Agricultural machinery operation path planning method and system based on regional division

Abstract

The present invention relates to a method and system for planning agricultural machinery operation paths based on regional division, comprising the following steps: obtaining the size of an agricultural machinery operation area and obstacles within the area to form environmental map information; using the agricultural machinery location as the starting coordinate and the warehouse location as the ending coordinate, and performing environmental modeling based on the obtained environmental map information; obtaining an initial feasible path group from the starting coordinate to the ending coordinate, treating obstacles in the environmental map information as circular obstacles, using the coordinates of the circle centers as connecting nodes for the boundaries of each sub-area in regional division, extending to the next node using the ending coordinate as the initial node, connecting the initial node with each extended node to implement regional division, obtaining multiple groups of sub-areas, performing path search within the sub-areas, obtaining the optimal path within the sub-areas, and obtaining the final optimal feasible path after updating.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural equipment, and in particular to a method and system for planning agricultural machinery operation paths based on regional division. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Modern agriculture has begun to integrate unmanned driving technology into the use of agricultural machinery to realize autonomous navigation functions. Unmanned operations have been realized in operations such as plowing and sowing. However, after the crops are harvested, the work of transporting them back to the warehouse still needs to be completed with the help of manually driven agricultural machinery.

[0004] The basis for realizing unmanned autonomous navigation of agricultural machinery is to ensure that it can achieve accurate path planning. This is the core issue for its ability to perform precise operations. When encountering obstacles during operation, the obstacle avoidance problem during transportation must also be considered when planning the path. How to plan an effective path in an area with obstacles to achieve minimum energy consumption of agricultural machinery.

[0005] To address the above issues, traditional path planning methods can find a feasible path from the starting point to the target point in a short period of time during path planning. However, redundant nodes will be generated during the planning process, resulting in an increase in the length of the feasible path and the time of path planning, which will indirectly lead to an increase in the working cost of agricultural machinery operations. Some intelligent algorithms are also often used in path planning algorithm research, such as genetic algorithms, ant colony algorithms and whale algorithms. Although they can solve the globally optimal feasible path, they have weak performance when performing local path searches and are prone to premature convergence, resulting in reduced accuracy of the final feasible path. Summary of the Invention

[0006] To address the technical issues presented in the aforementioned background technology, the present invention provides a method and system for agricultural machinery operation path planning based on regional division. Applied during the post-harvest crop transport back to the warehouse, the method uses a particle swarm algorithm to generate a set of initial feasible paths. Taking into account the characteristics of obstacles in the environmental map and the number of intersections between the initial path and the centerline, the regional division method decomposes the environmental map into multiple sub-regions, narrowing the search range for feasible paths and reducing search complexity. This avoids the local optimal solution problem that is prone to occur during global search, and improves the efficiency of the next stage of refined search. During the refined search stage, fractional calculus is introduced into a hybrid algorithm combining particle swarm and gravitational search to enhance the algorithm's global convergence. Energy consumption constraints are also considered. When obstacles are encountered, dynamic parameters are introduced to adjust the search step size, allowing for the planning of a route with the lowest possible energy consumption, based on the energy consumption constraints.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] A first aspect of the present invention provides an agricultural machinery operation path planning method based on region division, comprising the following steps:

[0009] Obtaining the size of the agricultural machinery operation area and the environmental map information formed by obstacles within the area;

[0010] The agricultural machinery location is used as the starting coordinate, the warehouse location is used as the ending coordinate, and environmental modeling is performed based on the acquired environmental map information;

[0011] Obtain the initial feasible path group from the starting coordinates to the end coordinates, regard the obstacles in the environmental map information as circular obstacles, use the coordinates of the circle center as the connection node of the sub-region boundary in the region division, use the end coordinates as the initial node to extend to the next node, connect the initial node with each extended node to realize region division, obtain multiple groups of sub-regions and perform path search in the sub-regions to obtain the optimal path in the sub-region. After updating, obtain the final optimal feasible path.

[0012] Implement regional division, specifically:

[0013] The line connecting the centers of each obstacle in the environment map is defined as the center line. The location of the end point coordinate is used as the parent node. The obstacles at both ends of the center line with the most intersections with the feasible path group are used as child nodes of the parent node. The child nodes serve as the parent nodes of the next level nodes and extend until they reach the final node, which is the starting coordinate.

[0014] Except for the starting coordinate point and the end coordinate point, each node can only be a child node of a parent node; each node must be traversed, that is, the decomposed sub-region space is a simply connected region.

[0015] When the child node of the parent node happens to be the starting coordinate point, but the connection line is blocked by other connected nodes, the boundary line of the operation area is used as the connection line between the nodes to form a new sub-area.

[0016] With the set radius around the obstacle as a reference, a safety zone is formed around the obstacle, and each safety zone has a set energy consumption weight.

[0017] The initial feasible path group from the starting coordinate to the end coordinate is obtained by the standard particle swarm algorithm.

[0018] Perform path search in the sub-area to obtain the optimal path in the sub-area, and obtain the final required optimal feasible path through comparison, specifically:

[0019] The optimal path within the sub-region is obtained through the particle swarm algorithm;

[0020] Use fractional calculus to update the velocity term of the particle swarm algorithm;

[0021] In the position update item, the position information of the particle swarm algorithm is updated by setting the energy consumption weight;

[0022] By updating the speed and position, the optimal path in the sub-area is obtained, and when the requirements or the maximum iteration are met, the final optimal feasible path is obtained.

[0023] A second aspect of the present invention provides a system for implementing the above method, comprising:

[0024] The information collection module is configured to: obtain the size of the agricultural machinery operation area and the environmental map information formed by obstacles in the area;

[0025] The information processing module is configured to: use the location of the agricultural machinery as the starting coordinate and the location of the warehouse as the ending coordinate, and perform environmental modeling based on the acquired environmental map information;

[0026] The path planning module is configured to: obtain an initial feasible path group from the starting coordinates to the end coordinates, regard the obstacles in the environmental map information as circular obstacles, use the coordinates of the center of the circle as the connection node of the boundary of each sub-area in the area division, extend to the next node with the end coordinate as the initial node, connect the initial node with each extended node to realize area division, obtain multiple groups of sub-areas and perform path search in the sub-areas to obtain the optimal path in the sub-area, and obtain the final optimal feasible path after updating.

[0027] A third aspect of the present invention provides a computer device.

[0028] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the agricultural machinery operation path planning method based on area division as described above are implemented.

[0029] Compared with the existing technology, one or more of the above technical solutions have the following beneficial effects:

[0030] 1. Through regional division, the environmental area for path planning is divided into several sub-areas. By performing a refined search on each sub-area, the optimal path within the sub-area is obtained, and the final optimal feasible path is obtained through comparison. This can avoid the problem of local optimality in the algorithm, while reducing the complexity of the search and improving the search efficiency.

[0031] 2. During the region division process, except for the starting and ending coordinates, each node can only be a child node of a parent node to prevent the appearance of closed areas. At the same time, each node must be traversed, that is, the decomposed sub-region space is a single-connected region to avoid the appearance of multi-connected regions and ensure that all obstacles can participate in the region decomposition.

[0032] 3. During the area division process, in order to avoid the emergence of multi-connected areas, the boundary line of the working area itself is also taken into account. When the child node of the parent node happens to be the starting coordinate point, but the connecting line is blocked by other connected nodes, the boundary line of the working area is taken into account to form a complete sub-area.

[0033] 4. When performing a detailed search of sub-regions after region division, combining the particle swarm algorithm with the gravitational search algorithm can effectively improve the search capability of the particle swarm algorithm. At the same time, combined with the memory characteristics of fractional calculus, the particle search speed is updated and the convergence performance of the algorithm is improved. Energy consumption factors are taken into account in the path planning constraints, and the particle search step size is adjusted through dynamic parameters to obtain a feasible path with low energy consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0035] Figure 1 This is a schematic diagram of a process flow for agricultural machinery operation path planning based on area division provided by one or more embodiments of the present invention;

[0036] Figures 2(a)-2(c) These are schematic diagrams of area division in the agricultural machinery path planning process provided by one or more embodiments of the present invention;

[0037] Figure 3(a)-3(b) They are two infeasible states when dividing sub-regions in the agricultural machinery path planning process provided by one or more embodiments of the present invention;

[0038] Figures 4(a)-4(b) They are schematic diagrams of two division methods considering boundary lines in the agricultural machinery path planning process provided by one or more embodiments of the present invention;

[0039] Figure 5 This is a schematic diagram of energy consumption level classification in the agricultural machinery path planning process provided by one or more embodiments of the present invention;

[0040] Figure 6 It is a structural diagram of an agricultural machinery path planning system provided by one or more embodiments of the present invention. DETAILED DESCRIPTION

[0041] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0042] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0043] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0044] As described in the background technology, traditional path planning methods can find a feasible path from the starting point to the target point in a short time during path planning. However, redundant nodes will be generated during the planning process, resulting in an increase in the length of the feasible path and the time of path planning, which will indirectly lead to an increase in the working cost of agricultural machinery operations. Some intelligent algorithms are also often used in path planning algorithm research, such as genetic algorithms, ant colony algorithms and whale algorithms. Although they can solve the globally optimal feasible path, they have weak performance when performing local path searches and are prone to premature convergence, resulting in reduced accuracy of the final feasible path.

[0045] The particle swarm algorithm is widely used in optimization problem research. It has fewer parameters, simple rules, and is easy to implement. When applied to path planning, it has achieved good results. However, its convergence speed is slow in the later stage and it is easy to produce local optimal solutions. Therefore, the following embodiment provides a method and system for agricultural machinery operation path planning based on regional division. According to the particle swarm algorithm, a set of initial feasible paths is obtained. Combined with the characteristics of obstacles in the environmental map and the number of intersections between the initial path and the center line, the environmental map is decomposed into multiple sub-regions through regional division, narrowing the search range of feasible paths, reducing the complexity of the search, avoiding the local optimal solution problem that is easy to produce during global search, and improving the efficiency of the next stage of refined search. In the refined search stage, fractional calculus is introduced into a hybrid algorithm combining particle swarm and gravitational search to enhance the global convergence of the algorithm. At the same time, energy consumption constraints are taken into account. When encountering obstacles, dynamic parameters are introduced to adjust the search step size. Based on the energy consumption constraints, a route with the lowest possible energy consumption can be planned.

[0046] Example 1:

[0047] like Figure 1 As shown, the agricultural machinery operation path planning method based on regional division includes the following steps:

[0048] Step 1: Collect environmental map information (including area size and obstacles within the area) through sensors and image acquisition modules on agricultural machinery;

[0049] Step 2: Convert the collected environmental information into a two-dimensional image and perform environmental modeling. Use the GPS positioning system to determine the location of the agricultural machinery and set it as the starting coordinates, and the end point as the warehouse location.

[0050] Step 3: Obtain a set of initial feasible paths through the particle swarm algorithm and divide the environment map into different sub-regions by combining the region partitioning method;

[0051] Step 4: Particle swarm optimization is combined with the gravitational search algorithm. Fractional-order calculus and energy consumption constraints are introduced in particle updates to form a fractional-order particle swarm gravitational search algorithm to perform path search in the sub-region.

[0052] Step 5: Initialize the information of particle swarm speed, position, mass, etc.

[0053] Step 6: Calculate the fitness value and update the particle acceleration, mass, etc.

[0054] Step 7: Update the position and velocity of the particle swarm;

[0055] Step 8: Iteration ends / the optimal feasible path is found and path planning is completed.

[0056] The above process divides the environmental area for path planning into several sub-areas through regional division, realizing refined search, which can avoid the problem of local optimality in the algorithm, reduce the complexity of the search, and improve the search efficiency; combining the particle swarm algorithm with the gravitational search algorithm can effectively improve the search ability of the particle swarm algorithm, and at the same time, combine the memory characteristics of fractional calculus to update the particle search speed and improve the convergence performance of the algorithm; taking energy consumption factors into account in the path planning constraints, the particle search step size is adjusted through dynamic parameters to obtain a feasible path with low energy consumption.

[0057] Specifically:

[0058] Environmental modeling information collection

[0059] Before conducting path search and planning, it is necessary to obtain environmental information about the entire operating area. To ensure more accurate and realistic environmental information, the first step in environmental modeling is to install a lidar and machine vision system on the unmanned agricultural machinery to form an environmental information acquisition module. Machine vision uses cameras to capture and process environmental images. Through the unmanned agricultural machinery and the collection system, information is collected from the working area to obtain the overall environmental information, including the overall size of the environmental map and some obstacles in the area. With the help of the information processing module, the collected information is converted into a two-dimensional image. The laser radar can realize the ranging function and accurately collect information within the area, including the boundary and obstacles in the area, such as the distance between the obstacle boundary and the area boundary, the obstacle coordinate information, etc. The agricultural machinery is equipped with a laser ranging radar, an image acquisition module and a GPS positioning system module, which can determine the location of the unmanned agricultural machinery and the distance between it and the obstacle in real time, and plan the path. When it reaches the area with obstacles, it determines whether it enters the high-energy consumption area (the area near the obstacle). When the planned path does not meet the current needs, it is re-planned through the path planning module according to the location of the unmanned agricultural machinery to obtain a feasible path that meets the needs, thereby completing the automated transportation operation of the unmanned agricultural machinery.

[0060] Regional optimization decomposition

[0061] 1. Regional decomposition

[0062] For the division of spatial sub-regions, the division method will be different depending on the actual needs. When connecting and dividing sub-regions, if the connection point of the region dividing line is at the geometric vertex of the obstacle, the geometric boundary, or the tangent of the arc, there may be a large difference in the number of obstacles in the sub-region, or even a situation where there are no obstacles in some sub-regions, resulting in a large difference in search performance within the region. In this embodiment, the obstacles in the environment map are set as circular obstacles, and the coordinates of the center of the obstacle circle are regarded as the connection nodes of the region boundary in the region division. Starting from the initial node and extending to the next level node, the spatial region is decomposed by connecting the initial node with each extended node. Such region division can avoid the phenomenon of no obstacles in the region mentioned above.

[0063] like Figure 2(a)-Figure 2(c) The diagram of area division is shown in the figure. The black filled circle represents the obstacle, the square represents the starting position of the unmanned agricultural machinery, and the target position is represented by a five-pointed star.

[0064] In the early stage of region division, a standard particle swarm algorithm is used to obtain a group of feasible paths from the starting point to the target point (as shown by the dotted line in Figure 2(a)). The line connecting the centers of the two obstacles is defined as the center line (as shown by the black solid line in Figure 2(b)). The location of the target point is used as the initial node (defined as the parent node in this embodiment, and the next extended node is defined as the child node). The two obstacles with the center line that intersects the feasible path group the most are used as child nodes of the parent node (such as child node 1 and child node 2) to form a sub-region. These two child nodes serve as the parent nodes of the next group of nodes and are connected to the next node. The starting coordinate point is the final node (as shown by the black arrow in Figure 2(c)).

[0065] Among them, each node can only be a child node of a parent node (except the starting point and the target point), which can prevent the appearance of closed areas (not including the starting point) in the process, as shown in Figure 3(a); at the same time, each node must be traversed, that is, the decomposed sub-region space must be a single-connected region, avoiding the appearance of multiply-connected regions (as shown in Figure 3(b)), ensuring that all obstacles in the graph can participate in the region decomposition.

[0066] 2. Consideration of spatial boundaries

[0067] In the process of region decomposition, in order to avoid the emergence of multi-connected regions, the boundary lines within the spatial region itself are also taken into account. When the child node of the parent node happens to be the starting point, but the connection line is blocked by other connected nodes, the boundary line is taken into account and a new sub-region is formed to avoid this situation. Figure 4(a)-Figure 4(b) As shown in Figure 4(a), considering the division of the boundary line, it is impossible to form a feasible sub-region because the connecting line (black dashed line) has to pass through the obstacle. At this time, the boundary area should be considered, as shown by the black solid line in Figure 4(b), so as to form a complete spatial sub-region.

[0068] After regional division, a more detailed search range can be obtained, the complexity of path search can be reduced, and refined search can be performed. This can also avoid some interference from local optimal positions in global search to a certain extent, and improve search efficiency and capabilities.

[0069] 3. Energy consumption classification

[0070] During the planning and search process for feasible paths, agricultural machinery generally needs to perform obstacle avoidance near obstacles and consider the turning costs of agricultural machinery, etc., which results in higher energy consumption compared to areas without obstacles. To minimize the energy consumption of agricultural machinery, this embodiment sets a safe zone around the obstacle, taking into account the energy loss caused by agricultural machinery avoiding obstacles near the obstacle. By dividing the safe energy consumption levels, energy loss values ​​within different ranges are determined. By incorporating energy consumption factors into the constraints of particle updates, particles can search under multi-factor constraints to obtain better search results. Therefore, this part of the content is mainly defined in two aspects (energy consumption classification takes circular obstacles as an example):

[0071] According to the characteristics of the selected space obstacle, with the radius of the obstacle (r) as a reference, the maximum impact area is based on its radius to 0.5r, that is, the maximum periphery of the safe area. Figure 5 The width of the ring area is 0.2r, 0.3r, and 0.5r, and the area boundary lines are defined as L1, L2, and L3 respectively;

[0072] Regarding the setting of energy consumption weights: in the range (r, L1), the energy consumption weight coefficient is K1; in the range (L1, L2), the energy consumption weight coefficient is K2; (L2, L3), the energy consumption coefficient is K3; the range outside (L3) is normal driving energy consumption; when within the obstacle radius r, obstacle avoidance fails and no feasible path can be formed.

[0073] Energy consumption calculation formula:

[0074]

[0075] Among them, e0 represents the energy consumption coefficient of the non-obstacle area, K i Represents the energy consumption coefficient of each energy consumption division layer around the obstacle, L path Represents the total path length of the agricultural machinery in the non-obstacle area, L ob Indicates the path length within the obstacle area.

[0076] When the obstacle area is other geometric bodies, the geometric center is used as the center, and the distance from the geometric center to the boundary of the obstacle is used as the benchmark reference distance to set the corresponding energy consumption area.

[0077] Fractional-order particle swarm gravitational search algorithm

[0078] After the environmental map is refined through the region division method, in order to further improve the global convergence and search capability of the search, the particle swarm algorithm is combined with the gravitational search algorithm in the refined search stage. By introducing fractional calculus, the velocity term of the particle swarm algorithm is updated to improve the global convergence; the energy consumption problem is introduced in the position update term to dynamically adjust and update the position information of the particle swarm algorithm; by performing refined searches on each sub-region, the optimal path within the sub-region is obtained, and through comparison, the final globally feasible path is obtained.

[0079] The standard particle swarm algorithm speed update and position update are as follows:

[0080]

[0081]

[0082] In the above formula, v i represents the particle velocity, c1 and c2 are particle learning factors, and r1 and r2 represent random numbers between (0,1).

[0083] If we assume that there is an attractive force between particles, the magnitude of which is proportional to the mass of the particle itself and inversely proportional to the distance between the particles, this force can enable particles with good fitness to attract other particles to approach, and search iteratively towards the global optimal position. At this time, the particle gravitational search algorithm is formed, and its speed update formula is shown as follows:

[0084]

[0085] in, Is each with quality The particle acceleration, f t is the fitness, f best With f worst They represent the best and worst fitness of the particle swarm respectively. is the force of other particles on the current particle, ε is a random number, L ij is the Euclidean distance between particles, is the gravitational constant that depends on the constants g and α. In addition, t and T represent the current and maximum number of iterations, respectively.

[0086] By transposing the terms, equation (4) is transformed into the following form:

[0087]

[0088] The inertia weight ω is a random number between 0 and 1. When its value is 1, the left side of Equation (5) can be expanded with the help of fractional calculus, as shown below:

[0089]

[0090] In the above formula, α is a random coefficient. As a method of particle update, a good convergence performance can be obtained by selecting a value between 0 and 1.

[0091] In the above equation, fractional calculus is introduced into the velocity update term, linking the particle velocity update to the history terms of the previous four moments to calculate the particle velocity at the next moment. These history terms help improve global convergence.

[0092] For the improved algorithm, during the search process, it is necessary to appropriately increase or decrease the step size during the search iteration to enable the spatial search to obtain a more accurate solution. At the same time, it is also possible to adaptively adjust the position update parameter during the iteration process. This embodiment introduces a dynamic parameter β to dynamically adjust the particle update search step size, as shown in the following formula:

[0093]

[0094] Among them, E(x id ) is the energy consumption at the current position, E(x id ) avg is the average energy consumption.

[0095] At this time, the particle position update term is:

[0096]

[0097] The above process divides the environmental area for path planning into several sub-areas through regional division, realizing refined search, which can avoid the problem of local optimality in the algorithm, reduce the complexity of the search, and improve the search efficiency; combining the particle swarm algorithm with the gravitational search algorithm can effectively improve the search ability of the particle swarm algorithm, and at the same time, combine the memory characteristics of fractional calculus to update the particle search speed and improve the convergence performance of the algorithm; taking energy consumption factors into account in the path planning constraints, the particle search step size is adjusted through dynamic parameters to obtain a feasible path with low energy consumption.

[0098] Example 2:

[0099] like Figure 6 As shown, the system for implementing the above method includes:

[0100] The information collection module is configured to: obtain the size of the agricultural machinery operation area and the environmental map information formed by obstacles in the area;

[0101] The information processing module is configured to: use the location of the agricultural machinery as the starting coordinate and the location of the warehouse as the ending coordinate, and perform environmental modeling based on the acquired environmental map information;

[0102] The path planning module is configured to: obtain an initial feasible path group from the starting coordinates to the end coordinates, regard the obstacles in the environmental map information as circular obstacles, use the coordinates of the center of the circle as the connection node of the boundary of each sub-area in the area division, extend to the next node with the end coordinate as the initial node, connect the initial node with each extended node to realize area division, obtain multiple groups of sub-areas and perform path search in the sub-areas to obtain the optimal path in the sub-area, and obtain the final optimal feasible path after updating.

[0103] By dividing the environment area for path planning into several sub-areas through regional division, a refined search can be achieved, which can avoid the problem of local optimality in the algorithm, reduce the complexity of the search, and improve the search efficiency.

[0104] Example 3:

[0105] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the agricultural machinery operation path planning method based on area division as described in the first embodiment above are implemented.

[0106] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. The agricultural machinery operation path planning method based on regional division is characterized by: The following steps are involved: Obtaining the size of the agricultural machinery operation area and the environmental map information formed by obstacles within the area; The agricultural machinery location is used as the starting coordinate, the warehouse location is used as the ending coordinate, and environmental modeling is performed based on the acquired environmental map information; Obtain an initial feasible path group from the starting coordinates to the end coordinates. Treat obstacles in the environmental map information as circular obstacles. Use the coordinates of the circle center as the connection node of each sub-region boundary in the region division. Use the end coordinates as the initial node to extend to the next node. Connect the initial node with each extended node to realize region division. Obtain multiple groups of sub-regions and perform path search in the sub-regions to obtain the optimal path in the sub-region. After updating, obtain the final optimal feasible path. To achieve regional division, specifically: define the line connecting the centers of each obstacle in the environmental map information as the center line, take the location of the end point coordinates as the parent node, and take the obstacles at both ends of the center line with the most intersections with the feasible path group as the child nodes of the parent node; the child node serves as the parent node of the next level node and extends until reaching the final node, which is the starting coordinate.

2. The method for agricultural machinery operation path planning based on regional division according to claim 1, characterized in that: When implementing region division, except for the starting coordinate point and the end coordinate point, each node can only be a child node of a parent node; each node must be traversed, that is, the decomposed sub-region space is a simply connected region.

3. The method for agricultural machinery operation path planning based on regional division according to claim 1, characterized in that: When implementing area division, if the child node of the parent node happens to be the starting coordinate point, but the connection line is blocked by other connected nodes, the boundary line of the working area is used as the connection line between the nodes to form a new sub-area.

4. The method for planning agricultural machinery operation paths based on regional division according to claim 1, characterized in that: When implementing area division, a set radius around the obstacle is used as a reference to form a safe area around the obstacle, and each safe area has a set energy consumption weight.

5. The method for planning agricultural machinery operation paths based on regional division according to claim 1, wherein: The initial feasible path group from the starting coordinate to the end coordinate is obtained by the standard particle swarm algorithm.

6. The method for planning agricultural machinery operation paths based on regional division according to claim 1, characterized in that: Perform path search in the sub-area to obtain the optimal path in the sub-area, and obtain the final required optimal feasible path through comparison, specifically: The optimal path within the sub-region is obtained through the particle swarm algorithm; Use fractional calculus to update the velocity term of the particle swarm algorithm; In the position update item, the position information of the particle swarm algorithm is updated by setting the energy consumption weight; By updating the speed and position, the optimal path in the sub-area is obtained, and when the requirements or the maximum iteration are met, the final optimal feasible path is obtained.

7. The method for planning agricultural machinery operation paths based on regional division according to claim 6, characterized in that: The speed update is as follows: in, is the inertia weight, is the particle velocity, , is the particle learning factor, is the particle acceleration, 、 is a random number between (0,1), The current iteration number.

8. The agricultural machinery operation path planning system based on regional division is characterized by: include: The information collection module is configured to: obtain the size of the agricultural machinery operation area and the environmental map information formed by obstacles in the area; The information processing module is configured to: use the location of the agricultural machinery as the starting coordinate and the location of the warehouse as the ending coordinate, and perform environmental modeling based on the acquired environmental map information; The path planning module is configured to: obtain an initial feasible path group from the starting coordinates to the end coordinates, treat obstacles in the environmental map information as circular obstacles, use the coordinates of the circle center as the connection node of the boundary of each sub-region in the region division, extend the next node from the end coordinate as the initial node, connect the initial node with each extended node to realize the region division, obtain multiple groups of sub-regions, and perform path search within the sub-regions to obtain the optimal path within the sub-regions, and obtain the final optimal feasible path after updating; To achieve regional division, specifically: define the line connecting the centers of each obstacle in the environmental map information as the center line, take the location of the end point coordinates as the parent node, and take the obstacles at both ends of the center line with the most intersections with the feasible path group as the child nodes of the parent node; the child node serves as the parent node of the next level node and extends until reaching the final node, which is the starting coordinate.

9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the method for planning an agricultural machinery operation path based on area division as described in any one of claims 1 to 7 are implemented.

Images