CN104978607A - Greedy genetic algorithm-based pot seedling thin planting and transplantation path optimization method - Google Patents
Greedy genetic algorithm-based pot seedling thin planting and transplantation path optimization method Download PDFInfo
- Publication number
- CN104978607A CN104978607A CN201510346330.0A CN201510346330A CN104978607A CN 104978607 A CN104978607 A CN 104978607A CN 201510346330 A CN201510346330 A CN 201510346330A CN 104978607 A CN104978607 A CN 104978607A
- Authority
- CN
- China
- Prior art keywords
- cave
- genetic algorithm
- coding
- dish
- population
- 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
- 230000002068 genetic effect Effects 0.000 title claims abstract description 46
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000005457 optimization Methods 0.000 title claims abstract description 12
- 238000002054 transplantation Methods 0.000 title abstract 9
- 210000000349 chromosome Anatomy 0.000 claims description 15
- 230000002759 chromosomal effect Effects 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 6
- 239000003550 marker Substances 0.000 claims description 6
- 230000013011 mating Effects 0.000 claims description 6
- 238000001467 acupuncture Methods 0.000 claims description 5
- 230000035772 mutation Effects 0.000 claims description 5
- 238000005295 random walk Methods 0.000 claims description 5
- 108090000623 proteins and genes Proteins 0.000 claims description 3
- 235000019994 cava Nutrition 0.000 claims description 2
- 238000012248 genetic selection Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 4
- 239000012636 effector Substances 0.000 description 4
- 238000001514 detection method Methods 0.000 description 2
- 230000003862 health status Effects 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
Abstract
The invention discloses a greedy genetic algorithm-based pot seedling thin planting and transplantation path optimization method. Health information of pot seedlings in a transplantation hole tray of a greenhouse pot seedling thin planting and transplantation machine is obtained through machine vision, and healthy seedling hole positions in the transplantation hole tray and empty hole positions in a target hole tray are subjected to label coding respectively; a greedy genetic selection principle is that the empty holes in the target hole tray are partitioned by column for carrying out current path optimization of a local genetic algorithm; the coding of a column of empty holes in the target hole tray is integrated with the coding of unplanned seedling holes in the transplantation hole tray, random path coding is generated, an initial population of the local genetic algorithm is formed, the selection, crossing, variation and reinsertion operations are circularly carried out until preset convergence generations, and individuals with maximum population fitness serve as local optimal paths; and the successive columns of planned local optimal paths are combined to generate a thin planting and transplantation path of the whole target hole tray. According to the method, an optimal path for greenhouse pot seedling thin planting and transplantation can be generated, the transplantation efficiency can be improved, and the real-time planning requirements of a control system are met.
Description
Technical field
The present invention relates to the method for transplanting for pot seedling in agricultural machinery, especially relate to a kind of pot seedling thin planting based on Greedy genetic algorithm and transplant method for optimizing route.
Background technology
In greenhouse hole plate seedling growth, the pot seedling in the dish of high density cave needs thin planting to be transplanted in the dish of low-density cave, and the planting percent simultaneously in the dish of cave is between 80-95%.Tradition transplants operation by manually carrying out identification, and efficiency is low, and labour intensity is large, and the consistance of transplanted seedling is bad; Greenhouse pot seedling thin planting transplanter, by Machine Vision Detection pot seedling health status and positional information, is captured by end effector and transplants, can solve the problem.
Disk hole cave, low-density cave is many, the randomness of the interior healthy seedling present position of high density cave dish is large, and the priority selectivity that transplanter control end effector is transplanted from initial point to each cavities is many, namely carries out the variable-length that thin planting transplants path; Because hole data volume is large, the method for the shortest optimal path of controller calculating sifting cannot meet the requirement of realtime control.The paths planning method of this kind of pot seedling thin planting transplanting has to be developed.
Summary of the invention
The object of the present invention is to provide a kind of pot seedling thin planting based on Greedy genetic algorithm to transplant method for optimizing route, the travel distance of greenhouse pot seedling thin planting transplanter end effector can be reduced, improve operating efficiency.
In order to achieve the above object, the technical solution used in the present invention is:
By machine vision, oneself knows the health and fitness information of the interior pot seedling of transplanting cave dish of greenhouse pot seedling thin planting transplanter in the present invention, carries out label coding respectively to healthy seedling acupuncture point and position, Pan Nei hole, object cave in the dish of transplanting cave; Greed heredity selects the cavities of circling or whirl in the air of cave for the purpose of excellent principle to carry out the current path optimization of Local genetic algorithm by row subregion; Object cave coil certain row Hole coding with transplant cave coil in do not plan have seedling cavities coding comprehensive, generate the initial population that random walk coding forms Local genetic algorithm, circulation is carried out selecting, intersect, make a variation and heavy update until preset convergence times, using maximum for population's fitness individuality as this local optimum path; The local optimum path of successively each row planning is merged, namely generates whole object cave dish thin planting and transplant path.
Describedly respectively label coding is carried out to healthy seedling acupuncture point and position, Pan Nei hole, object cave in the dish of transplanting cave, the each cavities being specially the high transplanting cave dish of density and low density object cave dish is fixed in the position of transplanter system, by order from top to bottom, from left to right, arithmetic number mark is carried out to healthy seedling cavities in the dish of transplanting cave, carry out negative real number mark to each Hole in the dish of object cave by order from top to bottom, from left to right, label coding is actual is thus implied with cavities position and pot seedling health and fitness information.
Described object cave cavities of circling or whirl in the air carries out the current path optimization of Local genetic algorithm by row subregion, be specially object cave to circle or whirl in the air cavities negative flag coding hole pressing dish row subregion, by from left to right or dextrosinistral row order, that does not plan in successively coiling with transplanting cave has seedling cavities negative flag coding comprehensive, carries out the current optimum path planning of Local genetic algorithm.
Described object cave coil certain row Hole coding with transplant cave coil in do not plan have seedling cavities coding comprehensive, generate the initial population that random walk coding forms Local genetic algorithm, circulation is carried out selecting, intersect, make a variation and heavy update until preset convergence times, using maximum for population's fitness individuality as this local optimum path, concrete Local genetic algorithm process prescription is as follows:
A) the initial population generation method of Local genetic algorithm is specially: the negative flag coded set that certain row Hole is coiled in hypothesis goal cave is {-1,-2,-3,-4,-5,-6,-7,-8}, that does not plan in the dish of transplanting cave has the positive label coding collection of seedling cavities to be { 1, 2, 3, 48, 49, 50}, then transplant path to intersect at random from initial point and positive and negative label coding collection, can be formed as (0, 3,-2, 8,-4, 9,-7, 10,-1, 7,-6, 13,-3, 5,-5, 16,-8, 0) item chromosome of initial population, algorithm arranges and generates some chromosome, namely initial population is formed,
B) individual population's fitness is specially: the position of the transplanting coordinate system of the actual mapping of the coding in each chromosome is known, then the path of concrete each chromosomal mapping also can calculate, and is set to
l(
x), wherein
l minwith
l maxrepresent population the shortest chromosomal and longest path respectively; Defining individual population's fitness is
c=(
l max-
l(
x))/(
l max-
l min);
C) the Local genetic algorithm selection carried out that circulates is operating as: initial population as parent, according to random sequence, with chromosome fitness for select probability, select probability
m>
cchromosome as progeny population;
D) the Local genetic algorithm interlace operation carried out that circulates is: after initial population selects operation to generate progeny population, randomly ordered, carries out interlace operation; 1. hypothesis has O=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,19 ,-3,13 ,-6,15 ,-8,26 ,-5,0), P=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,7 ,-5,28 ,-8,5 ,-6,16 ,-3,0), Q=(0,12 ,-6,30 ,-3,19 ,-8,11 ,-1,16 ,-2,5 ,-4,14 ,-5,10 ,-7,0) three progeny population chromosomes; 2. produce 2 random number j between 1 to 16 and k, wherein j is as mating indicating bit, and k is as mating step-length, then three chromosomal jth of filial generation+1 are gone forward one by one to j+k position exchange; If j+k>=16, order is taken as 16; If j=8, k=4, can obtain intersecting rear individuality: O
1=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,16 ,-2,5 ,-6,15 ,-8,26 ,-5,0), P
1=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,19 ,-3,13 ,-8,5 ,-6,16 ,-3,0), Q
1=(0,12 ,-6,30 ,-3,19 ,-8,11 ,-1,7 ,-5,28 ,-4,14 ,-5,10 ,-7,0); 3. scanning is except position after the intersection of initial point 0, if identical, with 800 replacements, O
2=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,16,800,5 ,-6,15 ,-8,26 ,-5,0), P
2=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,19 ,-3,800 ,-8,5 ,-6,16,800,0), Q
2=(0,12 ,-6,30 ,-3,19 ,-8,11 ,-1,7 ,-5,28 ,-4,14,800,10 ,-7,0); 4. scanning is except the previous step individuality of initial point 0, uses an effectively position to replace several 800 successively; If several 800 are in even bit, then these row all marker number order scanning in the dish of object cave are contrasted except each even bit after initial point 0 with a position, if do not occur, then replace 800 with this; If several 800 are in odd bits, then there is the scanning of seedling marker number order to contrast except each odd-even bit after initial point 0 with a position by transplanting do not plan in the dish of cave all, the mark do not occurred is produced one to replace 800 at random; O can be obtained
3=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,16 ,-3,5 ,-6,15 ,-8,26 ,-5,0), P
3=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,19 ,-3,28 ,-8,5 ,-6,16 ,-5,0), Q
3=(0,12 ,-6,30 ,-3,19 ,-8,11 ,-1,7 ,-5,28 ,-4,14 ,-2,10 ,-7,0);
E) the Local genetic algorithm mutation operation carried out that circulates is: the population produced above-mentioned interlace operation is randomly ordered, carries out mutation operation; Produce 2 random number r between 1 to 16 and s, as 2 the variation positions of individuality except initial point 0:
If the label coding of variation position is negative, then from object cave other negative flag of dish row coding Stochastic choice one, from variation individuality, scanning finds this value to exchange with variation position: suppose r=2, offspring individual O
3=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,16 ,-3,5 ,-6,15 ,-8,26 ,-5,0), then except-4 reference numerals-1 ,-2 ,-3 ,-5, produce a random number in-6 ,-7 ,-8}, be set to-7, then new offspring individual O after variation
4=(0,14 ,-7,9 ,-2,8 ,-4,12 ,-1,16 ,-3,5 ,-6,15 ,-8,26 ,-5,0);
If the label coding of variation position is just, then there is the positive label coding Stochastic choice of seedling one to exchange from transplanting do not plan in the dish of cave all, and travel through the offspring individual after gene replacement, if there is same positive label coding, then replace with former label coding number: suppose s=9, offspring individual P
3=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,19 ,-3,28 ,-8,5 ,-6,16 ,-5,0), then except 19 reference numerals 1,2,3 ..., produce a random number in 48,49,50}, be set to 20, then new offspring individual P after variation
3=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,20 ,-3,28 ,-8,5 ,-6,16 ,-5,0);
F) the Local genetic algorithm heavy update carried out that circulates is: to above-mentioned initial population through selecting, intersecting, the offspring individual that produces after variation carries out fitness calculating, heavily inserting initial population replaces the most unconformable parent individual, keeps initial population scale;
Local genetic algorithm, by selecting above-mentioned initial population circulation, intersecting, make a variation and heavy update, arrives default convergence times and stops, getting the maximum individuality of population's fitness in this convergence generation as this local optimum path.
Until all object caves coil each row local optimum path obtain after, according to from left to right or dextrosinistral row order merge, generate whole object cave dish thin planting transplant path.
The beneficial effect that the present invention has is:
The present invention obtains high density by Machine Vision Detection and transplants cave dish pot seedling health status and positional information, utilization Greedy genetic algorithm is technological means, complete rapid Optimum high density and transplant cave dish pot seedling to dish thin planting transplanting path, low-density object cave, thus improve end effector operating efficiency.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the pot seedling thin planting transplanting method for optimizing route that the present invention is based on Greedy genetic algorithm.
Fig. 2 is the label coding figure transplanting cave dish and object cave dish.
Fig. 3 optimizes algebraically variation diagram in the genetic optimization of local.
Fig. 4 is that single-row cavities local genetic optimization result is coiled in object cave.
Fig. 5 is that after each row order synthesis, thin planting transplants path profile.
Embodiment
Below in conjunction with method flow diagram and embodiment, the invention will be further described.
Method flow diagram of the present invention is as shown in Figure 1: the present invention knows the health and fitness information of pot seedling in the dish of cave by machine vision, carries out label coding (as shown in Figure 2) respectively to healthy seedling acupuncture point and position, Pan Nei hole, object cave in the dish of transplanting cave.Greed heredity selects the cavities of circling or whirl in the air of cave for the purpose of excellent principle to carry out the current path optimization of Local genetic algorithm by row subregion; Object cave coil certain row Hole coding with transplant cave coil in do not plan have seedling cavities coding comprehensive, generate the initial population that random walk coding forms Local genetic algorithm, circulation is carried out selecting, intersect, make a variation and heavy update until preset convergence times (as shown in Figure 3), using maximum for population's fitness individuality as this local optimum path (as shown in Figure 4); The local optimum path of successively each row planning is merged, namely generates whole object cave dish thin planting and transplant path (as shown in Figure 5).
P1: each cavities of the transplanting cave dish that density is high and low density object cave dish is fixed in the position of transplanter system, transplant pot seedling health and fitness information in the dish of cave also to be obtained by machine vision, by order from top to bottom, from left to right, arithmetic number mark is carried out to healthy seedling cavities in the dish of transplanting cave, carry out negative real number mark to each Hole in the dish of object cave by order from top to bottom, from left to right, label coding is actual is thus implied with cavities position and pot seedling health and fitness information.As shown in Figure 2, object cave dish 4 × 8 specifications, Hole label coding collection be-1 ,-2 ,-3 ... ,-32}; Transplant cave dish 5 × 10 specifications, containing the strain of healthy pot seedling 42, label coding collection be 1,2,3 ..., 40,41,42}.
P2: object cave cavities of circling or whirl in the air carries out the current optimum path planning of genetic algorithm by row subregion, be specially object cave to circle or whirl in the air cavities negative flag coding hole pressing dish row subregion, by from left to right or dextrosinistral row order, that does not plan in successively coiling with transplanting cave has seedling cavities negative flag coding comprehensive, carries out the initial population of Local genetic algorithm optimization.Model is transplanted, object cave dish left column label coding {-1 ,-2 ,-3 ,-4 ,-5 ,-6 for Fig. 2 thin planting,-7 ,-8}, transplant cave dish current do not planned the label coding of seedling cavities 1,2,3 ... 40,41,42}, from initial point, mutually random cross-synthesis, can be formed as (0,3,-2,8 ,-4,9 ,-7,10,-1,7 ,-6,13 ,-3,5,-5,16 ,-8,0) item chromosome of initial population, arranges and generates some chromosome, namely form initial population.
P3: the position of the transplanting coordinate system of the actual mapping of the coding in each chromosome is known, then the path of concrete each chromosomal mapping also can calculate, and is set to
l(
x), wherein
l minwith
l maxrepresent population the shortest chromosomal and longest path respectively; Defining individual population's fitness is
c=(
l max-
l(
x))/(
l max-
l min).Initial population as parent, according to random sequence, with chromosome fitness for select probability, select probability
m>
cchromosome as progeny population.
P4: after initial population selects operation to generate progeny population, randomly ordered, carries out interlace operation.1. hypothesis has O=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,19 ,-3,13 ,-6,15 ,-8,26 ,-5,0), P=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,7 ,-5,28 ,-8,5 ,-6,16 ,-3,0), Q=(0,12 ,-6,30 ,-3,19 ,-8,11 ,-1,16 ,-2,5 ,-4,14 ,-5,10 ,-7,0) three progeny population chromosomes; 2. produce 2 random number j between 1 to 16 and k, wherein j is as mating indicating bit, and k is as mating step-length, then three chromosomal jth of filial generation+1 are gone forward one by one to j+k position exchange; If j+k>=16, order is taken as 16; If j=8, k=4, can obtain intersecting rear individuality: O
1=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,16 ,-2,5 ,-6,15 ,-8,26 ,-5,0), P
1=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,19 ,-3,13 ,-8,5 ,-6,16 ,-3,0), Q
1=(0,12 ,-6,30 ,-3,19 ,-8,11 ,-1,7 ,-5,28 ,-4,14 ,-5,10 ,-7,0); 3. scanning is except position after the intersection of initial point 0, if identical, with 800 replacements, O
2=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,16,800,5 ,-6,15 ,-8,26 ,-5,0), P
2=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,19 ,-3,800 ,-8,5 ,-6,16,800,0), Q
2=(0,12 ,-6,30 ,-3,19 ,-8,11 ,-1,7 ,-5,28 ,-4,14,800,10 ,-7,0); 4. scanning is except the previous step individuality of initial point 0, uses an effectively position to replace several 800 successively; If several 800 are in even bit, then these row all marker number order scanning in the dish of object cave are contrasted except each even bit after initial point 0 with a position, if do not occur, then replace 800 with this; If several 800 are in odd bits, then there is the scanning of seedling marker number order to contrast except each odd-even bit after initial point 0 with a position by transplanting do not plan in the dish of cave all, the mark do not occurred is produced one to replace 800 at random; O can be obtained
3=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,16 ,-3,5 ,-6,15 ,-8,26 ,-5,0), P
3=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,19 ,-3,28 ,-8,5 ,-6,16 ,-5,0), Q
3=(0,12 ,-6,30 ,-3,19 ,-8,11 ,-1,7 ,-5,28 ,-4,14 ,-2,10 ,-7,0).
P5: the population produced above-mentioned interlace operation is randomly ordered, carries out mutation operation.Produce 2 random number r between 1 to 16 and s, as 2 the variation positions of individuality except initial point 0:
If the label coding of variation position is negative, then from object cave other negative flag of dish row coding Stochastic choice one, from variation individuality, scanning finds this value to exchange with variation position: suppose r=2, offspring individual O
3=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,16 ,-3,5 ,-6,15 ,-8,26 ,-5,0), then except-4 reference numerals-1 ,-2 ,-3 ,-5, produce a random number (being set to-7) in-6 ,-7 ,-8}, then new offspring individual O after variation
4=(0,14 ,-7,9 ,-2,8 ,-4,12 ,-1,16 ,-3,5 ,-6,15 ,-8,26 ,-5,0);
If the label coding of variation position is just, then there is the positive label coding Stochastic choice of seedling one to exchange from transplanting do not plan in the dish of cave all, and travel through the offspring individual after gene replacement, if there is same positive label coding, then replace with former label coding number: suppose s=9, offspring individual P
3=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,19 ,-3,28 ,-8,5 ,-6,16 ,-5,0), then except 19 reference numerals 1,2,3 ..., produce a random number (being set to 20) in 48,49,50}, then new offspring individual P after variation
3=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,20 ,-3,28 ,-8,5 ,-6,16 ,-5,0).
P6: the Local genetic algorithm heavy update carried out that circulates is: to above-mentioned initial population through selecting, intersecting, the offspring individual that produces after variation carries out fitness calculating, heavily insert initial population and replace the most unconformable parent individual, keep initial population scale.
P7: Local genetic algorithm judges whether that arriving default convergence times stops, and does not arrive, carries out selection, intersection, the variation of P3 to P6 to population circulation and weighs update; If arrive, enter next step P8.
P8: arrive and preset convergence times stopping, and defeated also optimum results.As shown in Figure 3, got for the 70th generation as convergence times, the maximum individuality of population's fitness is (0,4 ,-4,11 ,-3,2 ,-1,1 ,-2,3 ,-5,5 ,-8,7 ,-7,6 ,-6,0) as this local optimum path, as shown in Figure 4.
P9: judge that object cave is coiled each row and whether all completed local path optimization, if do not complete, circulates in order by column from P2 step, if Fig. 4 is by order from left to right; If each row all complete, enter next step P10.
P10: the local optimum path merging all row, the thin planting generating whole object cave dish transplants path.As shown in Figure 5, merge way is (0,4 ,-4,11 ,-3,2 ,-1,1 ,-2,3 ,-5,5 ,-8,7 ,-7,6 ,-6,14 ,-15,15 ,-16,22,-13,18 ,-9,9 ,-10,8 ,-11,10 ,-12,12 ,-14,23 ,-24,24 ,-23,13 ,-21,19 ,-18,17 ,-20,30,-22,21 ,-19,16 ,-17,36 ,-30,25 ,-31,33 ,-29,20 ,-28,29 ,-27,28 ,-25,27 ,-26,31 ,-32,0).
Claims (5)
1. the pot seedling thin planting based on Greedy genetic algorithm transplants a method for optimizing route, and by machine vision, oneself knows the health and fitness information of the interior pot seedling of transplanting cave dish of greenhouse pot seedling thin planting transplanter; It is characterized in that: respectively label coding is carried out to healthy seedling acupuncture point and position, Pan Nei hole, object cave in the dish of transplanting cave; Greed heredity selects the cavities of circling or whirl in the air of cave for the purpose of excellent principle to carry out the current path optimization of Local genetic algorithm by row subregion; Object cave coil certain row Hole coding with transplant cave coil in do not plan have seedling cavities coding comprehensive, generate the initial population that random walk coding forms Local genetic algorithm, circulation is carried out selecting, intersect, make a variation and heavy update until preset convergence times, using maximum for population's fitness individuality as this local optimum path; The local optimum path of successively each row planning is merged, namely generates whole object cave dish thin planting and transplant path.
2. a kind of pot seedling thin planting based on Greedy genetic algorithm according to claim 1 transplants method for optimizing route, it is characterized in that: describedly respectively label coding is carried out to healthy seedling acupuncture point and position, Pan Nei hole, object cave in the dish of transplanting cave, the each cavities being specially the high transplanting cave dish of density and low density object cave dish is fixed in the position of transplanter system, healthy seedling cavities in the dish of transplanting cave is pressed from top to bottom, order from left to right carries out arithmetic number mark, each Hole in the dish of object cave is pressed from top to bottom, order from left to right carries out negative real number mark, label coding is actual is thus implied with cavities position and pot seedling health and fitness information.
3. a kind of pot seedling thin planting based on Greedy genetic algorithm according to claim 1 transplants method for optimizing route, it is characterized in that: described object cave cavities of circling or whirl in the air carries out the current path optimization of Local genetic algorithm by row subregion, be specially object cave to circle or whirl in the air cavities negative flag coding hole pressing dish row subregion, by from left to right or dextrosinistral row order, that does not plan in successively coiling with transplanting cave has seedling cavities negative flag coding comprehensive, carries out the current optimum path planning of Local genetic algorithm.
4. a kind of pot seedling thin planting based on Greedy genetic algorithm according to claim 1 transplants method for optimizing route, it is characterized in that: described object cave coil certain row Hole coding with transplant cave coil in do not plan have seedling cavities coding comprehensive, generate the initial population that random walk coding forms Local genetic algorithm, circulation is carried out selecting, intersect, make a variation and heavy update until preset convergence times, using maximum for population's fitness individuality as this local optimum path, concrete Local genetic algorithm process prescription is as follows:
A) the initial population generation method of Local genetic algorithm is specially: the negative flag coded set that certain row Hole is coiled in hypothesis goal cave is {-1,-2,-3,-4,-5,-6,-7,-8}, that does not plan in the dish of transplanting cave has the positive label coding collection of seedling cavities to be { 1, 2, 3, 48, 49, 50}, then transplant path to intersect at random from initial point and positive and negative label coding collection, can be formed as (0, 3,-2, 8,-4, 9,-7, 10,-1, 7,-6, 13,-3, 5,-5, 16,-8, 0) item chromosome of initial population, algorithm arranges and generates some chromosome, namely initial population is formed,
B) individual population's fitness is specially: the position of the transplanting coordinate system of the actual mapping of the coding in each chromosome is known, then the path of concrete each chromosomal mapping also can calculate, and is set to
l(
x), wherein
l minwith
l maxrepresent population the shortest chromosomal and longest path respectively; Defining individual population's fitness is
c=(
l max-
l(
x))/(
l max-
l min);
C) the Local genetic algorithm selection carried out that circulates is operating as: initial population as parent, according to random sequence, with chromosome fitness for select probability, select probability
m>
cchromosome as progeny population;
D) the Local genetic algorithm interlace operation carried out that circulates is: after initial population selects operation to generate progeny population, randomly ordered, carries out interlace operation; 1. hypothesis has O=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,19 ,-3,13 ,-6,15 ,-8,26 ,-5,0), P=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,7 ,-5,28 ,-8,5 ,-6,16 ,-3,0), Q=(0,12 ,-6,30 ,-3,19 ,-8,11 ,-1,16 ,-2,5 ,-4,14 ,-5,10 ,-7,0) three progeny population chromosomes; 2. produce 2 random number j between 1 to 16 and k, wherein j is as mating indicating bit, and k is as mating step-length, then three chromosomal jth of filial generation+1 are gone forward one by one to j+k position exchange; If j+k>=16, order is taken as 16; If j=8, k=4, can obtain intersecting rear individuality: O
1=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,16 ,-2,5 ,-6,15 ,-8,26 ,-5,0), P
1=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,19 ,-3,13 ,-8,5 ,-6,16 ,-3,0), Q
1=(0,12 ,-6,30 ,-3,19 ,-8,11 ,-1,7 ,-5,28 ,-4,14 ,-5,10 ,-7,0); 3. scanning is except position after the intersection of initial point 0, if identical, with 800 replacements, O
2=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,16,800,5 ,-6,15 ,-8,26 ,-5,0), P
2=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,19 ,-3,800 ,-8,5 ,-6,16,800,0), Q
2=(0,12 ,-6,30 ,-3,19 ,-8,11 ,-1,7 ,-5,28 ,-4,14,800,10 ,-7,0); 4. scanning is except the previous step individuality of initial point 0, uses an effectively position to replace several 800 successively; If several 800 are in even bit, then these row all marker number order scanning in the dish of object cave are contrasted except each even bit after initial point 0 with a position, if do not occur, then replace 800 with this; If several 800 are in odd bits, then there is the scanning of seedling marker number order to contrast except each odd-even bit after initial point 0 with a position by transplanting do not plan in the dish of cave all, the mark do not occurred is produced one to replace 800 at random; O can be obtained
3=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,16 ,-3,5 ,-6,15 ,-8,26 ,-5,0), P
3=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,19 ,-3,28 ,-8,5 ,-6,16 ,-5,0), Q
3=(0,12 ,-6,30 ,-3,19 ,-8,11 ,-1,7 ,-5,28 ,-4,14 ,-2,10 ,-7,0);
E) the Local genetic algorithm mutation operation carried out that circulates is: the population produced above-mentioned interlace operation is randomly ordered, carries out mutation operation; Produce 2 random number r between 1 to 16 and s, as 2 the variation positions of individuality except initial point 0:
If the label coding of variation position is negative, then from object cave other negative flag of dish row coding Stochastic choice one, from variation individuality, scanning finds this value to exchange with variation position: suppose r=2, offspring individual O
3=(0,14 ,-4,9 ,-2,8 ,-7,12 ,-1,16 ,-3,5 ,-6,15 ,-8,26 ,-5,0), then except-4 reference numerals-1 ,-2 ,-3 ,-5, produce a random number in-6 ,-7 ,-8}, be set to-7, then new offspring individual O after variation
4=(0,14 ,-7,9 ,-2,8 ,-4,12 ,-1,16 ,-3,5 ,-6,15 ,-8,26 ,-5,0);
If the label coding of variation position is just, then there is the positive label coding Stochastic choice of seedling one to exchange from transplanting do not plan in the dish of cave all, and travel through the offspring individual after gene replacement, if there is same positive label coding, then replace with former label coding number: suppose s=9, offspring individual P
3=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,19 ,-3,28 ,-8,5 ,-6,16 ,-5,0), then except 19 reference numerals 1,2,3 ..., produce a random number in 48,49,50}, be set to 20, then new offspring individual P after variation
3=(0,13 ,-7,8 ,-4,9 ,-2,10 ,-1,20 ,-3,28 ,-8,5 ,-6,16 ,-5,0);
F) the Local genetic algorithm heavy update carried out that circulates is: to above-mentioned initial population through selecting, intersecting, the offspring individual that produces after variation carries out fitness calculating, heavily inserting initial population replaces the most unconformable parent individual, keeps initial population scale;
Local genetic algorithm, by selecting above-mentioned initial population circulation, intersecting, make a variation and heavy update, arrives default convergence times and stops, getting the maximum individuality of population's fitness in this convergence generation as this local optimum path.
5. a kind of pot seedling thin planting based on Greedy genetic algorithm according to claim 1 transplants method for optimizing route, it is characterized in that: after the local optimum path acquisition that each row are coiled in all object caves, according to from left to right or dextrosinistral row order merge, generate whole object cave dish thin planting transplant path.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510346330.0A CN104978607B (en) | 2015-06-23 | 2015-06-23 | A kind of pot seedling thin planting transplanting method for optimizing route based on Greedy genetic algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510346330.0A CN104978607B (en) | 2015-06-23 | 2015-06-23 | A kind of pot seedling thin planting transplanting method for optimizing route based on Greedy genetic algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104978607A true CN104978607A (en) | 2015-10-14 |
| CN104978607B CN104978607B (en) | 2018-04-20 |
Family
ID=54275089
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510346330.0A Active CN104978607B (en) | 2015-06-23 | 2015-06-23 | A kind of pot seedling thin planting transplanting method for optimizing route based on Greedy genetic algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104978607B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106598043A (en) * | 2016-11-08 | 2017-04-26 | 中国科学院自动化研究所 | High-speed pickup path optimizing method of parallel robots facing dynamic objects |
| CN107145961A (en) * | 2017-03-24 | 2017-09-08 | 南京邮电大学 | A kind of tour schedule planing method based on improved adaptive GA-IAGA |
| CN109459052A (en) * | 2018-12-28 | 2019-03-12 | 上海大学 | A kind of sweeper complete coverage path planning method |
| CN110889552A (en) * | 2019-11-26 | 2020-03-17 | 中国计量大学 | Automatic apple boxing path optimization method based on optimal parameter genetic algorithm |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102870534A (en) * | 2012-09-21 | 2013-01-16 | 浙江大学 | Genetic algorithm based automatic bowl transporting route optimization method of plug seedlings |
| CN102939815A (en) * | 2012-10-11 | 2013-02-27 | 江苏大学 | Seedling taking and seedling planting path planning for pot seedling transplanting robot |
| CN103955753A (en) * | 2014-04-14 | 2014-07-30 | 浙江大学 | Automatic plug seedling transplanting path optimization method based on artificial ant colony |
-
2015
- 2015-06-23 CN CN201510346330.0A patent/CN104978607B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102870534A (en) * | 2012-09-21 | 2013-01-16 | 浙江大学 | Genetic algorithm based automatic bowl transporting route optimization method of plug seedlings |
| CN102939815A (en) * | 2012-10-11 | 2013-02-27 | 江苏大学 | Seedling taking and seedling planting path planning for pot seedling transplanting robot |
| CN103955753A (en) * | 2014-04-14 | 2014-07-30 | 浙江大学 | Automatic plug seedling transplanting path optimization method based on artificial ant colony |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106598043A (en) * | 2016-11-08 | 2017-04-26 | 中国科学院自动化研究所 | High-speed pickup path optimizing method of parallel robots facing dynamic objects |
| CN107145961A (en) * | 2017-03-24 | 2017-09-08 | 南京邮电大学 | A kind of tour schedule planing method based on improved adaptive GA-IAGA |
| CN109459052A (en) * | 2018-12-28 | 2019-03-12 | 上海大学 | A kind of sweeper complete coverage path planning method |
| CN110889552A (en) * | 2019-11-26 | 2020-03-17 | 中国计量大学 | Automatic apple boxing path optimization method based on optimal parameter genetic algorithm |
| CN110889552B (en) * | 2019-11-26 | 2023-05-23 | 中国计量大学 | Apple automatic boxing path optimization method based on optimal parameter genetic algorithm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104978607B (en) | 2018-04-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104978607A (en) | Greedy genetic algorithm-based pot seedling thin planting and transplantation path optimization method | |
| CN106779372B (en) | Agricultural machinery scheduling method based on improved immune taboo algorithm | |
| Jiang et al. | A simple and fast hypervolume indicator-based multiobjective evolutionary algorithm | |
| CN111562785A (en) | Path planning method and system for collaborative coverage of cluster robots | |
| CN101968860A (en) | Order sorting method and system | |
| CN106340061B (en) | A kind of mountain area point cloud filtering method | |
| Botzheim et al. | Bacterial memetic algorithm for offline path planning of mobile robots | |
| CN105426969A (en) | Game strategy generation method of non-complete information | |
| CN108803332A (en) | Based on the paths planning method for improving biogeography | |
| CN103279332A (en) | Data flow parallel processing method based on GPU-CUDA platform and genetic algorithm | |
| CN102870534B (en) | Genetic algorithm based automatic bowl transporting route optimization method of plug seedlings | |
| CN106611231A (en) | Hybrid particle swarm tabu search algorithm for solving job-shop scheduling problem | |
| CN108413963A (en) | Bar-type machine people's paths planning method based on self study ant group algorithm | |
| AR103075A1 (en) | IMPROVED METHODS OF MOLECULAR CULTURE | |
| CN104463328A (en) | Order crossover multi-filial-generation genetic algorithm for solving traveling salesman problem | |
| CN111427341B (en) | Robot shortest expected time target searching method based on probability map | |
| Akkus et al. | Automated land reallotment using genetic algorithm | |
| Xiao-Ting et al. | Flight path planning based on an improved genetic algorithm | |
| Beed et al. | A study of the genetic algorithm parameters for solving multi-objective travelling salesman problem | |
| CN115451974B (en) | Power equipment routing inspection path planning method, system, equipment and medium | |
| Cano | Parameters for a genetic algorithm: An application for the order batching problem | |
| CN112214885A (en) | Irrigation area crop irrigation water quantity optimal distribution method under insufficient irrigation condition | |
| CN116307214A (en) | Automatic cotton distribution method based on NSGA-II algorithm and related device | |
| CN112632777B (en) | II-type bilateral assembly line balancing method and system for household appliance assembly line | |
| CN103955753A (en) | Automatic plug seedling transplanting path optimization method based on artificial ant colony |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20151014 Assignee: ZHEJIANG KALUOLAINA WEAVING Co.,Ltd. Assignor: ZHEJIANG SCI-TECH University Contract record no.: X2021330000803 Denomination of invention: A greedy genetic algorithm based method for optimizing transplanting path of potted seedlings with sparse planting Granted publication date: 20180420 License type: Common License Record date: 20211215 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20151014 Assignee: Zhejiang Hongyuan Intelligent Technology Co.,Ltd. Assignor: ZHEJIANG SCI-TECH University Contract record no.: X2023330000751 Denomination of invention: A greedy genetic algorithm based method for optimizing the transplanting path of potted seedlings with sparse planting Granted publication date: 20180420 License type: Common License Record date: 20230928 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20151014 Assignee: Wenzhou Chaoxin Software Development Co.,Ltd. Assignor: ZHEJIANG SCI-TECH University Contract record no.: X2023330000834 Denomination of invention: A greedy genetic algorithm based method for optimizing the transplanting path of potted seedlings with sparse planting Granted publication date: 20180420 License type: Common License Record date: 20231109 Application publication date: 20151014 Assignee: GLOBALS ELECTRONICS (WENZHOU) Co.,Ltd. Assignor: ZHEJIANG SCI-TECH University Contract record no.: X2023330000820 Denomination of invention: A greedy genetic algorithm based method for optimizing the transplanting path of potted seedlings with sparse planting Granted publication date: 20180420 License type: Common License Record date: 20231107 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20151014 Assignee: Zhejiang Boshang Textile Co.,Ltd. Assignor: ZHEJIANG SCI-TECH University Contract record no.: X2024330000028 Denomination of invention: A greedy genetic algorithm based method for optimizing the transplanting path of bowl seedlings with sparse planting Granted publication date: 20180420 License type: Common License Record date: 20240403 |