Method and system for planning field operation path of agricultural machinery equipment in hilly and mountainous regions

Abstract

The invention relates to the technical field of agricultural machinery operation, in particular to a field operation path planning method and system for agricultural machinery equipment in hilly and mountainous regions, and the method comprises the following steps: obtaining terrain data of the hilly and mountainous regions and operation data of the agricultural machinery equipment, so as to construct an environment equipment cooperation data set; analyzing the environment equipment collaborative data set to obtain a field operation segmentation area and a field operation priority as an agricultural machinery operation analysis result; based on the agricultural machinery operation analysis result, establishing a four-wheel steering kinematics model and an agricultural machinery trafficability constraint model as mechanical constraint conditions of agricultural machinery equipment; generating an initial agricultural machine operation path by combining the agricultural machine operation analysis result and the agricultural machine equipment mechanical constraint condition; and performing multi-target adaptive adjustment on the initial agricultural machine operation path according to an agricultural machine operation mode to obtain an optimal agricultural machine operation path. The method is combined with mechanical characteristics of agricultural machinery equipment, the accuracy of path planning is improved, and complex terrains of hills and mountains are effectively adapted.

Description

Technical Field

[0001] This invention relates to the field of agricultural machinery operation technology, and in particular to a method and system for planning field operation paths for agricultural machinery equipment in hilly and mountainous areas. Background Technology

[0002] Currently, the planning of field operation paths for agricultural machinery in hilly and mountainous areas mainly relies on a combination of global navigation satellite systems (GNSS) and inertial measurement units (IMUs) for positioning and navigation. Existing technologies typically plan regular geometric paths such as bow-shaped, spiral, and zigzag patterns in a two-dimensional plane based on pre-acquired farmland boundary information. Satellite positioning determines the absolute position of the agricultural machinery in the field, thereby controlling the machinery to travel along a preset baseline trajectory. Gyroscopes and accelerometers are used to compensate for attitude changes to maintain the accuracy of straight-line travel. At the control level, classic proportional-integral-derivative (PI-DI) controllers or fuzzy control algorithms are often used to provide feedback adjustment to the angle of the machinery's steering wheels to track the predetermined path. Some technologies introduce digital elevation models to coarsely compensate for slope gradients or set fixed obstacle-free zones to attempt to adapt to complex terrain.

[0003] However, existing technologies have limitations, including: First, path planning is generally based on the assumption of a two-dimensional plane and idealized movement of agricultural machinery, seriously neglecting the dynamic coupling relationship between the unique three-dimensional terrain undulations of hilly areas and the mechanical structure of the agricultural machinery itself, resulting in insufficient feasibility of the planned paths; Second, existing solutions mostly adopt a single, fixed operation mode, failing to make differentiated adaptive adjustments to the trajectory of agricultural machinery according to different operation modes, making it difficult to balance operation quality and efficiency; Finally, there are barriers at the data utilization level, with terrain data and agricultural machinery operation status data being isolated from each other, failing to form a collaborative dataset for analysis and decision-making, resulting in a lack of comprehensive measurement of dynamic energy consumption, terrain passability, and task priority in agricultural machinery field operation path planning, making agricultural machinery poorly adaptable, inefficient, and wasteful of energy when operating in complex hilly areas. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a method and system for planning field operation paths for agricultural machinery in hilly and mountainous areas.

[0005] To achieve the above objectives, in a first aspect, the present invention provides a method for planning field operation paths for agricultural machinery in hilly and mountainous areas. The method includes the following steps: acquiring hilly and mountainous terrain data and agricultural machinery operation data to construct an environmental equipment collaborative dataset; parsing the environmental equipment collaborative dataset to obtain field operation segmentation areas and field operation priorities as agricultural machinery operation analysis results; establishing a four-wheel steering kinematic model and an agricultural machinery passability constraint model as mechanical constraints for the agricultural machinery based on the agricultural machinery operation analysis results; generating an initial agricultural machinery operation path by combining the agricultural machinery operation analysis results and the agricultural machinery mechanical constraints; and performing multi-objective adaptive adjustments on the initial agricultural machinery operation path according to the agricultural machinery operation mode to obtain the optimal agricultural machinery operation path. This invention comprehensively considers the complex terrain of hilly and mountainous areas and the characteristics of agricultural machinery, achieving scientific and accurate path planning, effectively balancing operation efficiency, operation quality, and agricultural machinery adaptability, avoiding operation interruptions or energy waste caused by unreasonable paths, and significantly improving the stability, safety, and overall benefits of agricultural machinery field operations in hilly and mountainous areas.

[0006] Optionally, acquiring hilly terrain data and agricultural machinery operation data to construct an environmental equipment collaborative dataset includes: acquiring the hilly terrain data using lidar and collecting the agricultural machinery operation data based on onboard sensors; performing spatiotemporal registration and data alignment on the hilly terrain data and the agricultural machinery operation data to form a multidimensional data set; and performing feature extraction and semantic annotation on the multidimensional data set to generate the environmental equipment collaborative dataset. This invention constructs a collaborative dataset to ensure the comprehensiveness, accuracy, and relevance of the data; it provides high-quality data support for subsequent operational area analysis, constraint model establishment, and path optimization, reducing path planning errors caused by data mismatch.

[0007] Optionally, the step of parsing the environmental equipment collaborative dataset to obtain field operation segmentation areas and field operation priorities as agricultural machinery operation parsing results includes: based on the environmental equipment collaborative dataset, using a dynamic clustering algorithm to divide continuous areas with similar terrain features into the same operation sub-region as the field operation segmentation areas; constructing a multi-factor priority evaluation function to evaluate the field operation segmentation areas and obtain the field operation priorities. This invention uses a dynamic clustering algorithm to rationally segment operation areas, combines multi-factor priority evaluation to clarify the operation sequence, adapts to the highly heterogeneous characteristics of hilly and mountainous terrain, avoids inefficiency caused by blind operations, achieves scientific planning of operation areas, and improves the logic and targeting of path planning.

[0008] Optionally, the step of constructing a multi-factor priority evaluation function to evaluate the field operation segmentation areas and obtain the field operation priority includes: obtaining priority evaluation factors for the field operation segmentation areas, including soil suitability index, crop growth status index, agricultural machinery-area distance, and operation urgency; determining the dynamic weight coefficients of the priority evaluation factors, and constructing the multi-factor priority evaluation function in combination with the remaining agricultural machinery battery power; obtaining a comprehensive priority score for the field operation segmentation areas based on the multi-factor priority evaluation function; and ranking the field operation segmentation areas according to the comprehensive priority score to obtain the field operation priority. This invention, by extracting multi-dimensional priority evaluation factors, combining dynamic weight coefficients, and combining agricultural machinery battery power to construct an evaluation function, achieves precise quantification of operation area priorities, ensures that priority ranking aligns with actual field conditions, guarantees priority implementation in key operation areas, and improves the rationality of path planning.

[0009] Optionally, establishing a four-wheel steering kinematic model and an agricultural machinery passability constraint model based on the agricultural machinery operation analysis results as mechanical constraints for agricultural machinery equipment includes: determining the field operation segmentation area to be planned based on the agricultural machinery operation analysis results, and obtaining the terrain height from the hilly and mountainous terrain data of the field operation segmentation area; obtaining agricultural machinery attribute parameters, including the agricultural machinery wheelbase, agricultural machinery turning radius, adjustable agricultural machinery track, and agricultural machinery chassis ground clearance; establishing the four-wheel steering kinematic model based on Ackermann steering geometric constraints, combined with the agricultural machinery wheelbase, agricultural machinery track, and agricultural machinery turning radius; and conducting passability constraint analysis based on the terrain height and the agricultural machinery chassis ground clearance to construct the agricultural machinery passability constraint model. This invention accurately characterizes the characteristics of agricultural machinery and terrain adaptability requirements through agricultural machinery equipment mechanical constraints, effectively avoiding situations where the planned path exceeds mechanical constraints, and improving the agricultural machinery adaptability of the path.

[0010] Optionally, generating the initial agricultural machinery operation path by combining the agricultural machinery operation analysis results and the mechanical constraints of the agricultural machinery equipment includes: determining the global operation sequence of the field operation segmentation areas based on the field operation priority; generating internal operation paths based on the boundary geometric features of the field operation segmentation areas, including multi-machine collaborative task decomposition areas, operation travel directions, operation strips, and connecting turning paths; verifying the internal operation paths based on the mechanical constraints of the agricultural machinery equipment to obtain internal operation optimization paths; and connecting the internal operation optimization paths based on the global operation sequence to generate the initial agricultural machinery operation path. The initial agricultural machinery operation path constructed by this invention ensures the rationality of the internal operation paths of each area and achieves orderly connection of global operations, avoiding operational chaos or path conflicts between areas and improving the overall feasibility of path planning.

[0011] Optionally, the step of verifying the internal operation path based on the mechanical constraints of the agricultural machinery to obtain the optimized internal operation path includes: verifying whether the radius of curvature of the connecting steering path meets the minimum turning radius constraint of the agricultural machinery using the four-wheel steering kinematic model; verifying whether the elevation changes of the path points and the crop growth status of the operation strip are within the safe passage range of the agricultural machinery chassis based on the agricultural machinery passability constraint model; for path segments that do not meet the mechanical constraints of the agricultural machinery, using path smoothness and operation continuity as optimization indicators, using a sequential quadratic programming algorithm to obtain optimized path segments that meet the mechanical constraints of the agricultural machinery; and integrating the optimized path segments and the internal operation path of the region to obtain the optimized internal operation path. This invention verifies path curvature and elevation changes through mechanical constraints, optimizes unqualified path segments using a sequential quadratic programming algorithm, and ensures path smoothness and operation continuity while meeting mechanical constraints, thereby improving the reliability and adaptability of the internal path of the region.

[0012] Optionally, the step of performing multi-objective adaptive adjustment of the initial agricultural machinery operation path according to the agricultural machinery operation mode to obtain the optimal agricultural machinery operation path includes: acquiring the agricultural machinery operation mode, including sowing mode, spraying mode, and fertilization mode, and determining the operation parameter configuration set corresponding to the agricultural machinery operation mode; constructing a multi-objective optimization function for the agricultural machinery operation mode based on the operation parameter configuration set; and iteratively optimizing the path parameters of the initial agricultural machinery operation path using an adaptive weighted particle swarm optimization algorithm based on the environmental equipment collaborative dataset to solve the multi-objective optimization function and obtain the optimal agricultural machinery operation path. This invention constructs multi-objective optimization functions for different operation modes, uses an adaptive weighted particle swarm optimization algorithm to iteratively optimize path parameters, adapts to the needs of multiple operation modes, balances operation quality, energy consumption, and efficiency, realizes dynamic adaptive adjustment of the path, and improves the operational efficiency of agricultural machinery in different operation scenarios.

[0013] Optionally, constructing the multi-objective optimization function for the agricultural machinery operation mode based on the operation parameter configuration set includes: extracting key operation parameters of the agricultural machinery operation mode based on the operation parameter configuration set; establishing multi-dimensional optimization objectives, including operation quality, energy consumption, and operation efficiency, and determining the balance adjustment coefficient of the multi-dimensional optimization objectives based on the key operation parameters; and fusing the multi-dimensional optimization objectives according to the balance adjustment coefficient to construct the multi-objective optimization function. This invention, by constructing a multi-objective optimization function, matches the core requirements of different operation modes, making the optimization function fit the actual operation scenario, providing a scientific and accurate objective guide for optimal path solving, and improving the pertinence of path optimization.

[0014] Secondly, this invention provides a field operation path planning system for agricultural machinery in hilly and mountainous areas. The system executes the field operation path planning method for agricultural machinery in hilly and mountainous areas provided by this invention. The system includes an input device, an output device, a processor, and a memory, which are interconnected. The memory stores a computer program, which includes program instructions, and the processor is configured to call the program instructions. This invention, through the collaboration of high-performance hardware, achieves automatic planning of agricultural machinery operation paths in hilly and mountainous areas, improves path planning efficiency, reduces human error, ensures the reliability and practicality of the planning results, and provides stable hardware support for agricultural machinery field operation path planning. Attached Figure Description

[0015] Figure 1 This is a flowchart of a method for planning field operation paths for agricultural machinery in hilly and mountainous areas, according to an embodiment of the present invention. Figure 2 This is a framework diagram of a field operation path planning system for agricultural machinery equipment in hilly and mountainous areas, according to an embodiment of the present invention. Detailed Implementation

[0016] Specific embodiments of the present invention will now be described in detail. It should be noted that the embodiments described herein are for illustrative purposes only and are not intended to limit the invention. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that these specific details are not necessary to practice the invention. In other instances, well-known circuits, software, or methods have not been specifically described to avoid obscuring the invention.

[0017] Throughout this specification, references to "an embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in connection with that embodiment or example is included in at least one embodiment of the invention. Therefore, the phrases "in an embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily refer to the same embodiment or example. Furthermore, specific features, structures, or characteristics can be combined in one or more embodiments or examples in any suitable combination and / or sub-combination. Moreover, those skilled in the art will understand that the illustrations provided herein are for illustrative purposes and are not necessarily drawn to scale.

[0018] Please see Figure 1 One embodiment of the present invention provides a method for planning field operation paths for agricultural machinery in hilly and mountainous areas, the method comprising the following steps: S1. Obtain hilly and mountainous terrain data and agricultural machinery operation data to construct an environmental equipment collaborative dataset.

[0019] In this embodiment, for hilly and mountainous terrain data, a drone or a lidar mounted on a fixed point is used to scan the area to obtain high-precision three-dimensional point cloud data of the work area. After processing, the three-dimensional point cloud data is used to generate a digital elevation model (DEM) and a digital surface model with centimeter-level precision to accurately represent the elevation undulation, slope changes, and three-dimensional distribution of obstacles (such as field ridges, rocks, and ditches) of the terrain.

[0020] In this embodiment, the acquisition of agricultural machinery operation data relies on the on-board sensor network constructed by the on-board sensors of the agricultural machinery; the wheel steering angle is calculated by measuring the piston rod displacement of the electric steering cylinder; the vehicle attitude is sensed in real time by the inertial measurement unit (IMU) installed on the chassis, and the chassis pitch angle and roll angle are output to determine the real-time attitude of the agricultural machinery on the slope; the drive motor current is read from the controller of the four wheel hub drive motors, which is a key indicator reflecting the real-time load and torque demand of the agricultural machinery and thus estimating energy consumption; the remaining battery power comes from the battery management system of the customized lithium iron phosphate battery pack (model YHD768125TPA, 72V / 115Ah), and the remaining battery power is the core boundary condition that constrains the continuous operation time of the agricultural machinery (≥5 hours).

[0021] After the initial data acquisition is completed, the raw data undergoes spatiotemporal registration and alignment to form a unified multidimensional dataset. In the temporal dimension, a unified Global Positioning System (GPSTS) timestamp is applied to all data streams to ensure a strict correspondence between terrain information and agricultural machinery status information at any given time. In the spatial dimension, coordinate transformation is used to unify the point cloud data acquired by the lidar to the geodetic coordinate system or local operational coordinate system used for agricultural machinery navigation, achieving centimeter-level data alignment. For example, the pitch angle measured by the IMU and the slope value measured by the lidar are cross-validated and calibrated to ensure data consistency. Ultimately, each spatial location point not only contains three-dimensional terrain information but also is associated with the agricultural machinery's current steering angle, attitude, motor current, and battery level, forming a multidimensional dataset with strict spatiotemporal correlation.

[0022] Furthermore, feature extraction and semantic annotation are performed on the multidimensional dataset. In the feature extraction stage, key terrain features such as ridgelines, valley lines, continuous steep slopes (e.g., slope > 15°), and relatively flat areas are extracted from the DEM. Simultaneously, agricultural machinery operation data is analyzed to extract features representing energy consumption and operational frequency, such as average current load and steering angle change frequency. In the semantic annotation stage, based on the extracted features, clear semantic labels are assigned to different geographical areas and agricultural machinery states. For example, areas with gentle slopes and no obstacles are labeled as "high-efficiency operating areas"; low-lying areas with drastic chassis pitch angle changes and a risk of bottoming out are labeled as "careful passage areas" or "avoidance areas"; and long uphill sections with persistently high drive motor current and high energy consumption are labeled as "high-energy-consumption areas." After feature extraction and semantic annotation, the multidimensional dataset is transformed into an environmental equipment collaborative dataset rich in decision-making information.

[0023] In an optional embodiment, the environmental equipment collaborative dataset can explicitly indicate whether the drive motor current of a seeding and protection machine with a chassis height of 600mm and a weight of 350kg will exceed the limit when turning at a specific steering angle on a particular slope, and whether the battery power can support it to complete subsequent operations. The environmental equipment collaborative dataset not only accurately describes the terrain undulations and obstacles in the field spatially, but also correlates the mechanical characteristics, real-time status, and energy consumption level of the agricultural machinery itself in the temporal dimension. This deep collaboration between environmental information and agricultural machinery status ensures that agricultural machinery path planning is not based on the geometric lines of an idealized two-dimensional map, but rather on a practically feasible operational plan generated based on real constraints of the physical world, guaranteeing the feasibility, safety, and economy of the planned path for a specific agricultural machine on a specific terrain.

[0024] S2. Parse the environmental equipment collaborative dataset to obtain the field operation segmentation area and field operation priority as the agricultural machinery operation parsing result.

[0025] Specifically, S2 includes the following steps: S21. Based on the environmental equipment collaborative dataset, a dynamic clustering algorithm is used to divide continuous areas with similar terrain features into the same sub-region for field operations as the field operation segmentation area.

[0026] In this embodiment, after obtaining the deeply fused environmental equipment collaborative dataset, the work area is scientifically divided to obtain field work segmentation areas in order to adapt to the fragmented and diverse characteristics of hilly and mountainous terrain. Dynamic clustering algorithms (such as Density-Based Spatial Clustering of Applications with Noise (DBSCAN)) are used to parse the dataset to achieve automatic and reasonable division of field work segmentation areas.

[0027] Specifically, the input feature vector of the dynamic clustering algorithm consists of key terrain and accessibility features in the environmental equipment collaborative dataset, mainly including: elevation value, slope value, and obstacle identifier for each grid cell or point cloud partition. The clustering process integrates multi-dimensional terrain features for similarity measurement. For example, continuous plots with gentle slopes (e.g., less than 15 degrees, referencing the unloaded climbing ability of agricultural machinery), continuous elevation, and no large obstacles are automatically grouped into the same "efficient operation sub-region"; conversely, areas with abrupt slopes, deep ditches, or raised field ridges (whose height difference may exceed the vertical obstacle-crossing ability of agricultural machinery by 200mm-300mm or threaten the chassis height of 600mm) are identified as boundaries or independent regions. By dynamically adjusting the neighborhood radius and minimum sample number parameters of the clustering algorithm, subtle changes in terrain features are captured, ensuring that the segmentation results reflect both the macroscopic continuity of the terrain and accurately identify microscopic obstacle zones unsuitable for continuous operation. Ultimately, the output divides the entire farmland operation area into several continuous and accessible sub-areas with similar internal terrain features as field operation segmentation areas. These sub-areas form the basic units for subsequent path planning, laying the foundation for adopting differentiated operation strategies on plots with different characteristics. This avoids the problem that planning a single continuous path on complex terrain may lead to obstructed movement or low efficiency.

[0028] S22. Construct a multi-factor priority evaluation function to evaluate the field operation segmentation area and obtain the field operation priority.

[0029] First, priority assessment factors for each field operation sub-region are acquired and quantified: the soil suitability index is obtained through a network of soil sensors deployed in the field or through pre-sampling data, and quantified indicators include soil moisture, compaction, and pH value; for example, in the sowing mode, areas with moderate moisture and loose soil will receive a higher index score. The crop growth status index relies on multispectral cameras mounted on agricultural machinery or drones, and assesses crop growth and health by calculating remote sensing indicators such as the normalized vegetation index; for sowing and planting protection machines, in the spraying mode, areas with missing leaves or abnormal leaf color (indicating pests and diseases) will have a significantly lower index, indicating higher operational needs. The agricultural machinery-region distance is a dynamic factor, referring to the geometric distance from the current real-time location of the agricultural machinery or the preset starting point to the center of gravity of each operation sub-region. This factor is related to the non-operational energy consumption of the agricultural machinery, and for agricultural machinery with fixed endurance, it is a key indicator affecting the overall task completion rate. The urgency level of the operation is preset and dynamically adjusted by the agricultural expert system according to agronomic requirements. For example, for areas with sudden outbreaks of pests, the urgency level of the spraying operation will be set to the highest level.

[0030] Secondly, since the core objectives of different operation modes vary, the importance of priority evaluation factors is not constant, thus requiring the determination of dynamic weight coefficients. When constructing a multi-factor priority evaluation function, a dynamic weight coefficient based on the analytic hierarchy process (AHP) or entropy weighting method is used to assign weight coefficients to priority evaluation factors that change with time and task. For example, in the sowing mode, the soil suitability index is given the highest weight to ensure the optimal environment for seed germination; the crop growth status index and operation urgency have relatively lower weights. In the spraying mode, the crop growth status index and operation urgency are prioritized and their weights are increased; the weight of soil suitability is reduced. The weight of the distance between the agricultural machinery and the area is usually maintained at a basic value across all modes to control energy consumption, but when the agricultural machinery's battery charge is low, its weight can be adaptively increased to prioritize nearby operations and ensure the safe return of the agricultural machinery for charging.

[0031] Based on soil suitability index, crop growth status index, agricultural machinery-region distance, and operational urgency and their dynamic weighting coefficients, a multi-factor priority evaluation function is constructed in conjunction with the remaining battery power of agricultural machinery, satisfying the following relationship: in, For comprehensive priority scoring, , For dynamic weighting coefficients, Soil suitability index This is a crop growth status index. For agricultural machinery - regional distance, To assess the urgency of the task, This refers to the remaining battery power of agricultural machinery.

[0032] Then, after determining the quantified values ​​and corresponding dynamic weights of each factor, a comprehensive priority score for each field operation segment is calculated using a multi-factor priority evaluation function. Before calculation, all factors need to be normalized to eliminate the influence of dimensions and ensure comparability within the same numerical range. For example, the original values ​​of factors for each region are mapped to the interval [0,1], where 1 represents the optimal or most urgent state. Subsequently, the normalized factor values ​​are substituted into the aforementioned multi-factor priority evaluation function to calculate the comprehensive priority score. The comprehensive priority score objectively reflects the global value and necessity of prioritizing operations in that region under a specific operational mode.

[0033] Finally, based on the calculated comprehensive priority score, all field operation sub-regions are ranked from highest to lowest. The region with the highest score is considered the most priority target region under the current operation mode; the ranking result constitutes the final field operation priority. For example, a possible output sequence is: Region C (pesticide application, high urgency), Region A (sowing, optimal soil), Region B (fertilization, conventional). The field operation priority guides global path planning, determining the order in which agricultural machinery handles each field operation sub-region. In this way, it ensures that limited operation time (especially under battery range constraints) and resources are prioritized for the most critical, urgent, and high-efficiency operation segments, thereby improving the overall intelligence level and comprehensive economic benefits of agricultural machinery operations in hilly and mountainous areas.

[0034] S3. Based on the analysis results of the agricultural machinery operation, establish a four-wheel steering kinematic model and an agricultural machinery passability constraint model as mechanical constraint conditions for agricultural machinery equipment.

[0035] In this embodiment, the field operation segmentation areas to be planned are determined based on the agricultural machinery operation analysis results, and path planning is performed sequentially according to the priority of field operations. The terrain elevation is obtained from the hilly and mountainous terrain data of the field operation segmentation areas. The DEM data of the area is then called, and the terrain elevation (i.e., the degree of terrain undulation) within the area is accurately quantified by calculating the elevation extreme differences between grid cells. This identifies whether there are gullies with a depth exceeding the safety threshold or terrain uplifts that could cause the chassis to bottom out.

[0036] Furthermore, obtaining the agricultural machinery's attribute parameters determines the physical boundaries of its movement. Based on pre-stored agricultural machinery specifications, the machine's wheelbase, turning radius, adjustable track width, and chassis ground clearance are obtained.

[0037] One of the key features of the new agricultural machinery is the dynamically adjustable wheel track and chassis height, which is beneficial for subsequent path optimization.

[0038] It should be noted that the specifications of agricultural machinery (such as the wheelbase, track width, minimum turning radius of less than 1.5 meters, chassis ground clearance of 600mm, and gradeability of 28 degrees (unloaded) for seeders and planters) are inherent design values ​​of agricultural machinery and are pre-stored in the agricultural machinery model database.

[0039] A gradeability of 28 degrees (unloaded) refers to the maximum slope that an agricultural machine can safely and stably climb without carrying any work load (such as no pesticides, seeds, etc.). It is an important indicator for measuring the ability of agricultural machines to travel on slopes and is used to assess the passability of agricultural machines in complex hilly and mountainous terrain.

[0040] In this embodiment, a four-wheel steering kinematic model is established based on Ackermann steering geometry constraints, combined with the machine's wheelbase, track width, and turning radius. The core of Ackermann steering geometry constraints is to ensure that when the vehicle turns, the extensions of all wheels intersect at the same point—the Instantaneous Center of Rotation (ICR), thereby achieving pure rolling and avoiding tire slippage. For seeders and planters with independent four-wheel steering, this model uses the machine's wheelbase and track width as fixed dimensions and the minimum turning radius as the performance boundary to establish mathematical relationships. Through the four-wheel steering kinematic model, the theoretical steering angles required for each of the four wheels of the machine under any given path curvature can be calculated in reverse. Furthermore, it can be determined in advance whether the radius of curvature of connecting turning paths (such as field turns) in the planned path is less than the minimum turning radius (<1.5 meters) achievable by the machine. If the planned path involves a sharp turn, the model will immediately mark it as an infeasible path segment, thus ensuring that every planned instruction is within the mechanical capabilities of the agricultural machinery and guaranteeing that it can smoothly perform highly maneuverable actions such as turning around on the spot.

[0041] The above four-wheel steering kinematic model satisfies the following relationship: in, This refers to the steering angle of the inner front wheel. It is the arctangent function. For agricultural machinery wheelbase, This refers to the turning radius of the agricultural machinery. For the wheel track of agricultural machinery, This refers to the steering angle of the outer front wheel. This refers to the steering angle of the inner rear wheel. This is the steering angle of the outer rear wheel.

[0042] In this embodiment, to ensure the safe passage of agricultural machinery, a safety margin (such as 50mm) is introduced to cope with dynamic uncertainties such as soft ground and vehicle vibration. The passability constraint model of agricultural machinery is constructed by combining the terrain height extracted from DEM and the chassis ground clearance (600mm) retrieved from the agricultural machinery model database.

[0043] Specifically, the agricultural machinery passability constraint model uses a piecewise function to quantify the passability of any path point. The function output is a passability score between 0 and 1. When the terrain height at a point is not higher than the safety threshold (the difference between the chassis height and the safety margin), it is considered absolutely safe, and the passability score is 1. When the terrain height is between the safety threshold and the theoretical limit (i.e., between 550mm and 600mm), the passability score decreases linearly from 1 to 0, indicating that the risk gradually increases and the machine is in a "critical passable" state. This linear transition zone provides gradient information for the path optimization algorithm. When the terrain height reaches or exceeds the chassis height of the agricultural machinery (600mm), it is determined that "bottoming out" is inevitable, the passability score is 0, and the path is not feasible at this point.

[0044] The implementation process of the agricultural machinery passability constraint model in path planning is as follows: candidate paths are densely and discretely sampled, and their passability scores are calculated point by point. The final passability constraint for the entire path is that no point with a passability score of 0 is allowed, and the average passability score should be as close to 1 as possible. This model combines rigid mechanical limits with flexible safety assessments, serving as a guiding function that enables path optimization algorithms to proactively avoid high-risk terrain.

[0045] The above agricultural machinery passability constraint model satisfies the following relationship: in, For pass / fail scoring, For terrain height, This refers to the chassis height from the ground. For safety margin, This indicates a conditional statement.

[0046] S4. Generate an initial agricultural machinery operation path by combining the analysis results of the agricultural machinery operation and the mechanical constraints of the agricultural machinery equipment.

[0047] Specifically, S4 includes the following steps: S41. Determine the global operation sequence of the field operation segmentation area according to the field operation priority.

[0048] In this embodiment, the order of priority for field operations in the divided field operation areas is the global operation sequence.

[0049] S42. Generate an internal operation path based on the boundary geometric features of the field operation segmentation area, including the multi-machine collaborative task decomposition area, operation direction, operation strip, and connecting turning path.

[0050] In this embodiment, after determining a single field operation segmentation area, firstly, multi-agent autonomous collaboration and dynamic decomposition of regional tasks are performed on multiple agricultural machines to obtain a multi-machine collaborative task decomposition area. Secondly, geometric analysis is performed on the boundary polygon of the decomposition area to determine the operation travel direction. The minimum bounding rectangle of the boundary polygon is obtained, and the operation travel direction is set to be parallel to the direction of the long side of the minimum bounding rectangle, effectively reducing the number of turns during the operation. For slopes where operations are carried out along contour lines, the travel direction is set to be parallel to the contour lines to improve operational stability and safety. Next, based on the travel direction, a set of parallel lines covering the entire operation area are generated at intervals of agricultural machine operation width; these parallel lines are the operation strips. Then, connecting turning paths are planned for each operation strip, and specific turning areas are generated at both ends of the strip. Combining the turning characteristics of the agricultural machines, efficient turning paths, such as pear-shaped turns or fishtail turns, are designed within the turning areas. Finally, all the operation strips within the area and their connecting turning paths at both ends together constitute the complete internal operation path of the area.

[0051] S43. Based on the mechanical constraints of the agricultural machinery equipment, the internal operation path of the area is verified to obtain the optimized internal operation path.

[0052] After obtaining the operational path within the area, the connecting steering paths are validated using a four-wheel steering kinematic model. First, the geometric features of each turning segment are extracted, and its radius of curvature is calculated. Then, using the radius of curvature as input, the theoretical steering angles of the four wheels required for the agricultural machinery to complete the turning process of the connecting steering path are obtained based on the four-wheel steering kinematic model established according to Ackerman steering geometric constraints. This is then used to verify whether the radius of curvature of the turning path is greater than or equal to the inherent minimum turning radius of the agricultural machinery. If the calculation results show that the radius of curvature of the path is too small, i.e., the required steering angle exceeds the physical limit of the electric steering cylinder, then that turning path segment is marked as "infeasible".

[0053] Furthermore, all path points within the operational route within the area are traversed. The elevation changes and crop growth status of each path point are verified using an agricultural machinery passability constraint model. The locations and risk levels of all risk points that do not meet passability requirements are recorded. Operational strips with risk points are marked as "unqualified path segments" requiring further optimization. This verification process ensures that all path points on the operational strip are within the safe passage range of the agricultural machinery chassis, providing a safety guarantee for generating feasible operational routes.

[0054] In this embodiment, for the path segments that do not meet the mechanical constraints of agricultural machinery equipment in the above verification (i.e., path segments with too small turning radius or risk of bottoming out), the path smoothness and operation continuity are used as optimization indicators, and the sequential quadratic programming algorithm is used to perform local path correction to obtain the optimized path segments.

[0055] Specifically, in this scenario, the sequential quadratic programming algorithm uses unqualified path segments as initial solutions, the minimum turning radius of the agricultural machinery and the minimum safe ground clearance as constraints, and uses the total path length, the smoothness of curvature changes, and the continuity with the original qualified path as optimization objectives. It iteratively solves the problem to generate optimized path segments that meet the mechanical constraints. For example, a sharp bend can be planned as a gentle curve, or path points that dynamically adjust the wheelbase and ground clearance of the agricultural machinery to traverse complex terrain can be replanned.

[0056] In an optional embodiment, the optimization objective of the optimized path segment is quantified, satisfying the following relationship: in, To optimize the target quantization value, These are the weighting coefficients. The length of the path segment. The number of path points. For the index variable of the path point, Let be the curvature of the path point.

[0057] Furthermore, all optimized path segments generated by the sequential quadratic programming algorithm are seamlessly integrated with the original internal operation paths that meet the constraints. The integration process includes: geometric splicing to ensure the continuity of position and direction at connection points; and logical association to update the topological relationship of the entire path. After integration, the final optimized internal operation path is output, which is the optimal internal operation trajectory obtained within the operation sub-region under the premise of strictly adhering to the physical limits of agricultural machinery.

[0058] S44. Based on the global job sequence, connect the internal job optimization paths to generate the initial agricultural machinery job path.

[0059] In this embodiment, starting from the initial position of the agricultural machinery, a transfer path is first planned to reach the entrance point of the first work area in the global work sequence, including but not limited to the direction and speed of travel. Then, the optimized internal work path of that area is completely added to the total path. Next, an inter-area transfer path is planned from the exit point of the first area to the entrance point of the second area, and the optimized internal work path of the second area is added to the total path. This process continues until all areas in the global work sequence have been traversed. During the connection process, all inter-area transfer paths must comply with the mechanical constraints of the agricultural machinery, i.e., verified using a four-wheel steering kinematics model and an agricultural machinery passability constraint model, ensuring that the agricultural machinery meets the requirements for turning radius and chassis passability while traveling from one area to another. Finally, these independent transfer paths, internal work paths, and inter-area transfer paths are connected end-to-end in the order of the global work sequence to form a continuous and complete initial agricultural machinery work path covering the entire farmland work area from start to finish.

[0060] S5. Based on the agricultural machinery operation mode, the initial agricultural machinery operation path is adjusted in a multi-objective adaptive manner to obtain the optimal agricultural machinery operation path.

[0061] Specifically, S5 includes the following steps: S51. Obtain the agricultural machinery operation mode, including sowing mode, spraying mode and fertilization mode, and determine the operation parameter configuration set corresponding to the agricultural machinery operation mode.

[0062] In this embodiment, the agricultural machinery operation mode to be executed is accurately obtained by parsing the user-inputted task instructions or receiving the mode selection signal sent by the remote control terminal. After determining the operation mode, the corresponding operation parameter configuration set is called from the pre-stored operation mode parameter library: the operation parameter configuration set for the sowing mode includes key parameters such as the operating speed range (0.5m / s-1.2m / s), sowing width (1000mm-1100mm), number of sowing rows (2-8 rows), and seed metering speed based on the characteristics of the seeder; the operation parameter configuration set for the spraying mode includes parameters such as spray flow rate (10L / min-25L / min), spray boom width (6.5m), spray pressure, and tank capacity (200L); the operation parameter configuration set for the fertilization mode includes parameters such as the corresponding fertilizer application amount, fertilization depth, and operating speed.

[0063] It should be noted that the battery life of a single new energy vehicle for agricultural machinery is ≥5 hours.

[0064] S52. Construct a multi-objective optimization function for the agricultural machinery operation mode based on the set of operation parameters.

[0065] After obtaining the set of operational parameters, key operational parameters for each agricultural machinery operation mode are extracted based on this set. This is achieved through a parameter parsing engine, which selects core parameters that directly impact path optimization from the parameter configuration set according to different operation modes. For example, in the seeding mode, extracted key operational parameters include the optimal operating speed range, seeder width, and row spacing configuration; in the spraying mode, parameters such as effective spray width, droplet coverage density requirements, and pesticide flow rate range are extracted. By establishing a parameter correlation matrix, it is ensured that the extracted key operational parameters can support subsequent optimization target modeling.

[0066] It should be noted that key operational parameters have two important characteristics: First, they are directly related to path planning, such as the operation speed affecting the distribution density of path points and the machine width determining the division of operation strips; second, they are closely related to optimization objectives, such as the speed parameter affecting both operation efficiency and operation quality.

[0067] In this embodiment, a multi-dimensional optimization objective is constructed: the operation quality objective is quantified by indicators such as path tracking accuracy and operation uniformity; the energy consumption objective is based on an energy consumption model established by parameters such as motor current and travel distance; and the operation efficiency objective is measured by the operation area per unit time.

[0068] Furthermore, the balance adjustment coefficients for each objective are determined based on key operational parameters: in sowing mode, the operational quality adjustment coefficient is set higher to ensure uniform sowing and consistent depth; in spraying mode, the operational efficiency adjustment coefficient is weighted higher to quickly complete pest and disease control; when battery power is low, the energy consumption adjustment coefficient automatically increases to prioritize the completion of operational tasks. The allocation of balance adjustment coefficients is achieved through fuzzy reasoning, which can adaptively adjust the relative importance of each objective according to real-time operational conditions and agronomic requirements.

[0069] In this embodiment, a linear weighted sum method is used to fuse the balance adjustment coefficients and multi-dimensional optimization objectives to construct a multi-objective optimization function. First, the quantified data of each sub-objective are normalized to eliminate dimensional differences and ensure comparability between different objectives. Then, dynamically determined balance adjustment coefficients are assigned to the multi-dimensional optimization objectives to obtain the multi-objective optimization function. This function retains the independent characteristics of each sub-objective while reflecting the optimization focus under different operating modes through the balance adjustment coefficients, achieving the optimal balance of the multi-dimensional optimization objectives.

[0070] The above multi-objective optimization function satisfies the following relationship: in, For a multi-objective optimization function, To balance the adjustment parameters, For the quality of the work, For energy consumption, For work efficiency.

[0071] S53. Based on the environmental equipment collaborative dataset, the path parameters of the initial agricultural machinery operation path are iteratively optimized using an adaptive weighted particle swarm optimization algorithm to solve the multi-objective optimization function and obtain the optimal agricultural machinery operation path.

[0072] In this embodiment, the particle swarm is first initialized, and the key parameters of the initial agricultural machinery operation path are encoded as particle position vectors, including the velocity distribution of path points and the turning radius sequence. Each particle represents a candidate path optimization scheme. During the iteration process, the particles update their velocity and position based on their individual optimal position and the group optimal position. An adaptive weight adjustment strategy is adopted: a larger inertia weight is set in the early stage of iteration to enhance global exploration capability, and the weight is gradually reduced as iterations progress to strengthen local development capability. In each iteration, the fitness value of each path optimization scheme is calculated using a multi-objective optimization function. This fitness value comprehensively reflects the operation quality, energy consumption, and operation efficiency of the path. At the same time, the algorithm ensures that the optimized path always meets the mechanical constraints of the agricultural machinery through a constraint processing mechanism. During the iteration process, the individual optimal and group optimal solutions are dynamically tracked. When the maximum number of iterations is reached or the fitness value converges, the path parameters corresponding to the particle with the optimal fitness value are output. After decoding, the optimal agricultural machinery operation path is obtained. The adaptive weighted particle swarm optimization algorithm ensures that the optimal agricultural machinery operation path can achieve the best balance between operation quality, energy consumption, and operation efficiency in complex hilly and mountainous environments.

[0073] The adaptive weighted particle swarm optimization algorithm iteratively optimizes the path parameters of the initial agricultural machinery operation path, satisfying the following relationship: in, For particles In the The velocity vector of the generation, For adaptive inertia weights, For particles In the The velocity vector of the generation, The acceleration constant, These are random numbers that are uniformly distributed within the interval. For particles The individual's historical best position, For particles In the The position vector of the substitute, This is the best position in the global history. For particles In the The position vector of the substitute, This represents the maximum value of the adaptive inertia weight. This represents the minimum value of the adaptive inertia weight. This represents the current iteration number. This represents the maximum number of iterations.

[0074] Please see Figure 2 In an optional embodiment, the present invention provides a field operation path planning system for agricultural machinery in hilly and mountainous areas. The system includes an input device, an output device, a processor, and a memory, all interconnected. The memory stores a computer program comprising program instructions. The processor is configured to invoke the program instructions to execute specific steps as described in the relevant embodiments of the field operation path planning method for agricultural machinery in hilly and mountainous areas provided by the present invention. The field operation path planning system for agricultural machinery in hilly and mountainous areas provided by the present invention has a complete and stable structure, enhancing the overall applicability and practical application capability of the present invention.

[0075] In summary, this invention provides a method and system for planning field operation paths for agricultural machinery in hilly and mountainous areas. It constructs a collaborative dataset by integrating terrain data and agricultural machinery operation data, divides the operation area based on a dynamic clustering algorithm and evaluates priorities, establishes a four-wheel steering kinematic model and a passability constraint model as mechanical constraints, and generates an initial operation path including operation strips and connecting steering paths based on the characteristics of the operation area. According to different operation modes, an adaptive weighted particle swarm optimization algorithm is used for multi-objective optimization to obtain the optimal path that meets the requirements of hilly and mountainous operations. This invention is easy to understand, computationally simple, requires less work, and is convenient for engineering applications, providing a theoretical foundation and technical support for the further development of agricultural machinery operation technology.

[0076] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A method for planning field operation paths for agricultural machinery in hilly and mountainous areas, characterized in that, Includes the following steps: Acquire hilly and mountainous terrain data and agricultural machinery operation data to construct an environmental equipment collaborative dataset; The environmental equipment collaborative dataset is parsed to obtain the field operation segmentation area and field operation priority as the agricultural machinery operation parsing result; Based on the analysis results of agricultural machinery operations, a four-wheel steering kinematic model and an agricultural machinery passability constraint model are established as mechanical constraint conditions for agricultural machinery equipment. The initial agricultural machinery operation path is generated by combining the analysis results of the agricultural machinery operation and the mechanical constraints of the agricultural machinery equipment. The optimal agricultural machinery operation path is obtained by performing multi-objective adaptive adjustment on the initial agricultural machinery operation path according to the agricultural machinery operation mode.

2. The method for planning field operation paths for agricultural machinery in hilly and mountainous areas according to claim 1, characterized in that, The acquisition of hilly and mountainous terrain data and agricultural machinery operation data to construct an environmental equipment collaborative dataset includes: The terrain data of the hilly and mountainous areas is obtained by lidar, and the operation data of the agricultural machinery equipment is collected based on the on-board sensors of the agricultural machinery. The hilly terrain data and the agricultural machinery operation data are spatiotemporally registered and aligned to form a multidimensional data set. Feature extraction and semantic annotation are performed on the multidimensional dataset to generate the environmental device collaborative dataset.

3. The method for planning field operation paths for agricultural machinery in hilly and mountainous areas according to claim 1, characterized in that, The step of parsing the environmental equipment collaborative dataset to obtain field operation segmentation regions and field operation priorities as agricultural machinery operation parsing results includes: Based on the aforementioned environmental equipment collaborative dataset, a dynamic clustering algorithm is used to divide continuous areas with similar terrain features into the same operational sub-region as the field operation segmentation region. A multi-factor priority evaluation function is constructed to evaluate the field operation segmentation area and obtain the field operation priority.

4. The method for planning field operation paths for agricultural machinery in hilly and mountainous areas according to claim 3, characterized in that, The construction of a multi-factor priority evaluation function to evaluate the field operation segmentation areas and obtain the field operation priority includes: Obtain priority assessment factors for the field operation segmentation areas, including soil suitability index, crop growth status index, agricultural machinery-area distance, and operation urgency. Determine the dynamic weight coefficients of the priority evaluation factors, and combine them with the remaining power of agricultural machinery to construct the multi-factor priority evaluation function; The comprehensive priority score of the field operation segmentation area is obtained based on the multi-factor priority evaluation function. The field operation priority is obtained by sorting the field operation division areas according to the comprehensive priority score.

5. The method for planning field operation paths for agricultural machinery in hilly and mountainous areas according to claim 1, characterized in that, Based on the analysis results of the agricultural machinery operation, a four-wheel steering kinematic model and an agricultural machinery passability constraint model are established as mechanical constraint conditions for agricultural machinery equipment, including: Based on the analysis results of agricultural machinery operations, the field operation segmentation area to be planned is determined, and the terrain height is obtained from the hilly and mountainous terrain data of the field operation segmentation area; Obtain agricultural machinery attribute parameters, including the agricultural machinery wheelbase, agricultural machinery turning radius, adjustable agricultural machinery track, and agricultural machinery chassis ground clearance; Based on the Ackermann steering geometry constraints, the four-wheel steering kinematic model is established by combining the wheelbase, track width, and turning radius of the agricultural machinery. Based on the terrain height and the ground clearance of the agricultural machinery chassis, a passability constraint analysis is performed to construct the agricultural machinery passability constraint model.

6. The method for planning field operation paths for agricultural machinery in hilly and mountainous areas according to claim 1, characterized in that, The process of generating an initial agricultural machinery operation path by combining the agricultural machinery operation analysis results and the mechanical constraints of the agricultural machinery equipment includes: The global operation sequence for the field operation segmentation area is determined based on the field operation priority. Based on the boundary geometric features of the field operation segmentation area, an internal operation path is generated, including the multi-machine collaborative task decomposition area, operation direction, operation strip, and connecting turning path; Based on the mechanical constraints of the agricultural machinery equipment, the internal operation path of the area is verified to obtain the optimized internal operation path. Based on the global job sequence, the internal job optimization paths are connected to generate the initial agricultural machinery job path.

7. The method for planning field operation paths for agricultural machinery in hilly and mountainous areas according to claim 6, characterized in that, The process of verifying the internal operation path based on the mechanical constraints of the agricultural machinery equipment to obtain the optimized internal operation path includes: The curvature radius of the connecting steering path is verified using the four-wheel steering kinematic model to determine whether it meets the minimum turning radius constraint of agricultural machinery. Based on the aforementioned agricultural machinery passability constraint model, verify whether the elevation changes of the path points in the operating strip and the crop growth status are within the safe pass range of the agricultural machinery chassis; For path segments that do not meet the mechanical constraints of the agricultural machinery, the path smoothness and operation continuity are used as optimization indicators, and the sequential quadratic programming algorithm is used to obtain the optimized path segments that meet the mechanical constraints of the agricultural machinery. The optimized path segment and the internal operation path of the region are integrated to obtain the internal operation optimized path.

8. The method for planning field operation paths for agricultural machinery in hilly and mountainous areas according to claim 1, characterized in that, The step of performing multi-objective adaptive adjustment of the initial agricultural machinery operation path according to the agricultural machinery operation mode to obtain the optimal agricultural machinery operation path includes: The agricultural machinery operation modes are obtained, including sowing mode, spraying mode and fertilization mode, and the operation parameter configuration set corresponding to the agricultural machinery operation modes is determined; Construct a multi-objective optimization function for the agricultural machinery operation mode based on the set of operation parameter configurations; Based on the aforementioned environmental equipment collaborative dataset, an adaptive weighted particle swarm optimization algorithm is used to iteratively optimize the path parameters of the initial agricultural machinery operation path in order to solve the multi-objective optimization function and obtain the optimal agricultural machinery operation path.

9. The method for planning field operation paths for agricultural machinery in hilly and mountainous areas according to claim 8, characterized in that, The construction of the multi-objective optimization function for the agricultural machinery operation mode based on the set of operation parameters includes: Key operational parameters of the agricultural machinery operation mode are extracted based on the operational parameter configuration set; Establish multidimensional optimization objectives, including work quality, energy consumption, and work efficiency, and determine the balance adjustment coefficients of the multidimensional optimization objectives based on the key work parameters. The multi-dimensional optimization objectives are fused according to the balance adjustment coefficient to construct the multi-objective optimization function.

10. A field operation path planning system for agricultural machinery in hilly and mountainous areas, characterized in that, The system includes an input device, an output device, a processor, and a memory, which are interconnected. The memory stores a computer program, which includes program instructions. The processor is configured to invoke the program instructions to execute the field operation path planning method for agricultural machinery in hilly and mountainous areas as described in any one of claims 1-9.

Images