CN117235558A - Multi-field block route scheduling planning method for large-area hilly and mountainous areas - Google Patents
Multi-field block route scheduling planning method for large-area hilly and mountainous areas Download PDFInfo
- Publication number
- CN117235558A CN117235558A CN202311508910.6A CN202311508910A CN117235558A CN 117235558 A CN117235558 A CN 117235558A CN 202311508910 A CN202311508910 A CN 202311508910A CN 117235558 A CN117235558 A CN 117235558A
- Authority
- CN
- China
- Prior art keywords
- partition
- area
- algorithm
- planning
- route
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000005192 partition Methods 0.000 claims abstract description 93
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 69
- 230000002068 genetic effect Effects 0.000 claims abstract description 27
- 238000000137 annealing Methods 0.000 claims abstract description 11
- 238000005507 spraying Methods 0.000 claims abstract description 11
- 238000000638 solvent extraction Methods 0.000 claims abstract description 10
- 238000004364 calculation method Methods 0.000 claims abstract description 3
- 238000002922 simulated annealing Methods 0.000 claims description 10
- 238000010586 diagram Methods 0.000 claims description 7
- VYZAMTAEIAYCRO-UHFFFAOYSA-N Chromium Chemical compound [Cr] VYZAMTAEIAYCRO-UHFFFAOYSA-N 0.000 claims description 6
- 239000003814 drug Substances 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 239000007921 spray Substances 0.000 claims description 6
- 102100024058 Flap endonuclease GEN homolog 1 Human genes 0.000 claims description 5
- 101000833646 Homo sapiens Flap endonuclease GEN homolog 1 Proteins 0.000 claims description 5
- 230000003044 adaptive effect Effects 0.000 claims description 5
- 125000003275 alpha amino acid group Chemical group 0.000 claims description 5
- 150000001875 compounds Chemical class 0.000 claims description 5
- 239000013589 supplement Substances 0.000 claims description 4
- 238000007621 cluster analysis Methods 0.000 claims description 3
- 238000001816 cooling Methods 0.000 claims description 3
- 230000035772 mutation Effects 0.000 claims description 3
- IAZDPXIOMUYVGZ-UHFFFAOYSA-N Dimethylsulphoxide Chemical compound CS(C)=O IAZDPXIOMUYVGZ-UHFFFAOYSA-N 0.000 claims description 2
- 241000238631 Hexapoda Species 0.000 claims description 2
- 241000607479 Yersinia pestis Species 0.000 claims description 2
- 201000010099 disease Diseases 0.000 claims description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 2
- 239000003337 fertilizer Substances 0.000 claims description 2
- 241000196324 Embryophyta Species 0.000 description 34
- 241001122767 Theaceae Species 0.000 description 17
- 210000000349 chromosome Anatomy 0.000 description 4
- 229940079593 drug Drugs 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 101150044561 SEND1 gene Proteins 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 238000005496 tempering Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a multi-field block route scheduling planning method for a large-area hilly and mountainous area, which comprises the following steps: clustering and partitioning the operation area by using a Kmeans clustering algorithm, and planning out the optimal replenishment position of each partition to obtain a multi-area plant protection operation clustering partition map; traversing the single area of each partition according to a full-coverage route planning algorithm until all areas are traversed, and calculating plant protection spraying quantity of each partition after obtaining effective operation route of each area and the position of an unmanned aerial vehicle in-out point; and (3) marking out an optimal route sequence of scheduling planning of the operation area by adopting genetic algorithm rules of a self-adaptive dynamic annealing idea, and then determining the load quantity required by each frame according to full-coverage route calculation of each partition, thereby obtaining a unmanned plane plant protection task scheduling aerial graph. The multi-field route scheduling planning method for the large-area hilly mountain areas, which is provided by the technical scheme, can effectively improve the operation efficiency and population diversity of the algorithm, and simultaneously enhance the searching capability and optimizing precision of the algorithm.
Description
Technical Field
The invention relates to the technical field of unmanned aerial vehicle aerial survey, in particular to a multi-field route scheduling planning method for large-area hilly mountain areas.
Background
In agricultural production, plant protection operation is a very important link. The traditional plant protection operation modes mainly use machines such as a handheld sprayer, a ground plant protection machine and the like to spray, but the modes have the problems of low efficiency, high cost, environmental pollution and the like. Along with the continuous development of science and technology, unmanned aerial vehicle has obtained wide application in plant protection operation field, and it has advantages such as coverage area is big, operating efficiency is high, flexible operation, consequently becomes the new trend of plant protection operation.
However, in actual operation, in the plant protection operation of the unmanned aerial vehicle in a plurality of small and scattered areas in hilly and mountainous areas, the unmanned aerial vehicle needs to make multiple passes to and from the ground station and the field for energy replenishment due to the limitations of the capacity of the unmanned aerial vehicle medicine box, the capacity of the battery and the single plant protection capability, so that the planning problem of the plant protection operation sequence is relatively complex. At present, the conventional job scheduling method cannot effectively solve the problem, and a plant protection job scheduling method capable of performing targeted partition planning on a job area according to the plant protection capability of an unmanned aerial vehicle is needed, so that the job efficiency and the quality are improved.
Disclosure of Invention
The invention aims to provide a multi-field route scheduling planning method for a large-area hilly and mountainous area, which effectively solves the problem that the existing unmanned aerial vehicle is complicated to plan due to the limitation of the capacity of a medicine box, the capacity of a battery and the single-frame plant protection capability in plant protection operation.
In order to solve the technical problems, the invention adopts the following technical scheme:
step one: determining the number of optimal clustering partitions of the operation area by calculating a clustering contour coefficient value, then clustering and partitioning the operation area by using a Kmeans clustering algorithm, and planning out the optimal replenishment position of each partition to obtain a multi-area plant protection operation clustering partition map;
step two: traversing single areas of each partition according to a full-coverage route planning algorithm according to the operation clustering partition diagram obtained in the step one until all areas are traversed, and calculating plant protection spraying quantity of each partition after obtaining effective operation route of each area and the position of an unmanned aerial vehicle in-out point;
step three: and (3) obtaining the route entry and exit points of each zone and the vertexes of each zone based on the full coverage route of each zone obtained by the full coverage route planning algorithm in the step two, then marking out the optimal route sequence of the operation zone scheduling planning by adopting the genetic algorithm rule of the self-adaptive dynamic annealing idea, and then determining the load quantity required by each frame according to the full coverage route calculation of each zone, thereby obtaining the unmanned plane plant protection task scheduling air map.
The first step is as follows:
1-1: first, the number of different partitions is calculated byg（1<g<Total number of segments) to the contour coefficients of the clustered division of the discrete dot patterns formed by the vertices of each segmentS（i) (in [0,1 ]]The larger the value in the value range of (a) is, the better the partitioning effect is
Wherein:a（i) -an average value of the distances between each patch and the patch vertices within a certain partition;a 1 representing the distance between the first patch and the vertex of the patch in a certain partition;
b（i) -an average of the vertex distances of a partition to each of the tiles in the nearest partition;b 1 representing the vertex distance from a partition to the first tile in the nearest partition;
q-the number of tiles within a partition;
h-the number of tiles within a partition that is closest to the partition;
wherein the profile coefficientS(i) The closer to 1, the more compact the distance between the tiles in each partitioned partition, the greater the inter-partition distance, indicating the number of partitionsgThe more reasonable the cluster division; the partition number corresponding to the maximum contour coefficient is the optimal clustering partition number, and then the total operation area is divided into the optimal clustering partition numbergPartitioning to obtain a cluster analysis chart;
1-2: abstracting the optimal replenishment position planning problem into an optimization problem for solving minimum value points of feature points of a plurality of areas, in particular to solving Euclidean distancedIs the minimum value of (2):
wherein (1)>For the distance of the partition vertex coordinates to the replenishment point,c i for the coordinates of the vertices of the partition,ccoordinates of the replenishing points;
the algorithm flow for planning the position of the supplement point is as follows:
input: input initial discrete coordinate points {｝；
And (3) outputting: coordinates of the feed pointscAnd minimum value of the distance between the partition vertex coordinates and the replenishment point coordinatesd’；
The method specifically comprises the following steps of;
(1) input devicegIndividual partition vertex coordinatesrand(g,2)；
(2) For the generation ofgIndividual partition vertex coordinatesc i =｛Calculating its coordinates to the replenishment pointcEuclidean distance of (2)d；
(3) Solving Euclidean distanceMinimum value of (2)d’；
(4) Outputting the coordinates of the replenishing points until the square difference between the vertex coordinates of the partition and the coordinates of the replenishing points is minimumcAnd minimum value of the distance between the partition vertex coordinates and the replenishment point coordinatesd’。
The genetic algorithm of the adaptive dynamic annealing concept comprises the following steps:
3-1: the genetic algorithm stage adopts a method for randomly generating a population to obtain an initial population Chrom; if 123456, 213465, 345216 are all one of the chromosomes in the population, the number is not repeated; if the population number in the first step is set to 30, 30 randomly generated chromosomes are in the initial population;
3-2: setting the population size and the length of a first layer of genetic algorithm according to the number of areas needing to spray medicines in a decimal coding mode, wherein if 20 tea areas need to be applied, the population length is set to be 20+3=23; setting iteration times GEN1 of a genetic algorithm of a first layer as a constant, wherein the crossover probability adopts self-adaptive crossover operation probability, the variation probability adopts self-adaptive variation probability, and the code channel is set as a constant; the cross probability is set as formula (1), the variation probability is set as formula (2), and the ditch is set as 0.9-0.95;
（1）
in the formula (1), the components are as follows,is the average value of fitness of individuals to be crossed;/>is the minimum value of the fitness; />Is the maximum value of the fitness; />Is the average value of the fitness;k 1 is a constant value, and is used for the treatment of the skin,k 1 =1；
（2）
in the formula (2), the amino acid sequence of the compound,fitness for variant individuals; />Is the maximum value of the fitness; />Is the average value of the fitness;k 2 、k 3 is a constant value, and is used for the treatment of the skin,k 2 ∈[0,0.001]，k 3 =0.2；
3-3: the first 20% of the offspring solution set generated in the later stage of the genetic algorithm is selected to carry out a simulated annealing operation stage, and the improved metropolis criterion used by the second-layer simulated annealing algorithm refers to the formulas (3) and (4) and the cooling factoralphaIs provided as formula (5) and formula (6); the second layer of simulated annealing algorithm is used for avoiding the improvement of local optimum and optimizing the optimum solution value;
（3）
（4）
wherein,E t+1 representing a new energy state (new solution),E t indicating the energy state at this time (old solution),T m the adjustment coefficient representing the acceptance criterion is used,Trepresenting a preset initial temperature;E(xmax) AndE(xmin) Representing the maximum and minimum values of the objective function corresponding to the N possible solutions randomly selected from the first 20% of the solution set of the previous genetic algorithm,T 0 indicating an initial temperature;
（5）
（6）
wherein,T s the initial temperature is indicated as such,krepresenting the number of current iterations and,mis the number of changes corresponding to the intersection point of the logarithmic function and the exponential function, the DeltaN represents the metropolis rule to determine whether to accept the new solutionThe ratio of the total number of links to the number of times of receiving the current optimal solution;
3-4: detecting whether the iteration times exceeds the maximum iteration times GEN, returning to the step 3-3 when the iteration times are less than or equal to the GEN, otherwise, performing the step 3-5;
3-5: reading the job sequence in the Chrom corresponding to the shortest scheduling path, and recording the length of the shortest scheduling path;
3-6: the algorithm ends.
The method for planning the multi-field route in the large-area hilly mountain area adopts a mode of combining various algorithms, including a replenishing position planning method based on a Kmeans algorithm, a genetic algorithm of a self-adaptive dynamic annealing idea and the like, and adopts dynamic crossing and mutation strategies in the genetic algorithm, so that the operation efficiency and population diversity of the algorithm can be effectively improved; the improved Metropolis rule and the temperature control method of self-adaptive tempering are adopted, so that the stability of the probability of acceptance is enhanced, and the searching capability and optimizing precision of the algorithm are enhanced; the initial solution set of the annealing idea in the algorithm depends on the optimization result of the genetic algorithm part, and the whole algorithm depends on the annealing idea to prevent optimization from converging in advance. Performing task partitioning on the operation area, and considering single-frame endurance of the unmanned aerial vehicle; the iteration is needed for GEN1+GEN2 times, and GEN1 and GNE2 are respectively the maximum iteration times set by two layers of algorithms of the algorithm. According to the multi-field route scheduling planning method for the large-area hilly and mountainous areas, the operation amount is reasonably distributed in a targeted manner by considering the cruising ability, the loading ability and the flying speed of the unmanned aerial vehicle, and the loads of all the frames are reasonably distributed according to the planning result, so that the unmanned aerial vehicle can return to a supply point to supplement electric quantity and plant protection resources after the plant protection resources of all the regional operations are exhausted, and the flying task of the next frame operation area is prepared.
Drawings
FIG. 1 is a flow chart of a genetic algorithm of the adaptive dynamic annealing concept;
FIG. 2 is a diagram of a multiple tea field route planning task;
FIG. 3 is a schematic diagram of a routing within a monolithic region;
FIG. 4 is a monolithic vertex acquisition diagram;
FIG. 5 is a multiple tea field global airline scheduling plan aerial graph;
FIG. 6 is a diagram ofgClustering partition results at=3;
FIG. 7 is a schematic diagram of the result of replenishment position planning.
Detailed Description
The present invention will be specifically described with reference to examples below in order to make the objects and advantages of the present invention more apparent. It should be understood that the following text is intended to describe only one or more specific embodiments of the invention and does not limit the scope of the invention strictly as claimed.
The invention firstly utilizes Kmeans clustering algorithm to cluster and partition the operation area and plan the optimal replenishment position of each partition to obtain a multi-piece plant protection operation cluster partition map shown in figure 6.
Referring to fig. 2, the drone starts from the drone landing point H in fig. 2, then traverses each tile through the entry and exit points of each tile, and then returns to H. The route requiring inter-regional scheduling is the shortest. And adopting a full-coverage route planning algorithm shown in fig. 3 in the area, namely taking the longest edge of the boundary as parallel lines according to the width of the unmanned aerial vehicle, and turning around and flying the unmanned aerial vehicle until all areas of the film area are traversed when the unmanned aerial vehicle reaches other boundary boundaries. Each tile must have an in-point and an out-point. The scheduling must be done in and out from two points, but not specifically what point. Such as from Z2 in fig. 3, must exit from Z3, or from Z2 after entering from Z3.
After the effective operation range of each zone and the position of the unmanned aerial vehicle in-out point are obtained, the plant protection spraying quantity of each zone is calculated, after the optimal route sequence of the operation zone scheduling planning is drawn by the genetic algorithm rule of the self-adaptive dynamic annealing idea, the load quantity required by each frame is determined, and finally the unmanned aerial vehicle plant protection task scheduling air map is obtained.
The method for dispatching and planning the multi-field block route in the large-area hilly mountain area is described below by taking a tea field as an example, and comprises the following steps:
step 1: the population is expanded according to the following expansion method, for example, 20 patch scheduling tasks, and the english letters A, B, C … T are respectively corresponding to decimal numbers 1, 2, 3..20 one by one, as shown in table 1:
table 1 reference examples of coordinate values required for scheduling of each tile
| Number of sheet area | Numbering in algorithm | Midpoint x 0 | Midpoint y 0 |
| A | 1 | 790.49 | 43.04 |
| B | 2 | 790.49 | 100.64 |
| C | 3 | 790.49 | 155.41 |
| D | 4 | 764.52 | 188.16 |
| E | 5 | 747.59 | 50.38 |
| F | 6 | 753.79 | 93.86 |
| G | 7 | 741.94 | 152.02 |
| H | 8 | 723.30 | 217.52 |
| I | 9 | 697.89 | 61.11 |
| J | 10 | 697.89 | 98.38 |
| K | 11 | 697.89 | 132.26 |
| L | 12 | 692.24 | 173.48 |
| M | 13 | 684.34 | 244.06 |
| N | 14 | 656.67 | 63.37 |
| O | 15 | 613.20 | 63.37 |
| P | 16 | 635.22 | 117.58 |
| Q | 17 | 635.22 | 155.97 |
| R | 18 | 644.25 | 192.11 |
| S | 19 | 636.35 | 253.66 |
| T | 20 | 598.51 | 159.37 |
Step 2: according to the single plant protection efficiency of the selected plant protection unmanned aerial vehicle, clustering and partitioning are carried out on the operation areas of the multiple tea fields and the energy supply positions of the unmanned aerial vehicles are planned through a Kmeans algorithm-based supply position planning method, in the embodiment, 20 tea field operation areas with total areas of about 60 mu are taken as an example, and limitation of the cruising ability of the actual plant protection unmanned aerial vehicle is considered, and step 2: according to the single plant protection efficiency of the selected plant protection unmanned aerial vehicle, clustering and partitioning are carried out on the working areas of the multiple tea fields through a Kmeans algorithm-based replenishment position planning method, and the unmanned aerial vehicle energy replenishment positions are planned, in the embodiment, 20 tea field working areas with the total area of about 60 mu are taken as an example, considering the limitation of the cruising ability of the actual plant protection unmanned aerial vehicle, the total working area is required to be divided into a plurality of subareas, and fig. 6 is a view of dividing the total working area into 3 subareas @gCluster partition map of=3).
Wherein,gcalculated by the following method:
first, the number of different partitions is calculated byg（1<g<Total number of segments) to the contour coefficients of the clustered division of the discrete dot patterns formed by the vertices of each segmentS（i) (in [0,1 ]]The larger the value in the value range, the better the partitioning effect is specified):
wherein:a（i) -an average value of the distances between each patch and the patch vertices within a certain partition;a 1 representing the distance between the first patch and the vertex of the patch in a certain partition;
b（i) -an average of the vertex distances of a partition to each of the tiles in the nearest partition;b 1 representing the vertex distance from a partition to the first tile in the nearest partition;
q-the number of tiles within a partition;
h-within the partition closest to a certain partitionThe number of tiles;
wherein the profile coefficientS(i) The closer to 1, the more compact the distance between the tiles in each partitioned partition, the greater the inter-partition distance, indicating the number of partitionsgThe more reasonable the cluster division; the partition number corresponding to the maximum contour coefficient is the optimal clustering partition number, and then the optimal clustering partition number is usedg(the present embodiment givesg=3) dividing the total operation area into 3 partitions to obtain a cluster analysis chart;
step 3: abstracting the optimal replenishment position planning problem into an optimization problem for solving minimum value points of feature points of a plurality of areas, in particular to solving Euclidean distancedIs the minimum value of (2):
wherein (1)>For the distance of the partition vertex coordinates to the replenishment point,c i for the coordinates of the vertices of the partition,ccoordinates of the replenishing points;
the algorithm flow for planning the position of the supplement point is as follows:
input: input initial discrete coordinate points {｝；
And (3) outputting: coordinates of the feed pointscAnd minimum value of partition vertex coordinates and replenishment point distanced’；
The method comprises the following steps:
(1) input devicegIndividual partition vertex coordinatesrand(g,2)；
(2) For the generation ofgIndividual partition vertex coordinatesc i =｛Calculating its coordinates to the replenishment pointcEuclidean distance of (2)d；
(3) Solving Euclidean distanceMinimum value of (2)d’；
(4) Until the square difference between the partition vertex coordinates and the replenishment point coordinates is minimized, outputting the replenishment point coordinatescAnd minimum value of partition vertex coordinates and replenishment point distanced’；
The optimal replenishment point coordinates are shown in table 2:
table 2 three-partition optimal feed point planning result coordinates
| Number of sheet area | Numbering in algorithm | Midpoint x 0 | Midpoint y 0 |
| U | 21 | 768.47 | 111.93 |
| V | 22 | 659.17 | 199.45 |
| W | 23 | 678.30 | 105.93 |
Genetic algorithm of adaptive dynamic annealing concept referring to fig. 1;
step 4: the genetic algorithm stage adopts a method for randomly generating a population to obtain an initial population Chrom. If 123456, 213465, 345216 are all one of the chromosomes in the population, the numbers are not repeated. If the first population is set to 30, there are 30 randomly generated chromosomes in the initial population.
The population obtained by the first layer genetic algorithm needs to be expanded, and a custom expansion algorithm is adopted. If ABC represents a job sequence solved by the first layer algorithm, A represents 1, B represents 2, C represents 3, namely the job sequence is 1-2-3, and 1-4-2-5-3-6 is adopted to represent a real scheduling path after the point entering is considered. At this time, the extended real scheduling path is represented by 1 for point A1, 4 for point A2, 2 for point B1,5 for point B2,3 for point C1, and 6 for point C2. I.e. each patch plus a number of fields of tea indicates the second point of the tea zone. In the above-mentioned operation sequence of 1-4-2-5-3-6, at this time, each entry and exit point is regarded as a point of the problem of the traveling salesman, and the distance of 1-4-2-5-3-6-1 is calculated according to the neighborhood table of the query expansion, namely, the distance A1-A2-B1-B2-C1-C2-A1 is denoted as S1, and the distance is denoted as the complete distance. The real dispatching path distance of the multi-tea field=the complete distance-the sum of the connecting line distances of the in-out access points in each tea area, such as the length of S0=S1- (A1A2+B1B2+C1C2).
Step 5: and setting the population size and the length of the first layer genetic algorithm according to the number of tea areas needing to spray medicines in a decimal coding mode, wherein for example, 20 tea areas need to be applied, and the population length is set to be 20+3=23. The reason is that each tea area is represented by the most point position of the end point of each tea area calculated in the step 2, as shown by Z in fig. 4, the taking-off and landing point of the unmanned aerial vehicle in fig. 3 is also regarded as a point, and the scheduling task of 20 tea areas is regarded as the scheduling task of 23 tea areas when the taking-off and landing point of the unmanned aerial vehicle is placed at the last position of the population length. The iteration number GEN1 of the genetic algorithm of the first layer is set to be constant, for example, 500. The crossover probability adopts self-adaptive crossover operation probability, the variation probability adopts self-adaptive variation probability, and the code is set as a constant. The crossover probability is set as formula (1), the mutation probability is set as formula (2), and the ditch is set as 0.9-0.95.
（1）
In the formula (1), the components are as follows,is the average value of fitness of individuals to be crossed;/>is the minimum value of the fitness; />Is the maximum value of the fitness; />Is the average value of the fitness;k 1 is a constant value, and is used for the treatment of the skin,k 1 =1；
（2）
in the formula (2), the amino acid sequence of the compound,fitness for variant individuals; />Is the maximum value of the fitness; />Is the average value of the fitness;k 2 、k 3 is a constant value, and is used for the treatment of the skin,k 2 ∈[0,0.001]，k 3 =0.2。
step 6: the first 20% of the offspring solution set generated in the later period of the genetic algorithm is selected to carry out a simulated annealing operation stage, the improved metropolis criteria used by the second-layer simulated annealing algorithm are shown as formula (3) and formula (4), the maximum iteration number MaxOutIter of the outer layer loop of SA is set to be 100, the maximum iteration number MaxInIter of the inner layer loop is set to be 50, and the initial temperature T is set to be the initial temperature T 0 Set to 0.02, the cooling factor alpha is set to formula (5) and formula (6). The second layer simulated annealing algorithm is used for avoiding local optimization and optimizing the optimal solution value, and the improvement of the use of the second layer simulated annealing algorithm.
（3）
（4）
Wherein,E t+1 representing a new energy state (new solution),E t indicating the energy state at this time (old solution),T m the adjustment coefficient representing the acceptance criterion is used,Trepresenting a preset initial temperature;E(xmax) AndE(xmin) Representing the maximum and minimum values of the objective function corresponding to the N possible solutions randomly selected from the first 20% of the solution set of the previous genetic algorithm,T 0 indicating the initial temperature.
（5）
（6）
Wherein,T s the initial temperature is indicated as such,krepresenting the number of current iterations and,mis a value fixed in advance. The number of changes corresponding to the intersection point of the logarithmic function and the exponential function is used asm. For a pair oflogFunction derivation, can obtainm=15. The Δn represents the ratio of the total number of times that the metapolis rule determines whether to accept new solutions to the number of times that the current optimal solution is accepted.
Step 7: and (3) detecting whether the iteration times exceed the maximum iteration times GEN, returning to the step (6) when the iteration times are less than or equal to the GEN, otherwise, performing the step (8).
Step 8: and reading the job sequence in the Chrom corresponding to the shortest scheduling path, and recording the length of the shortest scheduling path.
Step 9: the algorithm ends.
Step 10: after obtaining the optimal route sequence of multi-tea-field dispatching planning, determining the number of times of the frame of each partition through a method (7), and establishing the unmanned aerial vehicle route according to the allocation of the frameAnd (3) calculating the load amount required by each frame according to a relation between the plant protection area and the formula (8). The spraying amount (mu amount) of each hectare is set by determining the severity of plant diseases and insect pests or the fertilizer requirement of crops in advanceωUnmanned aerial vehicle operation speedvWidth of spray widthw. Then the total range of the unmanned aerial vehicle is set asR j Unmanned aerial vehicle work areaSAmount of sprayingQThe method comprises the following steps of:
（7）
in the method, in the process of the invention,R j to the firstjThe total course of farmland plant protection operation during the setting up;R 1 representing the total course of farmland plant protection operation at the 1 st time;nrepresenting the last frame; when the coverage algorithm obtains the optimal full coverage route, the full coverage total range of each zone is obtained, and the workload of each plant protection zone can be obtained according to the formula (7).
（8）
In the formula (8), the amino acid sequence of the compound,L j is the firstjThe load amount required by the frame time;S j to the theoretical firstjThe total area of the operation is completed for the time;S j-1 to the theoretical firstj-1 total area of completed jobs;wthe width of the spray width of the plant protection unmanned aerial vehicle is set;vis the operation speed;Fis the spraying flow.
The invention expands the multi-zone route scheduling planning problem into a multi-travel-provider problem, and the number of the zones of the operation zone is assumed to be 12, the flight frame times of the unmanned aerial vehicle is assumed to be 3, the number of cities visited by each travel provider is not equal, data combinations with unequal numbers of {1,3,8,4}, {2,5,7,10}, and {11,6,9,12} 3 groups can be obtained, and then each group of traditional travel-provider problems can be solved.
In conclusion, the multi-field route scheduling planning method for the large-area hilly and mountainous areas can effectively solve the problem that the existing unmanned aerial vehicle is limited by the capacity of a medicine box, the capacity of a battery and the single-frame plant protection capability in plant protection operation and is complicated to plan.
While the embodiments of the present invention have been described in detail with reference to the drawings, the present invention is not limited to the above embodiments, and it will be apparent to those skilled in the art that various equivalent changes and substitutions can be made therein without departing from the principles of the present invention, and such equivalent changes and substitutions should also be considered to be within the scope of the present invention.
Claims (5)
1. The multi-field block route scheduling planning method for the large-area hilly and mountainous areas is characterized by comprising the following steps of:
step one: determining the number of optimal clustering partitions of a working area by calculating a clustering contour coefficient value, then clustering and partitioning the working area by using a Kmeans clustering algorithm, and planning out the optimal replenishment position of each partition to obtain a multi-area plant protection working clustering partition map;
step two: traversing single areas of each partition according to a full-coverage route planning algorithm according to the operation clustering partition diagram obtained in the step one until all areas are traversed, and calculating plant protection spraying quantity of each partition after obtaining effective operation route of each area and the position of an unmanned aerial vehicle in-out point;
step three: and (3) obtaining the route entry and exit points of each zone and the vertexes of each zone based on the full coverage route of each zone obtained by the full coverage route planning algorithm of the step two, then marking out the optimal route sequence of the operation zone scheduling planning by adopting the genetic algorithm rule of the self-adaptive dynamic annealing idea, and then determining the load quantity required by each frame according to the full coverage route calculation of each zone to obtain the unmanned plane plant protection task scheduling air map.
2. The method for multi-field block airlines scheduling planning of large-area hilly mountain area as set forth in claim 1, wherein the step one specifically includes:
1-1: first, the number of different partitions is calculated bygWherein 1 is<g<Total number of tiles, clustering and dividing discrete point patterns formed by the vertices of each tileProfile coefficientS（i），
；
Wherein:a（i) -an average value of the distances between each patch and the patch vertices within a certain partition;a 1 representing the distance between the first patch and the vertex of the patch in a certain partition;
b（i) -an average of the vertex distances of a partition to each of the tiles in the nearest partition;b 1 representing the vertex distance from a partition to the first tile in the nearest partition;
q-the number of tiles within a partition;
h-the number of tiles within a partition that is closest to the partition;
wherein the partition number corresponding to the maximum contour coefficient is the optimal clustering partition number, and the operation area is divided into the operation areas according to the optimal clustering partition numbergPartitioning to obtain a cluster analysis chart;
1-2: abstracting the optimal replenishment position planning problem into an optimization problem for solving minimum value points of feature points of a plurality of areas, in particular to solving Euclidean distancedIs the minimum value of (2):
；
wherein,for the distance of the partition vertex coordinates to the replenishment point coordinates,c i for the coordinates of the vertices of the partition,ccoordinates of the replenishing points;
the algorithm flow for planning the position of the supplement point is as follows:
input: input initial discrete coordinate points {｝；
And (3) outputting: coordinates of the feed pointscAnd minimum value of the distance between the partition vertex coordinates and the replenishment point coordinatesd’；
The method comprises the following steps:
(1) input devicegIndividual partition vertex coordinatesrand(g,2)；
(2) For the generation ofgIndividual partition vertex coordinatesc i =｛Calculating its coordinates to the replenishment pointcEuclidean distance of (2)d；
(3) Solving Euclidean distanceMinimum value of (2)d’；
(4) Outputting the coordinates of the replenishing points until the square difference between the vertex coordinates of the partition and the coordinates of the replenishing points is minimumcAnd minimum value of the distance between the partition vertex coordinates and the replenishment point coordinatesd’。
3. The method for multi-field block airlines scheduling planning in large-area hilly mountain areas according to claim 1, wherein the full coverage airlines planning algorithm specifically comprises: and (3) taking the longest edge of the boundary of the single area as parallel lines according to the width of the spraying width of the unmanned aerial vehicle, and turning around and flying the unmanned aerial vehicle when the unmanned aerial vehicle reaches the boundary of other boundaries until all areas of the area are traversed.
4. The method for multi-field block airline scheduling planning for large-area hilly mountainous areas according to claim 2, characterized in that the genetic algorithm of the adaptive dynamic annealing concept comprises the following steps:
3-1: the genetic algorithm stage adopts a method for randomly generating a population to obtain an initial population Chrom;
3-2: setting the population size and the length of a first layer of genetic algorithm according to the number of the areas needing to be sprayed with the medicine by adopting a decimal coding mode; setting iteration number GEN1 of the genetic algorithm of the first layer as a constant, wherein the crossover probability adopts an adaptive crossover operation probabilityThe rate, the variation probability adopts the self-adaptive variation probability, and the code is set as a constant; crossover probabilityThe formula (1) is set as the mutation probabilityThe formula (2) is set, and the ditch is set to be 0.9-0.95;
（1）
in the formula (1), the components are as follows,is the average value of fitness of individuals to be crossed; />Is the minimum value of the fitness; />Is the maximum value of the fitness;is the average value of the fitness;k 1 is a constant value, and is used for the treatment of the skin,k 1 =1；
（2）
in the formula (2), the amino acid sequence of the compound,fitness for variant individuals; />Is the maximum value of the fitness; />To adapt toA degree average value;k 2 、k 3 is a constant value, and is used for the treatment of the skin,k 2 ∈[0,0.001]，k 3 =0.2；
3-3: the first 20% of the offspring solution set generated in the later stage of the genetic algorithm is selected to carry out a simulated annealing operation stage, and the improved metropolis criterion used by the second-layer simulated annealing algorithm refers to the formulas (3) and (4) and the cooling factoralphaIs provided as formula (5) and formula (6); the second layer of simulated annealing algorithm is used for avoiding the improvement of local optimum and optimizing the optimum solution value;
（3）
（4）
wherein,E t+1 a new energy state is indicated and a new energy state is indicated,E t indicating the state of energy at this time,T m the adjustment coefficient representing the acceptance criterion is used,Trepresenting a preset initial temperature;E(x max ) AndE(x min ) Representing the maximum and minimum values of the objective function corresponding to the N possible solutions randomly selected from the first 20% of the solution set of the previous genetic algorithm,T 0 indicating an initial temperature;
（5）
（6）
wherein,T s indicating the initial set-up temperature of the device,krepresenting the number of current iterations and,mthe method is characterized in that the method comprises the steps of pre-fixing a value, changing times corresponding to intersection points of a logarithmic function and an exponential function, wherein DeltaN represents a ratio of total times of whether new solution links are accepted or not and times smaller than the current optimal solution are accepted by a metropolis rule;
3-4: detecting whether the iteration times exceeds the maximum iteration times GEN, returning to the step 3-3 when the iteration times are less than or equal to the GEN, otherwise, performing the step 3-5;
3-5: reading the job sequence in the Chrom corresponding to the shortest scheduling path, and recording the length of the shortest scheduling path;
3-6: the algorithm ends.
5. The method for multi-field block airline scheduling planning for large-area hilly mountainous areas according to claim 4, wherein:
after obtaining an optimal route sequence of scheduling and planning of an operation area, determining the number of times of setting up each partition through a formula (7), establishing a relation between the unmanned aerial vehicle route and a plant protection area according to the allocation of the times of setting up the partition, and calculating the load quantity required by each setting up through a formula (8); the spraying amount per hectare is set by determining the severity of plant diseases and insect pests or the fertilizer requirement of crops in advanceωUnmanned aerial vehicle operation speedvSpray breadth of plant protection unmanned aerial vehiclewThe method comprises the steps of carrying out a first treatment on the surface of the Then the unmanned aerial vehicle is always voyageR j Unmanned aerial vehicle work areaSAmount of sprayingQThe method comprises the following steps of:
（7）
in the method, in the process of the invention,R j to the firstjThe total course of farmland plant protection operation during the setting up;R 1 representing the total course of farmland plant protection operation at the 1 st time;nrepresenting the last frame; when the coverage algorithm obtains the optimal full coverage route, obtaining the full coverage total range of each zone, and obtaining the workload of each plant protection zone according to the formula (7);
（8）
in the formula (8), the amino acid sequence of the compound,L j is the firstj The load amount required by the frame time;S j to the theoretical firstj The total area of the operation is completed for the time;S j-1 to the theoretical firstj-1 total area of completed jobs;Fis the spraying flow.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311508910.6A CN117235558B (en) | 2023-11-14 | 2023-11-14 | Multi-field block route scheduling planning method for large-area hilly and mountainous areas |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311508910.6A CN117235558B (en) | 2023-11-14 | 2023-11-14 | Multi-field block route scheduling planning method for large-area hilly and mountainous areas |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117235558A true CN117235558A (en) | 2023-12-15 |
| CN117235558B CN117235558B (en) | 2024-03-19 |
Family
ID=89088441
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311508910.6A Active CN117235558B (en) | 2023-11-14 | 2023-11-14 | Multi-field block route scheduling planning method for large-area hilly and mountainous areas |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117235558B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109090076A (en) * | 2018-06-14 | 2018-12-28 | 华南农业大学 | A kind of unmanned plane sprays planing method |
| CN113159459A (en) * | 2021-05-21 | 2021-07-23 | 南京林业大学 | Multi-forest-area air route scheduling planning method based on fusion algorithm |
| CN113159369A (en) * | 2021-01-21 | 2021-07-23 | 安徽农业大学 | Optimized genetic algorithm-based multi-forest-area scheduling route planning method |
-
2023
- 2023-11-14 CN CN202311508910.6A patent/CN117235558B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109090076A (en) * | 2018-06-14 | 2018-12-28 | 华南农业大学 | A kind of unmanned plane sprays planing method |
| CN113159369A (en) * | 2021-01-21 | 2021-07-23 | 安徽农业大学 | Optimized genetic algorithm-based multi-forest-area scheduling route planning method |
| CN113159459A (en) * | 2021-05-21 | 2021-07-23 | 南京林业大学 | Multi-forest-area air route scheduling planning method based on fusion algorithm |
Non-Patent Citations (1)
| Title |
|---|
| 方树平 等: "基于VMM和DNiGA的直升机多林区植保作业航线规划", 农业工程学报, pages 70 - 77 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117235558B (en) | 2024-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Huang et al. | Large scale real-time ridesharing with service guarantee on road networks | |
| CN112016812B (en) | Multi-unmanned aerial vehicle task scheduling method, system and storage medium | |
| CN110308740A (en) | A kind of unmanned aerial vehicle group dynamic task allocation method towards mobile target tracking | |
| CN107515003B (en) | Method for planning flight route of airplane for patrolling power transmission line | |
| CN112925350B (en) | Multi-unmanned aerial vehicle distributed cooperative target searching method | |
| CN107123315A (en) | A kind of termination environment for considering ambient influnence is entered to leave the theatre Route optimization method | |
| CN113759971B (en) | Unmanned plane collaborative reconnaissance-oriented path planning method | |
| CN113139678A (en) | Unmanned aerial vehicle-vehicle combined distribution path optimization method and model construction method thereof | |
| CN107067824A (en) | Termination environment Route Network optimization method based on ambient influnence | |
| CN113131985B (en) | Multi-unmanned-aerial-vehicle data collection method based on information age optimal path planning | |
| CN113159459B (en) | Multi-forest-area air route scheduling planning method based on fusion algorithm | |
| CN110531782A (en) | Unmanned aerial vehicle flight path paths planning method for community distribution | |
| CN115730425A (en) | Unmanned aerial vehicle cluster complex area balanced coverage method, system, storage medium and terminal based on improved spanning tree | |
| Yu et al. | Balanced multi-region coverage path planning for unmanned aerial vehicles | |
| CN117235558B (en) | Multi-field block route scheduling planning method for large-area hilly and mountainous areas | |
| Gao et al. | Ant colony optimization for UAV-based intelligent pesticide irrigation system | |
| CN113703488A (en) | Multi-operation plant protection unmanned aerial vehicle path planning method based on improved ant colony algorithm | |
| CN115857543A (en) | Multi-unmanned aerial vehicle collaborative search method based on prior information | |
| CN109447471B (en) | Multi-robot multi-region collaborative search task allocation method | |
| CN114779820A (en) | Multi-destination unmanned aerial vehicle real-time flight path planning method with intelligent decision | |
| CN109726895B (en) | Multi-target-point task execution planning method and device | |
| CN111076727A (en) | Unmanned aerial vehicle orthographic multi-frame operation track planning algorithm | |
| Cao et al. | Path optimization of joint delivery mode of trucks and UAVs | |
| Zhang et al. | Cooperative Route Planning for Fuel-constrained UGV-UAV Exploration | |
| Zhu et al. | Data Collection in Wireless Sensor Networks: A Truck-Assisted Multi-UAV Method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |