CN114676901A

CN114676901A - Method for globally optimizing operation path of land leveler

Info

Publication number: CN114676901A
Application number: CN202210269280.0A
Authority: CN
Inventors: 魏潇洋
Original assignee: Zhenjiang Jintai Intelligent Technology Co ltd
Current assignee: Zhenjiang Jintai Intelligent Technology Co ltd
Priority date: 2022-03-18
Filing date: 2022-03-18
Publication date: 2022-06-28

Abstract

The invention discloses a land leveling machine operation path global optimization method for leveling farmland lands, wherein a digital elevation model of a field to be leveled is rasterized by adopting a square grid, the position weight of the field to be leveled is determined according to the positions of grid points in the field to be leveled, and the total earth volume, the target flat elevation and the planned excavation and filling earth volume of each grid of the field to be leveled are calculated in sequence; adopting an ant colony algorithm, taking the current load earth volume of each grid and the existing earth volume to be excavated and filled of the grid as criteria for node transfer, taking one of all circulating optimal paths with the highest path efficiency evaluation index as an optimal path, and smoothing the optimal path to obtain a global optimal path; the invention adopts the position weight of the grid points to obtain the total earth volume, ensures the accuracy of earth volume calculation, improves the ant colony algorithm, obtains the optimal operation path with balanced shortest path and highest efficiency, sets the path efficiency evaluation index and effectively improves the operation efficiency.

Description

Method for globally optimizing operation path of land leveler

Technical Field

The invention belongs to the field of agricultural machinery, and relates to a land leveler for leveling farmland land, in particular to a path planning method for land leveler operation.

Background

In agricultural production, the degree of land leveling of agricultural fields significantly affects the efficiency of water, fertilizer and drug use, and thus the demand for land leveling equipment has rapidly increased in recent years. The existing land leveling equipment can meet the requirement of agricultural production in the land leveling precision during small-area work, but has lower efficiency in the leveling work of large-area farmlands. At present, the manpower resources in agricultural production are short, experienced agricultural operators are in short supply, and the land leveling cost is high.

The existing land leveling equipment is mainly a land leveling shovel, automatic adjustment can be realized during land leveling shovel operation, and the automatic adjustment method comprises the following steps: the height, the attitude, the speed and other information of the land leveling blade are obtained through laser or a Global Navigation Satellite System (GNSS), and the attitude of the land leveling blade is adjusted in real time according to the information. This can achieve very high accuracy in a small range, but due to lack of path planning, the overall flatness can only be achieved by repeated rework during large-area work, which is time consuming and labor consuming.

For flat ground path planning, the shortest path is mainly planned, factors such as the planned excavation and filling volume of a field block, the loading capacity of a grader, the steering characteristic and the like need to be considered in land leveling operation, and the factors are not considered in the conventional shortest path planning process.

For the calculation of the earth volume, a commonly used calculation method of a Digital Elevation Model (DEM) is adopted. The principle is that mesh points are selected on a target land block according to a triangle or a square, the elevation of each mesh point is weighted and averaged to obtain an average elevation, and a weighting coefficient adopted in weighted averaging is estimated based on the relief condition around each mesh point, so that it is obvious that a certain error exists in the average elevation. In order to quantify the error, the excavation and filling volume of each grid point needs to be calculated according to the elevation difference between the elevation of each grid point and the estimated average elevation, the difference value between the planned excavation volume and the planned filling volume is the error, the weight coefficient of each grid point is adjusted until the difference value is zero, and then the more accurate average elevation is obtained.

The DEM grid method is also applicable to the calculation of the earth volume of farmland land leveling work. However, for most of farmlands needing to be leveled, the farmlands themselves are used for a long time, the levelness of the farmlands is good, and large relief is avoided.

The principle of the ant colony algorithm is as follows: the ant colony has n ants, each ant has own memory, nodes which cannot be accessed in the search are stored in the memory by a tabu table, nodes which can be accessed by the allowed table are stored in the allowed table, and pheromones released by the allowed table to the passed path in a cycle are stored in a matrix. Pheromones are volatilized to a certain extent, which affects the selection of ants on the branches of multiple paths, and pheromones accumulate faster on shorter paths, thereby attracting more and more ants to travel along the shortest path. Some ant colony algorithms also introduce heuristic functions to reflect the visibility of ants between nodes during path selection. Some shortest path planning uses ant colony algorithm to find the best, and the shortest path is obtained.

At present, the ant colony algorithm is applied to land scraper path planning, but because the factors of earth volume are not considered in land leveling operation, the land leveling path planning cursive rate is regarded as the ant colony algorithm problem to search the shortest path. If the ant colony algorithm is effectively used for flat ground path planning, the characteristics of land leveling work need to be improved on the key factors of the ant colony algorithm, such as node selection, taboo table updating rules, heuristic function setting and updating rules.

In view of the above, a need exists for a global optimization method for a grader working path to solve the above-mentioned problems.

Disclosure of Invention

The invention aims to provide a global optimization method for a land leveling machine operation path, which integrates rapid acquisition of farmland three-dimensional topography information required in farmland land leveling work and global path planning conforming to land leveling machine operation characteristics, and solves the problems of low working efficiency, low automation degree and overhigh labor cost of the conventional land leveling equipment.

In order to solve the problems, the invention discloses a land leveler operation path global optimization method which adopts the technical scheme that: the method comprises the following steps of collecting terrain information of a field to be leveled to obtain a digital elevation model of the field to be leveled, and the method further comprises the following steps:

step 1): adopting square grid rasterization to the digital elevation model, and determining the position weight of the digital elevation model according to the position of grid points in the field to be leveled;

step 2): according to the position weight of each grid point, the total earth volume V of the field to be leveled and the target flat ground height H of the field to be leveled are calculated in sequence_eAnd planned fill volume V for each grid_g；

Step 3): planning earth volume V in all grids by adopting ant colony algorithm_gThe maximum grid center point is used as the starting point, and the amount of the excavated earth volume V is calculated in all the grids _gThe grid closest to 0 is used as the terminal point, and the earth volume V is planned in all the grids_gAbsolute value of | V_g| reach the set earth volume threshold value V_sThe center point of the grid of (2) is taken as a node; with current load earthwork V of land leveling blade of land leveling machine of each grid_LAnd the amount of earth volume V to be excavated and filled existing in the grid_PAs a criterion of node transfer, judging whether the node should be put into a taboo table according to the load state of the land leveling shovel, and completing a round of circulation to obtain an optimal path P;

and step 4): calculating a path efficiency evaluation index I of the optimal path P, and updating cycle times and a taboo table;

step 5): judging whether the loop times or the path efficiency evaluation index I are reached, if so, taking the highest one of the path efficiency evaluation indexes I in the optimal paths P of all loops as the optimal path P_bFor the optimal path P_bAnd performing smoothing to obtain a global optimization path B (u).

The invention adopts the technical proposal to highlight the technical effects that:

1. the traditional digital elevation model grid method is that the elevations of all grid points of a grid are weighted and averaged to obtain average elevations, then the soil excavation amount and the soil filling amount are calculated according to the average elevations, and then the weights are corrected until the soil excavation amount and the soil filling amount are basically equal. According to the method, the flatness of the farmland is good, the change is smooth, the farmland area is large, the elevation weight significance is not large, the calculated amount is increased, the digital elevation model grid method is improved according to the characteristics of farmland leveling operation, the traditional complex elevation weight averaging method is abandoned, the position weight of grid points is adopted to obtain the total earth volume, the accuracy of earth volume calculation is guaranteed, and the calculated amount is greatly reduced.

2. Because the operation path planning of the land leveler is not simple shortest path planning and needs to comprehensively consider factors such as operation path length, machine tool load condition, planning workload and the like, the method improves the ant colony algorithm, selects the starting point, the terminal point and the path node of the ant colony algorithm according to the grid plan earth excavation and filling amount, establishes a relation between the priority of the advancing direction of the land leveler, the real-time earth load of the land leveler and the target node plan earth excavation and filling amount and the heuristic function of the ant colony algorithm, and affects the advancing probability between the nodes together with the pheromone reflecting the path length, thereby obtaining the optimal operation path with balanced shortest path and highest efficiency.

3. The invention sets the path efficiency evaluation index according to the operating efficiency characteristic of the land leveler, and reflects the proportion of the effective operation of the land leveler in the route, namely the proportion of the idle running state caused by no load or full load of the land leveler, by comparing the total excavating and filling amount of the land leveler in the whole route of the work of planning the path with the total planned excavating and filling amount of each grid on the route, thereby effectively improving the operating efficiency of the land leveler.

4. The method uses the second-order bezier curve to carry out smoothing treatment on the planned optimal path, adapts to the steering characteristic of the land leveler, provides global optimization of the operation path for land leveling work, obtains the accurate optimal path, and promotes the improvement of the working efficiency of the land leveler.

Drawings

In order to more clearly illustrate the technical solution of the present invention, the drawings needed for the detailed description will be briefly described below, it being understood that the following drawings only show some embodiments of the present invention and therefore should not be considered as limiting the scope, and that for a person skilled in the art, other related drawings may be obtained from these drawings without inventive effort.

FIG. 1 is a flow chart of a method for global optimization of a grader work path in accordance with the present invention;

FIG. 2 is a flow chart of a method of constructing the three-dimensional terrain model of FIG. 1;

FIG. 3 is a flow chart of a method of obtaining the planned cut and fill volume of FIG. 1;

FIG. 4 is a plot of the effect of the rasterized field of FIG. 1 along with associated parameters and grid points;

FIG. 5 is a plot of the four corner points and center point of any of the grids of FIG. 4;

FIG. 6 is a plot of nodes of the local grid of FIG. 4 and edges corresponding to paths between the nodes;

fig. 7 is a flow chart of the improved ant colony algorithm of fig. 1.

Detailed Description

In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

As shown in fig. 1, the terrain information of the field to be leveled is collected, and a three-dimensional terrain model of the field to be leveled is constructed. With reference to fig. 2, the device equipped with Global Navigation Satellite System (GNSS) positioning function is driven to perform one-time traversal idle running on the field to be leveled to acquire the terrain information of the field to be leveled, and the RTK positioning data of the device is periodically acquired through GNSS during idle running, that is, the RTK positioning data contains horizontal position X, Y data and altitude elevation H data, so as to obtain three-dimensional coordinate points (X, Y, H), record the data, and reconstruct the three-dimensional terrain model of the field to be leveled by using the three-dimensional coordinate points (X, Y, H). The altitude elevation H data can also be acquired by using a laser mode, and the GNSS method has better working environment adaptability and wider application, so the method adopts the GNSS mode to acquire the data. The established three-dimensional terrain model is the premise of subsequent path planning and other work.

As shown in fig. 3, according to the three-dimensional terrain model of the field to be leveled, the digital simulation of the ground terrain is realized through limited three-dimensional coordinate point (X, Y, H) data, and a Digital Elevation Model (DEM) is obtained. In order to reduce the requirement of computational complexity, the invention adopts a zero-order monomial digital geomorphic model, namely a digital elevation model.

And rasterizing the digital elevation model, wherein the rasterization mainly comprises a regular square grid and a triangular grid. Since the calculation of the earth volume by the regular square grid method is more favorable for using GPS measurement data, and the calculation is relatively simple, the square grid is used in the invention.

The field to be leveled is rasterized according to the set side length L, the local rasterized square grid effect of the field to be leveled is shown in figure 4, the outermost frame in figure 4 is the contour line of the field to be leveled, and most of the field fields are close to a rectangle, so the X axis is set to be parallel or close to be parallel to the short edge of the field, and the Y axis is set to be parallel or close to be parallel to the long edge of the field.

The method has the advantages that the field is rasterized, the ground characteristics of the field and the operation characteristics of the land scraper are considered, and the calculation complexity can be reduced under the condition of ensuring the precision. As most of farmlands needing land leveling are used as farmlands before leveling, the land leveling degree is good, the ground change trend is smooth, and steep rising and steep falling basically do not exist. Therefore, the data amount is further reduced by rasterization without causing large deviation. The side length L of the grid is set to be equal to the working width of the grader blade, taking into account the working and turning characteristics of the grader.

After rasterization, three-dimensional coordinates (X, Y, H) of each grid point are obtained, the weight of the grid point is determined according to the position of the grid point in the field to be smoothed, the position of the grid point in the field to be smoothed refers to four conditions of a middle point and an edge point of the square grid or an inward and outward angle drop point, and the condition is not limitedA complex estimation needs to be performed. As shown in FIG. 4, when the grid point is at the middle point m (X) of the square grid_m，Y_m，H_m) The position weight of the middle point m is determined to be 4, where X_m，Y_mHorizontal position of middle point m, H_mIs the elevation of the middle point m. When the grid point is at the inward corner point c1 (X) of the square grid_c1，Y_c1，H_c1) The position weight of the inward corner point c1 is determined to be 3, where X_c1，Y_c1Horizontal position of inward corner point c1, H_c1Is the elevation of the inward corner point c 1. When the grid point is at the edge point s (X) of the square grid_s，Y_s，H_s) The position weight of the edge point s is determined to be 2, where X_s，Y_sIs the horizontal position of the edge point s, H_sIs the elevation of the edge point s. When the grid point is at the outward corner point c2 (X) of the square grid_c2，Y_c2，H_c2) When it is determined that the position weight is 1, where X_c2，Y_c2Horizontal position of the outward corner point c2, H_c2The elevation of the outward corner point c 2.

After the position weight of each grid point is determined, the total earth volume V of the field to be leveled is calculated according to the following formula:

Wherein S is_gIs the area of a single grid; h_mThe elevation of the grid point at the middle point of the square grid is taken as the elevation of the grid point; h_c1The elevation of the grid point at an inward corner point of the square grid is shown; h_sThe elevation of the grid point at the edge point of the square grid is taken as the elevation; h_c2The elevation of the grid point at the outward corner point of the square grid is shown.

Calculating the average elevation H of the field to be leveled according to the total earth volume V_eI.e. target land level H_e：

Wherein S is the total area of the field to be leveled; ideally, the land leveling work does not require the carrying in or out of the earthwork, and both the amount of the excavated earthwork and the amount of the filled earthwork inside the field should be balanced, and thus, the target level land height H_eIt represents the desired elevation of the field after it has been leveled.

According to the target flat ground elevation H_eCalculating the planned excavation and filling volume V of each grid_gAs shown in fig. 3, the calculation formula is:

wherein H is the elevation of four corner points of the corresponding grid; when planned to dig and fill volume V_gWhen the grid is larger than 0, the grid needs to be dug; when planned to dig and fill volume V_gWhen the grid is smaller than 0, the grid needs to be filled with soil; when planned to dig and fill volume V_gWhen the value is equal to 0, it means that the grid earth volume does not need to be increased or decreased.

Four corner points and a center point of any one grid are shown in fig. 5, the four corner points are a1, a2, A3 and a4, and the center point is O.

The planned excavation and filling volume V of each grid is obtained_gThen, according to the horizontal position (X, Y) of the central point O of each grid and the planned volume V of the earth excavated and filled by the corresponding grid_gThe improved ant colony algorithm shown in fig. 7 is adopted to plan the path of the land scraper so as to obtain the optimal path P_bThe specific method comprises the following steps:

step 1: and setting a starting point, an end point and each path node of the ant colony algorithm.

Comparing planned excavation and fill volume V of each grid_gBy the size of (1), the volume of earth excavated in all grids is planned_gThe largest grid center point is taken as the starting point. Since the grader is unloaded initially, the planned cut and fill volume V is determined_gThe largest grid starts to work, so that the flat shovel can be fully loaded on the first grid, and the load utilization rate is maximized.

Cut and fill in all gridsVolume V_gThe grid closest to 0 is taken as the end point. Because the grader is actually running, the load should be emptied before reaching the end point, the empty load reaches the end point and the next round of work is started.

Planning and excavating earth volume V in all grids_gAbsolute value of | V_g| reach the set earth volume threshold value V_sThe central point of the grid is used as the node of the ant colony algorithm, and the set earth volume threshold value V_sHour, earth volume threshold value V _sShould be equal to the maximum load V of the land leveling shovel_LmaxRelated to, and should be much less than, the maximum load amount V_Lma。

Because of the unavoidable error in the earth volume measurement, the grader needs to perform at least one traversal operation to remove small errors in the operation after finishing the land leveling work according to the optimal path. Thus, the planned fill volume V_gVery small grids can be excluded from the necessary nodes in path planning because the traversal of the job path and the final traversal job can easily bring them to the flat goal.

And 2, step: parameters for initializing ant colony algorithm, including ant number N, cycle number N, and edge e corresponding to path between nodes_ijIntensity of pheromone of upper initial_ij(ii) a Edge e corresponding to path between nodes_ijDegree of heuristic of Upper initial η_ijAnd i and j represent two different nodes.

As shown in FIG. 6, the edge e corresponding to the path between each node_ijAs shown in FIG. 6, node O of the first grid₁(X₁,Y₁) Node O to the second grid₂(X₂,Y₂) The corresponding edge e of the path between₁₂Node O of the second grid₂(X₂,Y₂) Node O to the third grid₃(X₃,Y₃) The corresponding edge e of the path between₂₃Node O of the second grid₂(X₂,Y₂) Node O to the fourth grid 3₄(X₄,Y₄) The corresponding edge e of the path between ₂₄。

And 3, step 3: and setting the advancing rule of the ants according to the advancing mode of the land leveler.

Ants can only travel along the X-axis direction and the Y-axis direction, and when the ants travel from one grid node to the next grid node, the ants can only select grids in the front direction, the front left direction and the front right direction.

And 4, step 4: and calculating the forward probability of the ants when the ants are transferred between the nodes, and executing the node transfer according to the operation characteristics of the land scraper.

When any ant k transfers between nodes, the probability of transferring from the node i to the node j in the cycle time t is

The calculation formula is as follows:

wherein, tau_ijFor the edge e corresponding to the path between node i and node j_ijIntensity of pheromones on; eta_ijReflecting the heuristic degree from i to j for visibility factors between the nodes i and j; alpha is the importance index of pheromone; beta is an importance index of the heuristic factor.

At the time of node transfer, its rules need to be executed according to the operating characteristics of the grader. Because steering of the grader can cause additional work deflection, straight ahead is given a higher priority than left and right turns when the nodes are shifted.

After the grader passes through each grid, the current load earth volume V of the grader blade passing through each grid is recorded_LAnd the amount of earth volume V to be excavated and filled existing in the grid _PAnd the node transfer criterion is used. Volume V of existing earth to be excavated in grid_PIs initialized to its planned fill volume V_g。

Planned fill volume V for each grid_gWhen the earth needs to be dug, the planned earth volume is a positive value, and when the earth needs to be filled, the planned earth volume is a negative value. When the grader works, the grader blade is considered to be running empty and full for a long time, and should be avoided as much as possible for the sake of working efficiency.

Shovel the current load earth volume V on the level land_LLess than maximum load V_LmaxOne-half of (c), the next node should be loaded to avoid empty load, so the existing volume of earth to be excavated and filled V of the node is preferably selected_PThe current load earth volume V of the land leveling shovel_LAdding the maximum load V closest to the land scraper_LmaxThe node of (2).

The existing load V of the land scraper_LGreater than the maximum load V_LmaxOne-half of (c), the next node should be reduced in load to avoid full-load empty-running, so the amount of earth volume V to be excavated in the node grid is preferably selected_PCurrent load earth volume V of land leveling shovel_LThe nodes closest to 0 are added.

And 5: and judging whether the node should be put into a taboo table according to the load state of the land leveling shovel.

Current load earth volume V of land leveling blade _LThe working condition of the grader at the corresponding node is also reflected. If the land scraper is full or no load after passing through the node, the planned excavation and filling volume V representing the node_gIs not satisfied; on the other hand, if the land scraper is not in a full load or no load state after passing through the node, the planned excavation and filling volume V of the node is represented_gIs complete. If one of the nodes passes through the land leveler for the first time in the cycle, the planned cut and fill volume V is achieved_gRepresenting the planned fill volume V of the grid_gAnd if the node is small and does not need to be included in the planned path, the node is put into a taboo table of the loop of the current round.

Step 6: after one round of circulation is completed, recording the optimal path P of the circulation, calculating the path efficiency evaluation index I of the optimal path P, updating the circulation times and the taboo table according to the path efficiency evaluation index I, and updating the pheromone.

The calculation formula of the path efficiency evaluation index I is as follows:

wherein，V_dRepresenting the total cut of the grader path, V_fRepresenting the total fill volume, V, of the grader path_glRepresenting the earth excavation and filling quantity requirement of each grid of the path. The closer the path efficiency evaluation index I is to 1, the less empty running due to no load or full load of the land shovel in the representative path, and the higher the work efficiency.

Edge e corresponding to road between two nodes i and j_ijThe incremental calculation of the above pheromone factor with the cycle is as follows:

wherein, Δ τ_ij(t, t +1) is via edge e_ijAll ant opposite sides e of_ijThe pheromone amount contribution of;

is the ant k opposite side e_ijThe pheromone increment contributed above; ρ is the pheromone residual coefficient; q is the total amount of pheromone released by all ants after one cycle; l is_kIs the length of the path that the ants go out in the current cycle.

And 7: judging whether the number of times of circulation is reached or whether the circulation reaches the path efficiency evaluation index I, if so, ending the circulation, and taking the highest path efficiency evaluation index I in the optimal paths P of all the circulation as the optimal path P of the field leveling operation_bFor the optimal path P_bAnd performing smoothing processing and outputting a path planning result.

Using the efficiency evaluation index I as the selection optimal path P_bThe only criterion is that the influence of the no-load or full load of the land leveling shovel on the total operation time is far larger than the operation path length in the land leveling work, and the length of the optimal path P of each cycle is close to the theoretical shortest path length according to the principle of the ant colony algorithm, so the path length is not taken as the criterion for selecting the optimal path.

To accommodate the steering characteristics of the grader, at the optimal path P_bAll turning positions are smoothed by using a second-order bezier curve to obtain a second-order bezier curve B (u), namely a global optimization path B (u), wherein the expression of the second-order bezier curve B (u) is as follows:

B(u)＝(1-u)²P₀+2u(1-u)P₁+u²P₂ 0≤u≤1

take the example of the turning path from the first grid to the second grid to the third grid in fig. 6, where u is the argument of the curve; p is₀To the horizontal coordinate of the node of the preceding grid, i.e. the first node (X)₁,Y₁)；P₁The horizontal coordinate of the node of the grid occurring for steering, i.e. the second grid node (X)₂,Y₂)；P₂For turning to the node horizontal coordinate of the latter grid, i.e. the node horizontal coordinate (X) of the third grid₃,Y₃)。

Claims

1. A global optimization method for a land leveling machine operation path is used for acquiring terrain information of a field to be leveled to obtain a digital elevation model of the field to be leveled, and is characterized by comprising the following steps:

step 2): according to the position weight of each grid point, the total earth volume V of the field to be leveled and the target flat ground height H of the field to be leveled are calculated in sequence _eAnd planned fill volume V for each grid_g；

And step 3): planning earth volume V in all grids by adopting ant colony algorithm_gUsing the maximum grid center point as the starting point, and using the planned volume of filled earth V in all grids_gThe grid closest to 0 is taken as an end point, and the planned earth volume V is calculated in all the grids_gAbsolute value of (V)_g| reach the set earth volume threshold value V_sThe center point of the grid of (a) is taken as a node; with current load earthwork V of land leveling blade of land leveling machine of each grid_LAnd the amount of earth volume V to be excavated and filled existing in the grid_PAs a criterion for the node transfer,judging whether the node should be put into a tabu table according to the load state of the land leveling shovel, and completing a round of circulation to obtain an optimal path P;

step 4): calculating a path efficiency evaluation index I of the optimal path P, and updating cycle times and a taboo table;

2. The method of global optimization of a grader work path according to claim 1, wherein: the method comprises the steps of performing one-time traversal and idle running on a field to be leveled to acquire terrain information of the field to be leveled, periodically acquiring RTK positioning data through a GNSS (global navigation satellite system) to obtain three-dimensional coordinate points, and performing digital simulation on ground terrain by using the three-dimensional coordinate points to obtain a digital elevation model.

3. The global optimization method for a grader work path according to claim 1, wherein: in the step 1), the side length of the grid is set to be equal to the operation width of a land leveling shovel of the land leveling machine.

4. The global optimization method for a grader work path according to claim 1, wherein: in the step 1), when the grid point is at the middle point of the square grid, the position weight is 4; when the grid point is at the inward corner point of the square grid, the position weight is 3; when the grid point is at the edge point of the square grid, the position weight is determined to be 2; when a grid point is at an outward corner point of the square grid, its position weight is 1.

5. The method of global optimization of a grader work path according to claim 4, wherein: in step 2), according to the formula

Calculate the total earthwork of the field to be leveledQuantity V according to the formula

Calculating the target flat ground height H of the field to be leveled_eAccording to the formula amount

Calculating the planned excavation and filling volume V of each grid_g；S_gIs the area of a single grid; h_mThe elevation of the grid point at the middle point of the square grid is shown; h_c1The elevation of the grid point at an inward corner point of the square grid is shown; h_sThe elevation of the grid point at the edge point of the square grid is taken as the elevation; h_c2The elevation of the grid point at the outward corner point of the square grid is shown; s is the total area of the field to be leveled; and H is the elevation of four corner points of the corresponding grid.

6. The method of global optimization of a grader work path according to claim 1, wherein: and 3) setting that the ants can only select grids in the three directions of front, left and right when driving from one grid node to the next grid node according to the running mode of the land leveler, and the priority of the straight running is higher than that of the left and right turning.

7. The method of claim 6, wherein the method comprises: after the land leveler passes through each grid, the current load earth volume V is shoveled on the land_LLess than maximum load V_LmaxOne-half of (c), the amount of earth volume V to be excavated and filled existing in the node is preferably selected_PThe current load earth volume V of the land leveling shovel_LAdding the maximum load V closest to the land scraper_LmaxA node of (2); when the existing load V of the land scraper_LGreater than the maximum load V_LmaxOne-half of (c), the amount of earth volume V to be excavated in the node grid is preferably selected_PCurrent load earth volume V of land leveling shovel_LThe nodes closest to 0 are added.

8. Grader work according to claim 7The path global optimization method is characterized by comprising the following steps: if one node passes through the land leveler for the first time, the planned excavation and filling volume V is achieved_gThen put the node into the tabu table.

9. The method of global optimization of a grader work path according to claim 1, wherein: in the step 4): according to the formula

Calculating out path efficiency evaluation indexes I and V_dIs the total soil excavation amount of the land leveler path, V_fIs the total fill volume, V, of the land leveler path_glThe required soil excavating and filling amount of each grid of the path.

10. The global optimization method for a grader work path according to claim 1, wherein: in step 5), smoothing is performed by using a second-order bezier curve to obtain a second-order bezier curve b (u) ═ 1-u)²P₀+2u(1-u)P₁+u²U is a curve independent variable, and u is more than or equal to 0 and less than or equal to 1; p is₀Horizontal coordinate, P, of node for the grader steering to the previous grid₁Horizontal coordinates of nodes for steering generation grids, P₂Is the horizontal coordinate of the node that is steered to the next grid.