusting the initial operation path through a feedback adjustment model.

[0045] Real-time information on agricultural machinery operation status, including fuel consumption, operating speed, and steering delay, is collected. Sensors installed on the machinery provide this information. For example, fuel level sensors can monitor fuel consumption, speed sensors can measure operating speed, and steering sensors can capture steering delay parameters.

[0046] Calculate the remaining operating mileage based on fuel consumption and operating speed. Based on the fuel consumption rate and current remaining fuel of the agricultural machinery, combined with the operating speed, the mileage that the agricultural machinery can continue to operate can be calculated. Assuming the fuel consumption rate is (Unit: liters / kilometer), the current remaining fuel volume is (Unit: liter), the operating speed is (Unit: km / h), then the remaining operating mileage Dynamically compress or expand node spacing based on the remaining operating mileage. If the remaining operating mileage is short, the node spacing can be appropriately compressed to complete the task and improve operating efficiency. If the remaining operating mileage is long, the node spacing can be appropriately expanded to reduce frequent steering and energy consumption of agricultural machinery.

[0047] The smoothness threshold of the path's steering angle is adjusted based on the steering delay parameter to generate an optimized path. The steering delay parameter reflects the response speed of the agricultural machinery's steering mechanism, and the smoothness threshold of the path's steering angle is set accordingly. For example, a large steering delay indicates a slow response. To ensure operational safety and stability, the smoothness threshold should be increased to achieve smoother path steering. Conversely, the smoothness threshold can be appropriately decreased. By adjusting the smoothness threshold, the initial operation path is optimized to generate a path that better meets actual operational needs.

[0048] Example 5: This embodiment describes in detail the specific implementation of the path conflict detection module.

[0049] Map the obstacle coordinates to the grid coordinate system of the optimized path. Based on the grid division method used for the optimized path, convert the actual coordinates of the obstacle to the corresponding grid coordinate system. For example, if the grid size of the optimized path is 5 meters by 5 meters, and the actual coordinates of the obstacle are (x, y), then use the coordinate conversion formula to convert them to the coordinates (i, j) in the grid coordinate system.

[0050] Calculate the minimum distance between the path node and the obstacle. For each node on the optimized path, calculate the distance between it and the obstacle mapped to the grid coordinate system. The distance can be calculated using the Euclidean distance formula, which is: in Represents the distance between two points, and Denote the coordinates of the path nodes and obstacles in the grid coordinate system respectively. Find the minimum distance as the basis for evaluating the collision risk.

[0051] Based on the minimum distance and the speed of the agricultural machinery, a probability density function is used to predict the collision risk. Based on empirical data and actual testing, a probability density function relationship is established between the minimum distance, the speed of the agricultural machinery, and the collision risk value. For example, the smaller the minimum distance and the faster the agricultural machinery speed, the higher the collision risk value. If the collision risk value exceeds a preset threshold, the local path is replanned to bypass the obstacle, and the final agricultural machinery operation path is output. When replanning the local path, optimization algorithms such as the A* algorithm can be used to ensure the rationality and feasibility of the detour path.

[0052] Example 6: This embodiment comprehensively describes a method for determining a soil moisture threshold, a method for optimizing parameters of a constraint satisfaction mechanism, a method for adjusting a smoothness threshold, and a method for replanning a local path.

[0053] To determine soil moisture thresholds, the method first collects historical soil moisture data and correlates it with crop growth indicators to fit a soil moisture impact function. Through long-term field trials and data monitoring, crop growth data under different soil moisture conditions, such as yield and growth height, are collected. Using data analysis methods, a functional relationship between soil moisture and crop growth, such as a linear or nonlinear regression function, is fitted. Dynamic soil moisture threshold ranges are set based on crop type and growth cycle. Different crops have different soil moisture requirements at different growth stages. For example, soil moisture requirements are relatively low during the seedling stage, while higher requirements are seen during the flowering and grain filling stages. Based on these characteristics, dynamic soil moisture threshold ranges are set for different crops throughout their growth cycles. The dynamic soil moisture thresholds are adaptively adjusted based on real-time meteorological data, updating priority weights. Real-time meteorological data, such as precipitation and evaporation, can influence soil moisture. Adjusting the dynamic soil moisture thresholds based on this data allows for more accurate delineation of operational priority zones.

[0054] For the parameter optimization method of the constraint satisfaction mechanism, the minimum step size and maximum steering angle of the path nodes are set according to the grid cell size of the farmland digital elevation model. The grid cell size determines the accuracy of path planning, and the minimum step size and maximum steering angle are appropriately set based on the grid size. For example, if the grid cell size is small, a smaller minimum step size can be set to improve path planning accuracy. At the same time, a reasonable maximum steering angle is set based on the performance of the agricultural machinery. Through Monte Carlo simulation, different constraint parameter combinations are traversed to select the parameters with the optimal path coverage and energy consumption balance. Monte Carlo simulation simulates path planning under different constraint parameter combinations through multiple random experiments, calculates path coverage and energy consumption balance indicators, and selects the optimal parameter combination. Based on the iterative results, the constraints are dynamically adjusted. As the operation progresses, the constraints are continuously adjusted according to the actual situation to improve the adaptability of path planning.

[0055] Regarding the smoothness threshold adjustment method, a mapping relationship table is established between the steering angle and the response time of the agricultural machinery's steering mechanism. Through testing and data analysis of the agricultural machinery's steering mechanism, response time data at different steering angles is established to form a mapping relationship table. The initial smoothness threshold is set based on the response time, and the threshold offset is calculated in combination with the real-time steering delay parameter. The longer the response time, the larger the initial smoothness threshold. The threshold offset is calculated based on the difference between the real-time steering delay parameter and the average response time. The smoothness threshold is dynamically compensated through linear interpolation to generate an optimized path adapted to the current operating state. Based on the threshold offset, the smoothness threshold is adjusted using linear interpolation to make the path smoother and adapt to the actual operating conditions of the agricultural machinery.

[0056] For the local path replanning method, geographic data for a sub-area within a preset radius, centered around the collision risk node, is extracted. This preset radius is determined based on the machine's operating range and actual conditions. Geographic data, including terrain and obstacles, is then extracted within this sub-area. A greedy algorithm is used to rapidly generate a detour path within the sub-area and smoothly connect it to the global path. The greedy algorithm takes the optimal decision for each step, rapidly generating the detour path. Then, using algorithms such as spline curve fitting, the detour path is smoothly connected to the global path. The feasibility of the detour path is verified using an energy consumption evaluation model, and the final machine operation path is updated. The energy consumption evaluation model determines the feasibility of the detour path based on the machine's energy consumption along the detour path. If energy consumption is too high or does not meet actual operating requirements, the detour path is readjusted until it meets the requirements. Finally, the final machine operation path is updated and output.

[0057] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0058] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent planning of agricultural machinery operation paths based on Beidou positioning, characterized in that: include: Obtain Beidou positioning data and farmland geographic data sets; The Beidou positioning data includes the real-time coordinates, movement speed and operation status information of the agricultural machinery, and the farmland geographic data includes the terrain elevation distribution, soil moisture parameters and obstacle coordinates; Based on the Beidou positioning data, a dynamic filtering algorithm is used to eliminate positioning noise and generate a calibrated agricultural machinery trajectory sequence; Based on the farmland geographic data, a digital elevation model of the farmland is generated through rasterization processing, and operation priority areas are divided in combination with soil moisture parameters; A hybrid optimization algorithm is used to fuse the agricultural machinery trajectory sequence and the farmland digital elevation model to generate an initial operation path; the hybrid optimization algorithm includes a heuristic search and a constraint satisfaction mechanism; Based on real-time operation status information, the node spacing and steering angle of the initial operation path are dynamically adjusted through the feedback adjustment model to generate an optimized path; The optimized path is input into the path conflict detection module, and the collision probability is calculated in combination with the obstacle coordinates to output the final agricultural machinery operation path.

2. The method for intelligent planning of agricultural machinery operation paths based on Beidou positioning according to claim 1 is characterized in that: Eliminating positioning noise by a dynamic filtering algorithm includes: Perform outlier detection on Beidou positioning data to remove signal transition points and multipath interference data; The sliding window mean filter is used to smooth the real-time coordinates of the agricultural machinery and generate a denoised trajectory point set. The denoised trajectory point set is aligned with the farmland geographic coordinate system through the coordinate calibration algorithm, and the calibrated agricultural machinery trajectory sequence is output.

3. The method for intelligent planning of agricultural machinery operation paths based on Beidou positioning according to claim 1 is characterized in that: The method of generating a farmland digital elevation model by rasterization processing includes: Divide the grid cells according to the terrain elevation distribution data and calculate the slope and aspect characteristics of each cell; Based on soil moisture parameters and preset soil moisture thresholds, the priority weights of low-moisture and high-moisture areas are marked; The slope and aspect characteristics are integrated with priority weights to generate a multi-level farmland digital elevation model.

4. The method for intelligent planning of agricultural machinery operation paths based on Beidou positioning according to claim 1 is characterized in that: The hybrid optimization algorithm includes: Initialize the starting and ending points of the heuristic search and set path constraints based on the farmland digital elevation model; Through the constraint satisfaction mechanism, feasible path nodes are screened to generate a set of candidate paths; The energy consumption and time consumption of the candidate path set are evaluated, and the path with the highest comprehensive score is selected as the initial operation path.

5. The method for intelligent planning of agricultural machinery operation paths based on Beidou positioning according to claim 1 is characterized in that: The dynamic adjustment of the initial operation path through the feedback regulation model includes: Real-time collection of agricultural machinery operation status information including fuel consumption, operation speed and steering delay parameters; Calculate the remaining operating mileage based on fuel consumption and operating speed, and dynamically compress or expand node spacing; The smoothness threshold of the path steering angle is adjusted based on the steering delay parameter to generate an optimized path.

6. The method for intelligent planning of agricultural machinery operation paths based on Beidou positioning according to claim 1, characterized in that: The path conflict detection module includes: Map the obstacle coordinates to the grid coordinate system of the optimized path and calculate the minimum distance between the path node and the obstacle; Based on the minimum distance and the moving speed of the agricultural machinery, the collision risk value is predicted through the probability density function; If the collision risk value exceeds the preset threshold, the local path is replanned to bypass the obstacle and the final agricultural machinery operation path is output.

7. The method for intelligent planning of agricultural machinery operation paths based on Beidou positioning according to claim 3 is characterized in that: The method for determining the soil moisture threshold comprises: Collect historical soil moisture data and the correlation between crop growth indicators, and fit the soil moisture impact function; Set dynamic soil moisture threshold intervals based on crop type and growth cycle; Based on real-time meteorological data, the dynamic soil moisture threshold is adaptively modified and the priority weight is updated.

8. The method for intelligent planning of agricultural machinery operation paths based on Beidou positioning according to claim 4 is characterized in that: The parameter optimization method of the constraint satisfaction mechanism includes: According to the grid unit size of the farmland digital elevation model, the minimum step length and maximum turning angle of the path node are set; Through Monte Carlo simulation, we traverse different constraint parameter combinations and select the parameters with the best path coverage and energy consumption balance. Dynamically adjust constraints based on iteration results.

9. The method for intelligent planning of agricultural machinery operation paths based on Beidou positioning according to claim 5, characterized in that: The method for adjusting the smoothness threshold includes: Establish a mapping relationship table between steering angle and response time of agricultural machinery steering mechanism; Set the initial smoothness threshold based on the response time and calculate the threshold offset based on the real-time steering delay parameter; The smoothness threshold is dynamically compensated through linear interpolation to generate an optimized path that adapts to the current operation status.

10. The method for intelligent planning of agricultural machinery operation paths based on Beidou positioning according to claim 6, characterized in that: The local path replanning method includes: Taking the collision risk node as the center, extract the geographic data of the sub-area within the preset radius; A greedy algorithm is used to quickly generate detour paths in the sub-area and smoothly connect them with the global path; The feasibility of the detour path is verified through the energy consumption evaluation model, and the final agricultural machinery operation path is updated.

Images