Unmanned agricultural machine operation path planning algorithm

Abstract

The invention discloses an operation path planning algorithm for an unmanned agricultural machine. The method comprises the following steps: obtaining a field distribution diagram of a target area, carrying out grid division on the field distribution diagram of the target area, obtaining the number of grids required when the unmanned harvester is full of harvested crops, presetting N unmanned harvester operation paths on the field distribution diagram, screening out an operation path corresponding to the minimum value of a stroke S, and the path is used as an optimal harvesting operation path to recommend the unmanned harvester. The invention provides an unmanned agricultural machine operation path planning algorithm suitable for being used in cooperation with a small harvester and a small transfer truck, the optimal operation path planning can be provided when the unmanned agricultural machine is used in cooperation with the small harvester and the small transfer truck, efficient and accurate operation of the agricultural machine is facilitated, and the unmanned agricultural machine is particularly suitable for harvesting underground crops such as potatoes, potatoes and radishes; the implementation steps provided by the invention are simple.

Description

Technical Field

[0001] The present invention belongs to the field of agricultural machinery information technology, and particularly relates to an operation path planning algorithm for unmanned agricultural machinery. Background Art

[0002] With the rapid development of autonomous driving technology, more and more agricultural machinery is being tested for autonomous operation. Typically, high-quality farmland is divided into several small plots by tractor and tillage roads. Each plot has multiple entrances and exits, connected directly or indirectly via tractor or tillage roads. These entrances and exits are depicted on high-precision maps. Based on these maps, it is necessary to generate an operating path plan for the unmanned agricultural machinery to ensure optimal travel (including road travel between plots and tillage operations).

[0003] In the prior art, document CN115145257A discloses a path planning and precision operation method for harvesting with an unmanned harvester. This method includes path planning and precision operation steps. The method obtains parameter information such as the harvester's size, minimum turning radius, and harvesting width. It also obtains the location information of the field boundary points, connects the generated closed graph, and uses it as the field boundary. An all-terrain path is planned based on the land boundary, the operation starting point, and the operation width. After the harvester completes each inner circle, it reaches the common edge of the inner circle and the sweeping circle to perform the sweeping circle operation. However, this solution relies on a limited circular operation path for path planning, which is not an optimal operation path.

[0004] Currently, many farms in China are not suitable for using oversized agricultural machinery (hundreds of horsepower) and heavy-duty trucks. Most growers use small agricultural machinery, including small harvesters and small transfer trucks (such as tricycles). However, this requires frequent dumping (harvesting crops into a silo during harvesting and then dumping them into the cargo box of a transfer truck once the silo is full) and transfer (transfer trucks transporting harvested crops to crop storage areas) during the harvesting process. This results in complex operations and requires further optimization of operational efficiency. To address this issue, it is necessary to develop an unmanned agricultural machinery path planning algorithm suitable for use with small harvesters and small transfer trucks, at least to guide the efficient and intelligent operation of small agricultural machinery. Summary of the Invention

[0005] In response to the technical problems mentioned in the background technology, the purpose of the present invention is to provide an unmanned agricultural machinery operation path planning algorithm.

[0006] The present invention adopts the following technical solutions.

[0007] An unmanned agricultural machinery operation path planning algorithm includes the following steps: S01, obtaining a field distribution map of the target area, where the entrance and exit of the unmanned harvester are marked; S02, dividing the field distribution map of the target area into grids, where the grid length and width are equal to the single effective harvesting width of the unmanned harvester during operation (basically equal to the width of the harvesting tool of the unmanned harvester, which can be obtained by actual measurement); S03, obtaining the number of grids required when the unmanned harvester harvests a full warehouse of crops; the purpose of this step is to quickly and accurately determine the corresponding position when the unmanned harvester harvests a full warehouse of crops; S04: Preset N unmanned harvester operation paths on the field distribution map, and calibrate the position P (x, y) of the grid end corresponding to each operation path when the unmanned harvester harvests a full warehouse of crops. S05: Taking all positions P (x, y) on each operation path as the points that the unmanned harvester must pass through, calculate the distance S that the unmanned harvester travels through all positions P (x, y) on the field distribution map; S06: Filter out the operation path corresponding to the minimum value of the travel distance S, and recommend the path to the unmanned harvester as the optimal harvesting operation path.

[0008] Furthermore, the method further comprises the steps of: S07, using the position P (x, y) on the obtained optimal harvesting operation path as the point that the unmanned truck must pass through, and calculating the distance M from each position P (x, y) on the path to the crop storage area along several material transfer paths; S08: Filter out the material transfer path corresponding to the minimum value of the travel distance M, and recommend the material transfer path as the optimal material transfer operation path to the unmanned truck.

[0009] As a preferred option, the unmanned harvester harvesting a full silo of crops means that the silo of the unmanned harvester is full of crops, and the volume of the truck cargo box following the unmanned harvester is basically equal to an integer multiple of the silo volume, the volume of the truck cargo box is basically equal to one times the silo volume, and the inlet and outlet of the unmanned harvester are at the same location or at different locations.

[0010] As a preferred solution, the trip M is the driving distance of the unmanned truck along the closed-loop route formed by the unmanned harvester entrance, position P (x, y), the unmanned harvester exit, and the crop collection area.

[0011] As a preferred solution, the turning points of the unmanned harvester are calibrated on the field distribution map. When calculating the travel distance S, the distance that the unmanned harvester needs to travel during the turning process at the turning point (including the forward distance and reverse distance during the turning process) is set as a constant value. This distance can be based on the data obtained by the unmanned truck during the field test.

[0012] In the present invention, the method further includes: introducing a particle swarm optimization algorithm to perform a global search based on the existing itinerary S to further optimize the path selection process; A multi-objective path planning mechanism is introduced to integrate the goals of path length and energy consumption into the objective function. Based on different weight settings, multiple goals in path planning are optimized and the weights are adjusted in real time in the algorithm. Adjust the path of the unmanned harvester based on the feedback from the optimization algorithm and multi-objective planning.

[0013] Furthermore, initialize the particle swarm , where N is the number of nodes to be considered in path planning, each particle represents a path, and each particle represents the solution of path planning through an initial position and velocity, defining the velocity of each movement ; According to the standard update formula of the particle swarm optimization algorithm, the particle speed is adjusted and the particle position is updated and iterated continuously. The expression is: in, is the inertia weight, is the velocity of particle i in dimension j at step t, is the position of particle i at step t in dimension j, , is the acceleration constant, is a random number, is the historical optimal position of the particle, is the global optimal position; By continuously updating the position and velocity, the particle swarm searches in the solution space and obtains a path solution close to the optimal one. , providing a preliminary solution for subsequent path optimization.

[0014] Furthermore, the multi-objective path planning mechanism is as follows: based on the particle swarm optimization algorithm calculation, multi-objective optimization factors are added, and multi-objective path planning is introduced, and the objective function is defined as: in, , It is the coefficient for adjusting the path length and energy consumption to ensure that the weight of each target can be adjusted dynamically. Path solution for the current iteration The path length, Solution for the current iteration path path energy consumption; The objective function is to minimize the , thus obtaining the best path; The optimized multi-target path is fed back to the path planning system of the unmanned harvester, and the path planning is adjusted in real time.

[0015] Beneficial effects: The present invention provides an unmanned agricultural machinery operation path planning algorithm suitable for use with small harvesters and small material transfer trucks. It can provide optimal operation path planning when used in conjunction with them, which is conducive to efficient and precise operation of agricultural machinery, and is particularly suitable for harvesting underground crops such as potatoes, potatoes, and radishes; the implementation steps provided by the present invention are simple and can be achieved with a simple algorithm model. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a schematic diagram of planning path 1 on a certain field distribution map in Example 1; Figure 2 This is a schematic diagram of planning path 2 on a certain field distribution map in Example 1; Figure 3 This is a schematic diagram of planning path three on a certain field distribution map in Example 1; Figure 4 This is a schematic diagram of a material transfer path on a certain field distribution map in Example 1; Figure 5 This is a schematic diagram of the method flow of embodiment 2 of an unmanned agricultural machinery operation path planning algorithm proposed by the present invention. DETAILED DESCRIPTION

[0017] The following is a clear and complete description of the technical solutions of the present invention in conjunction with the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention. Example

[0018] A small unmanned harvester and a small unmanned truck are used to harvest potatoes in a field. The field has an entrance and exit for the agricultural machinery (the entrance and exit are located in the same location) located in the lower right corner of the field. The one-way distance between the entrance and exit for the agricultural machinery and the crop storage area is 1000 meters. The volume of the small unmanned harvester's silo is equal to the volume of the truck's cargo box, and each time the small unmanned harvester harvests an area 6 meters long and 2 meters wide, the silo is completely full. In this embodiment, an algorithm for planning the path of an unmanned agricultural machine operation includes the following steps: S01, obtain a field distribution map of the target area, on which the entrance and exit of the unmanned harvester are marked (corresponding to position P (x0, y0) in the figure); S02, the field distribution map of the target area is divided into grids, and the grid width is equal to the single effective harvesting width of the unmanned harvester during operation. The grid width and length after scaling are both 2m; S03, obtaining the number of grids required for the unmanned harvester to harvest a full silo of crops. According to the silo volume of the unmanned harvester, the number of grids required when the silo is full is 3, and the corresponding harvesting area is 6m*2m wide; S04: Preset N unmanned harvester operation paths on the field distribution map, and calibrate the position P (x, y) of the grid end corresponding to each time the unmanned harvester harvests a full warehouse of crops for each operation path, such as Figures 1 to 3 The calibrated P(x1,y1), P(x2,y2)…P(x 76 ,y 76 ));like Figure 1 As shown in FIG, the arrow direction shown in K1 represents the preset first unmanned harvester operation path, which is a harvesting operation along the length direction of the field distribution map; Figure 2 As shown in FIG, the arrow direction shown in K2 represents the preset second unmanned harvester operation path, which is a harvesting operation along the width direction of the field distribution map; Figure 3 As shown, the arrow direction shown by K3 represents the preset third unmanned harvester operation path, which is a circular direction / path along the field distribution map for harvesting; S05: Calculate the distance S that the unmanned harvester travels through all positions P(x,y) on the field distribution map, taking all positions P(x,y) on each operation path as points that the unmanned harvester must pass through. Demarcate the turning points of the unmanned harvester on the field distribution map, and when calculating the distance S, set the distance that the unmanned harvester needs to travel during the turning process at the turning point as a constant value. In this embodiment, the average distance that the unmanned harvester needs to travel during the turning process at the turning point is 8 meters. Then, the preset first unmanned harvester operation path distance S = 76 (number of positions P) * 2 + 11 (number of corresponding turning points) * 8 = 240m, the preset second unmanned harvester operation path distance S = 76 (number of positions P) * 8 + 18 (number of corresponding turning points) * 8 = 296m, and the preset third unmanned harvester operation path distance S = 76 (number of positions P) * 8 + 23 (number of corresponding turning points) * 8 = 336m; S06, filter out the operation path corresponding to the minimum value of the stroke S, which is as follows Figure 1 The first preset unmanned harvester operation path is shown, and the unmanned harvester is recommended as the optimal harvesting operation path; S07: Using position P (x, y) on the optimal harvesting path as a required point for the unmanned truck, calculate the distance M from each position P (x, y) on the path to the crop collection area along several transfer paths. The distance M represents the distance traveled by the unmanned truck along the closed loop formed by the unmanned harvester entrance, position P (x, y), the unmanned harvester exit, and the crop collection area. The position P (x 27 ,y 27 ) as an example, first preset multiple material transfer paths: the first material transfer path is as follows Figure 4 As shown in the middle path A, the corresponding travel distance M=(18+1000)*2=2036m; the second material transfer path is as follows Figure 4 As shown in the middle path C, the corresponding travel distance M=2025.46m; the third material transfer path is as follows Figure 4 As shown in the middle path B, the corresponding distance M=(18+1000)*2=2036m; S08, filter out the material transfer path corresponding to the minimum value of the stroke M, that is, Figure 4 The preset path C shown has the shortest travel distance and is recommended as the unmanned harvester's operating path. This transfer path is also recommended as the optimal transfer path for the unmanned truck. Finally, the optimal harvesting and transfer paths are stored and sent to the grower's management terminal. Example

[0019] Reference Figure 5 ,The method also includes introducing a particle swarm optimization algorithm to ,perform a global search based on the existing itinerary S to ,further optimize the path selection process.

[0020] Specifically, initialize the particle swarm , where N is the number of nodes to be considered in path planning, each particle represents a path, and each particle represents the solution of path planning through an initial position and velocity, defining the velocity of each movement ; According to the standard update formula of the particle swarm optimization algorithm, the particle speed is adjusted and the particle position is updated and iterated continuously. The expression is: in, is the inertia weight, is the velocity of particle i in dimension j at step t, is the position of particle i at step t in dimension j, , is the acceleration constant, is a random number, is the historical optimal position of the particle, is the global optimal position; Through this formula, the particle's speed and position will be updated in each iteration, and the particle will adjust its path based on its historical experience and global experience.

[0021] By continuously updating the position and velocity, the particle swarm searches in the solution space and obtains a path solution close to the optimal one. , providing a preliminary solution for subsequent path optimization.

[0022] The introduction of the particle swarm optimization algorithm enables a global search for path planning, avoiding the trap of local optimal solutions. By simulating the collaboration and competition of a swarm of particles, the particle swarm optimization algorithm can effectively explore a broader solution space, ensuring the optimal solution is found under complex operating conditions. It is particularly adaptable in complex terrain or when the task requirements are diverse.

[0023] A multi-objective path planning mechanism is introduced to integrate the goals of path length and energy consumption into the objective function. According to different weight settings, multiple goals in path planning are optimized and the weights are adjusted in real time in the algorithm.

[0024] Adjust the path of the unmanned harvester based on the feedback from the optimization algorithm and multi-objective planning.

[0025] Specifically, the multi-objective path planning mechanism is as follows: based on the particle swarm optimization algorithm calculation, multi-objective optimization factors are added, and multi-objective path planning is introduced, and the objective function is defined as: in, , It is the coefficient for adjusting the path length and energy consumption to ensure that the weight of each target can be adjusted dynamically. Path solution for the current iteration The path length, Solution for the current iteration path path energy consumption; The objective function is to minimize the , thus obtaining the best path; The optimized multi-target path is fed back to the path planning system of the unmanned harvester, and the path planning is adjusted in real time.

[0026] By combining multiple objectives such as path length, energy consumption, and operation time for comprehensive optimization, the algorithm can provide the optimal path for the unmanned harvester under different operating conditions, thereby effectively shortening the operation time, reducing energy consumption, and greatly improving operation efficiency and quality.

[0027] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An algorithm for planning the path of an unmanned agricultural machine operation, characterized in that the steps include: S01, obtaining a field distribution map of the target area, where the entrance and exit of the unmanned harvester are marked; S02, dividing the field distribution map of the target area into grids, where the grid width is equal to the effective single harvesting width of the unmanned harvester during operation; S03, obtaining the number of grids required for the unmanned harvester to harvest a full warehouse of crops; S04: Preset N unmanned harvester operation paths on the field distribution map, and calibrate the position P (x, y) of the grid end corresponding to each operation path when the unmanned harvester harvests a full warehouse of crops. S05: Taking all positions P (x, y) on each operation path as the points that the unmanned harvester must pass through, calculate the distance S that the unmanned harvester travels through all positions P (x, y) on the field distribution map; S06: Filter out the operation path corresponding to the minimum value of the travel distance S, and recommend the path to the unmanned harvester as the optimal harvesting operation path.

2. The unmanned agricultural machinery operation path planning algorithm according to claim 1, characterized in that: Also includes the steps: S07, using the position P (x, y) on the obtained optimal harvesting operation path as the point that the unmanned truck must pass through, and calculating the distance M from each position P (x, y) on the path to the crop storage area along several material transfer paths; S08: Filter out the material transfer path corresponding to the minimum value of the travel distance M, and recommend the material transfer path as the optimal material transfer operation path to the unmanned truck.

3. The unmanned agricultural machinery operation path planning algorithm according to claim 2, characterized in that: An unmanned harvester harvesting a full silo of crops means that the silo of the unmanned harvester is full of crops, and the volume of the truck box following the unmanned harvester is basically equal to an integer multiple of the volume of the silo, the volume of the truck box is basically equal to one times the volume of the silo, and the inlet and outlet of the unmanned harvester are at the same location or at different locations.

4. The unmanned agricultural machinery operation path planning algorithm according to claim 2, characterized in that: The trip M is the driving distance of the unmanned truck along the closed-loop route formed by the unmanned harvester entrance, position P (x, y), unmanned harvester exit, and crop collection area.

5. The unmanned agricultural machinery operation path planning algorithm according to claims 1-4, characterized in that: The turning point of the unmanned harvester is calibrated on the field distribution map. When calculating the travel distance S, the distance that the unmanned harvester needs to travel during the turning process at the turning point is set as a constant value.

6. The unmanned agricultural machinery operation path planning algorithm according to claim 5, characterized in that: The obtained optimal harvesting operation path and the obtained optimal material transfer operation path are stored and sent to the grower's management terminal.

7. The unmanned agricultural machinery operation path planning algorithm according to claim 1, characterized in that: The method further comprises the following steps: The particle swarm optimization algorithm is introduced to conduct global search based on the existing itinerary S to further optimize the path selection process; A multi-objective path planning mechanism is introduced to integrate the goals of path length and energy consumption into the objective function. Based on different weight settings, multiple goals in path planning are optimized and the weights are adjusted in real time in the algorithm. Adjust the path of the unmanned harvester based on the feedback from the optimization algorithm and multi-objective planning.

8. The unmanned agricultural machinery operation path planning algorithm according to claim 7, characterized in that: The particle swarm optimization algorithm includes the following steps: Initialize particle swarm , where N is the number of nodes to be considered in path planning, each particle represents a path, and each particle represents the solution of path planning through an initial position and velocity, defining the velocity of each movement ; According to the standard update formula of the particle swarm optimization algorithm, the particle speed is adjusted and the particle position is updated and iterated continuously. The expression is: in, is the inertia weight, is the velocity of particle i in dimension j at step t, is the position of particle i at step t in dimension j, , is the acceleration constant, is a random number, is the historical optimal position of the particle, is the global optimal position; By continuously updating the position and velocity, the particle swarm searches in the solution space and obtains a path solution close to the optimal one. , providing a preliminary solution for subsequent path optimization.

9. The unmanned agricultural machinery operation path planning algorithm according to claim 8, characterized in that: The method further includes: The multi-objective path planning mechanism is as follows: based on the particle swarm optimization algorithm calculation, multi-objective optimization factors are added, and multi-objective path planning is introduced, and the objective function is defined as: in, , It is the coefficient for adjusting the path length and energy consumption to ensure that the weight of each target can be adjusted dynamically. Path solution for the current iteration The path length, Solution for the current iteration path path energy consumption; The objective function is to minimize the , thus obtaining the best path; The optimized multi-target path is fed back to the path planning system of the unmanned harvester, and the path planning is adjusted in real time.

Images