CN117387634A - Color-changing wood forest zone unmanned aerial vehicle path multi-target planning method based on user preference - Google Patents
Color-changing wood forest zone unmanned aerial vehicle path multi-target planning method based on user preference Download PDFInfo
- Publication number
- CN117387634A CN117387634A CN202311707190.6A CN202311707190A CN117387634A CN 117387634 A CN117387634 A CN 117387634A CN 202311707190 A CN202311707190 A CN 202311707190A CN 117387634 A CN117387634 A CN 117387634A
- Authority
- CN
- China
- Prior art keywords
- reference vector
- unmanned aerial
- objective
- aerial vehicle
- initial reference
- 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 41
- 239000002023 wood Substances 0.000 title claims abstract description 24
- 239000013598 vector Substances 0.000 claims abstract description 97
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 34
- 238000013178 mathematical model Methods 0.000 claims abstract description 11
- 238000013461 design Methods 0.000 claims description 15
- 238000005457 optimization Methods 0.000 claims description 6
- 230000004083 survival effect Effects 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 abstract description 3
- 241000238631 Hexapoda Species 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 241000607479 Yersinia pestis Species 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
-
- 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
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Feedback Control In General (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a color-changing wood forest area unmanned aerial vehicle path multi-target planning method based on user preference, which comprises the steps of establishing a Maklink three-dimensional graph theory model of an unmanned aerial vehicle path environment; establishing a multi-objective optimized mathematical model of path planning; designing a variable space omega, acquiring sample points in the omega, generating a decision space pi, and generating an initial reference vector in the pi; adjusting the sample point governed by each initial reference vector; the user determines a most important objective function, adjusts the decomposition coefficient and generates an adjusted reference vector; using only the adjusted reference vector of the dominant sample pointF(X) Decomposing to obtain a decomposition expression; and taking each sample point of the population as a parent, generating a child through differential evolution, and carrying out population iteration according to the numerical value calculated by the decomposition expression. The invention can adaptively adjust the reference vector based on the path planning flow and user preference change caused by emergency, accelerates the convergence speed of path planning and improves the efficiency.
Description
Technical Field
The invention relates to the technical field of low-altitude remote sensing unmanned aerial vehicle path planning, in particular to a color-changing wood forest area unmanned aerial vehicle path multi-target planning method based on user preference.
Background
Patent publication No. CN116518982A discloses a multi-target planning method for a path of a remote sensing unmanned aerial vehicle monitored by low-altitude forestry. The method mentions that: one key constraint factor in the operation of low-altitude unmanned aerial vehicles is the planning of the flight path. In the low-altitude environment, fixed barriers such as trees, buildings and the like exist in the application scene of the unmanned aerial vehicle, various emergency situations such as flying birds, balloons, environmental wind and the like exist at the same time, and the unmanned aerial vehicle also needs to consider various factors such as energy consumption, timeliness, safety and the like when executing a flight task. The traditional means can not effectively and rapidly adjust the subsequent paths, so a multi-objective planning method is provided, the time and the consumed energy required by the operation of the unmanned aerial vehicle and the survival rate of the unmanned aerial vehicle are used as multi-objective functions in path planning, and a multi-objective solution is utilized to obtain a multi-party balanced optimized path scheme, so that the overall efficiency of the unmanned aerial vehicle is optimal in operation. However, this solution still has the following problems:
first, the color-changing wood refers to a situation that the color of leaves is changed due to occurrence of insect diseases and the like of trees in a forest area. When color-changing wood appears in a forest area, timely screening is needed. Therefore, unlike the path planning problem of unmanned aerial vehicles in the conventional forest zone, the path planning of the color-changing wood forest zone needs to consider further arrival confirmation after the color-changing wood is primarily identified, and the state confirmation of the color-changing wood is completely determined by the knowledge and experience of the user, so that the method has great subjectivity, and the subjectivity factors of the user influence the path planning result to a certain extent.
Secondly: aiming at complex application scenes, a good and reasonable flight path is made according to user preference, and the task execution efficiency of the low-altitude remote sensing unmanned aerial vehicle is determined. However, the patent with publication number CN116518982a cannot consider the application preference of the user to the unmanned aerial vehicle during path planning, and cannot efficiently complete all task requirements of the user. And after the patent finishes target planning, the user is dissatisfied with the result and cannot adjust the result according to the preference and special requirements.
Therefore, in the remote sensing scene of the low-altitude color-changing woodland, the rapid path planning of all complex influence factors in the application scene needs to be fully considered.
Disclosure of Invention
The invention aims to provide a color-changing wood forest zone unmanned aerial vehicle path multi-target planning method based on user preference, which solves the defects that the traditional path specification method only usually considers a single path influencing factor, has poor robustness of path results, cannot consider the application preference of a user to the unmanned aerial vehicle during path planning, and has low planning efficiency, poor precision and the like.
In order to achieve the above purpose, the technical scheme adopted by the invention is as follows: a color-changing wood forest zone unmanned aerial vehicle path multi-target planning method based on user preference comprises the following steps of;
s1, establishing a Maklink three-dimensional graph theory model of an unmanned aerial vehicle path environment based on initial environment information, wherein the initial environment information comprises an unmanned aerial vehicle initial position, a task target position and three-dimensional map information;
s2, setting an objective function and a constraint function when the unmanned aerial vehicle travels along a path, and establishing a multi-objective optimization mathematical model of path planningF(X)：
；
In the method, in the process of the invention,f 1 (X)~f N (X) For the 1 st to nth objective functions,Xomega is a design variable space, G is an expression of a constraint function;
s3, obtaining 3 in the design variable space omegak+2 sample points forming a multi-objective planned population, each sample point corresponding to an initial path, calculating a corresponding multi-objective function value and constraint function value, wherein, the firstjSample pointsX j The corresponding multiple objective function value isF(X j )，j=1~3k+2；
S4, generating a decision space pi according to all the multiple objective function values, generating a group of initial reference vectors Γ in the decision space pi, and comprisingInitial reference vector, 1bThe initial reference vectors are，b=1~/>，λ b1 ~λ Nb Respectively isf 1 (X)~f N (X) And (2) the decomposition coefficient ofλ b1 +λ b2 +…+λ bN =1；
S5, adjusting the sample point governed by each initial reference vector;
s6, determining a most important objective function according to the preference of the user, adjusting the decomposition coefficient, and generating an adjusted reference vector, wherein the reference vector comprises S6.1-S6.5;
s6.1, determining the S-th objective functionf s (X) S=1 to n, which is an important objective function;
s6.2, for the b-th initial reference vector Γ b According to the formulaλ bs ’=(1+β)λ bs Adjustment off s (X) Corresponding decomposition coefficientλ bs Obtaining the adjusted decomposition coefficientλ bs ’；
S6.3, calculating the adjustment parametersα，；
S6.4, according to the formulaCalculating adjusted decomposition coefficients corresponding to the rest objective functionsλ bm ’M=1 to n and m+.sGenerating an adjusted reference vector->；
S6.5, obtaining adjusted reference vectors corresponding to the rest initial reference vectors according to the steps S6.2-S6.4, and sharingThe sample point governed by the adjusted reference vector is identical to the initial reference vector before adjustment;
s7, only using the adjusted reference vector with the sample points to regulateF(X) Decomposing to obtain a decomposition expression, wherein the decomposition expression corresponding to the b-th adjusted reference vector is；
；
S8, taking each sample point of the population as a parent, generating a child through differential evolution, wherein the child belongs to and inherits the parent;
s9, forming a pair of sample points by the corresponding parent and offspring, substituting the sample points into the decomposition expression of the step S7 to calculate a numerical value for population iteration;
and S10, circulating the steps S4-S9 until the number of times of calculating the path by the Maklink three-dimensional graph theory model reaches a set value, stopping circulation, and outputting the current population as a multi-target optimal solution.
As preferable: s2, establishing a multi-objective optimized mathematical model of path planningF(X) Comprises the steps S2.1-S2.4;
s2.1, dividing a three-dimensional space grid in a Maklink three-dimensional graph theory model, and setting up a path on the three-dimensional space grid, wherein the mathematical form of the path is as followsX=(x 1 ,x 2 ,…,x k ) Whereinx k Is thatXIs the first of (2)kA number of design variables;
s2.2, setting N objective functions according to own preference by a user, and marking the N objective functions as follows in turnf 1 (X) ~f N (X)；
S2.3, setting n constraint functions to obtain an expression G of the n constraint functions;
s2.4 based onXInitializing a design variable space omega by the upper limit and the lower limit of the model, and establishing a multi-objective optimized mathematical model of path planningF(X)；
S2.2, a plurality of different types of image sensors are arranged on the unmanned aerial vehicle, the recognition grades of the different image sensors on the color-changing wood are different, and the objective function comprises, but is not limited to, time required by the unmanned aerial vehicle to travel along a path, consumed energy, survival rate of the unmanned aerial vehicle and recognition results of each image sensor on the color-changing wood.
As preferable: in S2.3, the number of the steps is,，g i (X) Is the firstiThe number of constraint functions is a function of the number of constraints,a i is the firstiThe control coefficients of the individual constraint functions,i=1~nthe constraint function includes obstacles encountered by the drone as it travels along the path.
As preferable: in S2.4, the design variable space Ω= {ω 1 ,ω 2 ,…,ω k And } wherein,ω k =[minx k , maxx k ]，ω k is thatx k Value interval of (min)x k 、maxx k Respectively isx k Lower and upper limits of (2).
As preferable: and S3, obtaining a sample point by using a Latin hypercube sampling method.
As preferable: s4, decision spaceWherein->Is the firstNIndividual objective functionsf N （X) Value interval of (min)f N （X）、maxf N （X) Respectively isf N （X) Lower and upper limits of (2).
As preferable: and S8, the differential evolution of the sample points is that for each initial reference vector, the sample points governed by the initial reference vector are put into a blank evolution pool, then the adjacent initial reference vectors of the initial reference vectors are found, the sample points governed by the adjacent initial reference vectors are added into the evolution pool, the differential evolution is adopted for the evolution pool, the sample points with the same scale as the evolution pool are obtained, and the sample points are stored into a standby offspring sample point set.
As preferable: in S9, the iteration of the population is specifically that if the numerical value of the offspring is small, the offspring is used for replacing the parent in the population, otherwise, the parent is reserved.
As preferable: s5 comprises the steps of S5.1-S5.2;
s5.1, calculating the included angle value from the multi-objective function value of each sample point to all initial reference vectors in a decision space II, finding the initial reference vector corresponding to the minimum included angle value, and adjusting the sample point to be governed by the initial reference vector;
s5.2, after the step S5.1 is completed, if the following conditions exist, the step is adjusted again;
case 1: when 3 isk+2≥If the number of the sample points governed by the initial reference vector is 0, adjusting any sample point to the initial reference vector from the initial reference vector governed by a plurality of sample points;
case 2: when 3 isk+2<If the number of the sample points governed by the initial reference vector exceeds 1, optionally reserving one sample point, and randomly dividing the rest sample points under the initial reference vector governed by zero sample points.
Compared with the prior art, the invention has the advantages that:
the color-changing wood forest zone unmanned aerial vehicle path multi-objective planning method based on user preference can enable a user to set content and quantity of objective functions of multi-objective path planning at will, for example, time required by unmanned aerial vehicle operation, consumed energy, survival rate of the unmanned aerial vehicle, special application requirements of the user and the like are used as multi-objective functions of path planning, and therefore the planned path can fully consider various complex influence factors in application scenes. And due to the characteristic of population searching in the multi-objective algorithm, the obtained result can be ensured to have robust stability in a certain interval, and the optimization objective can be increased or decreased at will under the complete optimization framework.
The method utilizes the user preference to generate a plurality of reference vectors in a decision space, and adjusts the internal weight value of the reference vectors according to the user preference; the multi-objective problem is decomposed by the reference vector, so that the solving of the multi-objective problem is simplified, the optimized calculated amount is reduced, the solving speed is increased, the population in the solving process can be converged more accurately, and the robustness and stability of the result are improved.
The invention updates and adaptively adjusts the reference vector based on the iteration process and the user preference change, and realizes the balance between the efficiency and the precision of the solving result.
In conclusion, the method and the device effectively solve the problems of low efficiency and low precision in the low-altitude remote sensing unmanned aerial vehicle multi-target path planning process, and improve the calculation efficiency and accuracy.
Drawings
FIG. 1 is a flow chart of the present invention.
Detailed Description
The invention will be further described with reference to the accompanying drawings.
Example 1: referring to fig. 1, a multi-objective planning method for unmanned aerial vehicle paths in a color-changing woodland based on user preference comprises the following steps of;
s1, establishing a Maklink three-dimensional graph theory model of an unmanned aerial vehicle path environment based on initial environment information, wherein the initial environment information comprises an unmanned aerial vehicle initial position, a task target position and three-dimensional map information;
s2, setting an objective function and a constraint function when the unmanned aerial vehicle travels along a path, and establishing a multi-objective optimization mathematical model of path planning, wherein the steps comprise S2.1-S2.4;
s2.1, dividing a three-dimensional space grid in a Maklink three-dimensional graph theory model, and setting up a path on the three-dimensional space grid, wherein the mathematical form of the path is as followsX=(x 1 ,x 2 ,…,x k )，Wherein the method comprises the steps ofx k Is thatXIs the first of (2)kA number of design variables;
s2.2, setting N objective functions according to own preference by a user, and marking the N objective functions as follows in turnf 1 (X) ~f N (X)；
S2.3, setting n constraint functions to obtain an expression G of the n constraint functions;
s2.4 based onXInitializing a design variable space omega by the upper limit and the lower limit of the model, and establishing a multi-objective optimized mathematical model of path planningF(X)；
S3, obtaining 3 in the design variable space omegak+2 sample points forming a multi-objective planned population, each sample point corresponding to an initial path, calculating a corresponding multi-objective function value and constraint function value, wherein, the firstjSample pointsX j The corresponding multiple objective function value isF(X j )，j=1~3k+2；
S4, generating a decision space pi according to all the multiple objective function values, generating a group of initial reference vectors Γ in the decision space pi, and comprisingInitial reference vector, 1bThe initial reference vector is->，b=1~/>，λ b1 ~λ Nb Respectively isf 1 (X)~f N (X) And (2) the decomposition coefficient ofλ b1 +λ b2 +…+λ bN =1；
S5, adjusting sample points governed by each initial reference vector, wherein the steps comprise S5.1-S5.2;
s5.1, calculating the included angle value from the multi-objective function value of each sample point to all initial reference vectors in a decision space II, finding the initial reference vector corresponding to the minimum included angle value, and adjusting the sample point to be governed by the initial reference vector;
s5.2, after the step S5.1 is completed, if the following conditions exist, the step is adjusted again;
case 1: when 3 isk+2≥If the number of the sample points governed by the initial reference vector is 0, adjusting any sample point to the initial reference vector from the initial reference vector governed by a plurality of sample points;
case 2: when 3 isk+2<If the number of the sample points governed by the initial reference vector exceeds 1, optionally reserving one sample point, and randomly dividing the rest sample points under the initial reference vector of the zero sample points governed by the initial reference vector;
s6, determining a most important objective function according to the preference of the user, adjusting the decomposition coefficient, and generating an adjusted reference vector, wherein the reference vector comprises S6.1-S6.5;
s6.1, determining the S-th objective functionf s (X) S=1 to n, which is an important objective function;
s6.2, for the b-th initial reference vector Γ b According to the formulaλ bs ’=(1+β)λ bs Adjustment off s (X) Corresponding decomposition coefficientλ bs Obtaining the adjusted decomposition coefficientλ bs ’；
S6.3, calculating the adjustment parametersα，；
S6.4, according to the formulaCalculating adjusted decomposition coefficients corresponding to the rest objective functionsλ bm ’M=1 to n and m+.sGenerating an adjusted reference vector->；
S6.5, obtaining adjusted reference vectors corresponding to the rest initial reference vectors according to the steps S6.2-S6.4, and sharingThe sample point governed by the adjusted reference vector is identical to the initial reference vector before adjustment;
s7, only using the adjusted reference vector with the sample points to regulateF(X) Decomposing to obtain a decomposition expression, wherein the decomposition expression corresponding to the b-th adjusted reference vector is；
；
S8, taking each sample point of the population as a parent, generating a child through differential evolution, wherein the child belongs to and inherits the parent;
s9, forming a pair of sample points by the corresponding parent and offspring, substituting the sample points into the decomposition expression of the step S7 to calculate a numerical value for population iteration;
and S10, circulating the steps S4-S9 until the number of times of calculating the path by the Maklink three-dimensional graph theory model reaches a set value, stopping circulation, and outputting the current population as a multi-target optimal solution.
In this embodiment S2.2, a plurality of different types of image sensors are disposed on the unmanned aerial vehicle, and the recognition grades of the different image sensors on the color-changing wood are different, and the objective function includes, but is not limited to, time required for the unmanned aerial vehicle to travel along the path, consumed energy, survival rate of the unmanned aerial vehicle, and recognition result of each image sensor on the color-changing wood.
In S2.3, the number of the steps is,，g i (X) Is the firstiThe number of constraint functions is a function of the number of constraints,a i is the firstiThe control coefficients of the individual constraint functions,i=1~nthe constraint function includes obstacles encountered by the drone as it travels along the path.
In S2.4, the design variable space Ω= {ω 1 ,ω 2 ,…,ω k And } wherein,ω k =[minx k , maxx k ]，ω k is thatx k Value interval of (min)x k 、maxx k Respectively isx k Lower and upper limits of (2).
And S3, obtaining a sample point by using a Latin hypercube sampling method.
S4, decision spaceWherein->Is the firstNIndividual objective functionsf N （X) Value interval of (min)f N （X）、maxf N （X) Respectively isf N （X) Lower and upper limits of (2).
And S8, the differential evolution of the sample points is that for each initial reference vector, the sample points governed by the initial reference vector are put into a blank evolution pool, then the adjacent initial reference vectors of the initial reference vectors are found, the sample points governed by the adjacent initial reference vectors are added into the evolution pool, the differential evolution is adopted for the evolution pool, the sample points with the same scale as the evolution pool are obtained, and the sample points are stored into a standby offspring sample point set.
In S9, the iteration of the population is specifically that if the numerical value of the offspring is small, the offspring is used for replacing the parent in the population, otherwise, the parent is reserved.
Regarding the recognition result of the color change wood by each image sensor in the objective function: when the color-changing wood forest zone is found, most of diseases and insect pests or special conditions in the zone are caused, so that the natural protection zone is required to be shot through an image sensor carried by an unmanned aerial vehicle, and the color-changing wood forest zone is required to be identified. However, the accuracy and the speed of the image recognition by different image sensors are different, so that the image recognition can be used as an objective function, the corresponding decomposition coefficients can be adjusted according to actual conditions, different requirements are met, and the number of times of calculating the path by the Maklink three-dimensional graph theory model is preset according to the actual conditions.
Example 2: on the basis of example 1, we present a specific method comprising the steps of:
s1, establishing a Maklink three-dimensional graph theory model of an unmanned aerial vehicle path environment based on initial environment information, specifically, acquiring space information of the unmanned aerial vehicle through an airborne camera according to a preset map and the unmanned aerial vehicle, and completing physical space modeling through SLAM;
s2, setting an objective function and a constraint function when the unmanned aerial vehicle travels along a path, and establishing a multi-objective optimization mathematical model of path planning, wherein the steps comprise S2.1-S2.4;
s2.1, dividing a three-dimensional space grid in a modeling space of a Maklink three-dimensional graph theory model, setting up a path on the three-dimensional space grid based on the Maklink graph theory, wherein the path comprises 10 control points, and the generation of the control points refers to a Maklink internal principle, so that the mathematical form of the path is as followsX=(x 1 ,x 2 ,…,x 10 )。
S2.2, setting N=5 objective functions according to own preference by a user, and marking the objective functions as follows in turnf 1 (X)~f 5 (X) The method comprises the steps of carrying out a first treatment on the surface of the Wherein the time required for the unmanned aerial vehicle to travel along the path is set asf 1 (X) The energy consumed when the unmanned aerial vehicle advances along the path isf 2 (X) The survival rate of the unmanned aerial vehicle is set as when the unmanned aerial vehicle advances along the pathf 3 (X) Target is the recognition result of color-changing wood by an image sensor in unmanned aerial vehiclef 4 (X) The color-changing wood is identified by another image sensor in the unmanned aerial vehicle as followsf 5 (X)。
S2.3, setting 4 constraint functions which are collision volume functions of 4 known barriers in the map in a three-dimensional space respectively to obtain 4 constraint functionsExpression of (2)，a 1 、a 2 、a 3 、a 4 The values of (2) are 0.25, respectively.
Based onXThe upper and lower limits of (1) initialize the design variable space omega,establishing a multi-objective optimized mathematical model for path planningF(X)：
；
S3, the same as the step S3 in the embodiment 1, wherein the number of sample points is 32.
S4, generating a decision space pi according to all the multiple objective function values,in this embodiment, <' > a->Generating a set of initial reference vectors Γ based on standard boundary crossing method in pi, comprising +.>Initial reference vector, 1bThe initial reference vectors areAnd->；
S5, adjusting the sample point governed by each initial reference vector, wherein the step is the same as the step S5 in the embodiment 1.
S6, the user determines a most important objective function according to the preference of the user, in the embodiment, the user considers that the influence factor of the path time on the task is most important, so the user selectsf 1 (X) As the most important objective function, then adjustment is requiredf 1 (X) Is of the decomposition coefficient of (2)λ b1 Generating an adjusted reference vector, including S6.1-S6.5;
s6.1, determining the 1 st objective functionf 1 (X) Is an important objective function;
s6.2, for the b-th initial reference vector Γ b According to the formulaCalculation ofλ b1 ’，β∈[0,0.5]Selecting in the embodiment;β=0.4, then->；
S6.3, calculating the adjustment parametersα，；
S6.4, according to the formulaCalculating adjusted decomposition coefficients corresponding to the rest objective functionsλ bm ’Obtainingλ b2 ’=0.008，λ b3 ’=0.361，λ b4 ’=0.159，λ b5 ’=0.150, provided b is 325 th, the 325 th adjusted reference vector is generated +.>。
S6.5, as in step S6.5 of example 1.
S7, only using the adjusted reference vector with the sample points to regulateF(X) Decomposing to obtain a decomposition expression, wherein the 325 th adjusted reference vector corresponds to the decomposition expression as follows：
；
S8, generating offspring by the father as in the step S8 in the embodiment 1;
s9, forming a pair of sample points by the corresponding parent and offspring, and carrying the calculated values in the decomposition expression in the step S7 for each pair of sample points to be used for population iteration, specifically, if the offspring value is small, replacing the parent in the population by the offspring, otherwise, reserving the parent, and if the sample points of the population after iteration are 32.
S10, setting convergence conditions as follows: the number of times the path is calculated based on the Maklink graph theory reaches 500 times. If the convergence condition is met, stopping circulation, and outputting the current population as a multi-target optimal solution; otherwise, repeating the steps S4-S9 until the convergence condition is met.
Compared with the traditional technology, the method provided by the invention has the advantages that the precision can be improved by 10% -15%, and the accuracy is improved by 25% -40%.
In addition, after the method is finished, if the result is not satisfied, the method can be manually participated in the step S6, and the user redefines a most important objective function according to the preference of the user and modifies the corresponding decomposition coefficient.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the invention.
Claims (9)
1. A color-changing wood forest zone unmanned aerial vehicle path multi-target planning method based on user preference is characterized by comprising the following steps of: comprises the following steps of;
s1, establishing a Maklink three-dimensional graph theory model of an unmanned aerial vehicle path environment based on initial environment information, wherein the initial environment information comprises an unmanned aerial vehicle initial position, a task target position and three-dimensional map information;
s2, setting an objective function and a constraint function when the unmanned aerial vehicle travels along a path, and establishing a multi-objective optimization mathematical model of path planningF(X)：
；
In the method, in the process of the invention,f 1 (X)~f N (X) For the 1 st to nth objective functions,Xomega is a design variable space, G is an expression of a constraint function;
s3, obtaining 3 in the design variable space omegak+2 sample points forming a multi-objective planned population, each sample point corresponding to an initial path, calculating a corresponding multi-objective function value and constraint function value, wherein, the firstjSample pointsX j The corresponding multiple objective function value isF(X j )，j=1~3k+2；
S4, generating a decision space pi according to all the multiple objective function values, generating a group of initial reference vectors Γ in the decision space pi, and comprisingInitial reference vector, 1bThe initial reference vector is->，b=1~/>，λ b1 ~λ Nb Respectively isf 1 (X)~f N (X) And (2) the decomposition coefficient ofλ b1 +λ b2 +…+λ bN =1；
S5, adjusting the sample point governed by each initial reference vector;
s6, determining a most important objective function according to the preference of the user, adjusting the decomposition coefficient, and generating an adjusted reference vector, wherein the reference vector comprises S6.1-S6.5;
s6.1, determining the S-th objective functionf s (X) S=1 to n, which is an important objective function;
s6.2, for the b-th initial reference vector Γ b According to the formulaλ bs ’=(1+β)λ bs Adjustment off s (X) Corresponding decomposition coefficientλ bs Obtaining the adjusted decomposition coefficientλ bs ’；
S6.3, calculating the adjustment parametersα，；
S6.4, according to the formulaCalculating adjusted decomposition coefficients corresponding to the rest objective functionsλ bm ’M=1 to n and m+.sGenerating an adjusted reference vector->；
S6.5, obtaining adjusted reference vectors corresponding to the rest initial reference vectors according to the steps S6.2-S6.4, and sharingThe sample point governed by the adjusted reference vector is identical to the initial reference vector before adjustment;
s7, only using the adjusted reference vector with the sample points to regulateF(X) Decomposing to obtain a decomposition expression, wherein the decomposition expression corresponding to the b-th adjusted reference vector is；
；
S8, taking each sample point of the population as a parent, generating a child through differential evolution, wherein the child belongs to and inherits the parent;
s9, forming a pair of sample points by the corresponding parent and offspring, substituting the sample points into the decomposition expression of the step S7 to calculate a numerical value for population iteration;
and S10, circulating the steps S4-S9 until the number of times of calculating the path by the Maklink three-dimensional graph theory model reaches a set value, stopping circulation, and outputting the current population as a multi-target optimal solution.
2. The user preference-based color-changing woodland unmanned aerial vehicle path multi-objective planning method according to claim 1, wherein: s2, establishing a multi-objective optimized mathematical model of path planningF(X) Comprises the steps S2.1-S2.4;
s2.1, dividing a three-dimensional space grid in a Maklink three-dimensional graph theory model, and setting up a path on the three-dimensional space grid, wherein the mathematical form of the path is as followsX=(x 1 ,x 2 ,…,x k ) Whereinx k Is thatXIs the first of (2)kA number of design variables;
s2.2, setting N objective functions according to own preference by a user, and marking the N objective functions as follows in turnf 1 (X) ~f N (X)；
S2.3, setting n constraint functions to obtain an expression G of the n constraint functions;
s2.4 based onXInitializing a design variable space omega by the upper limit and the lower limit of the model, and establishing a multi-objective optimized mathematical model of path planningF(X)；
S2.2, a plurality of different types of image sensors are arranged on the unmanned aerial vehicle, the recognition grades of the different image sensors on the color-changing wood are different, and the objective function comprises, but is not limited to, time required by the unmanned aerial vehicle to travel along a path, consumed energy, survival rate of the unmanned aerial vehicle and recognition results of each image sensor on the color-changing wood.
3. The user preference-based color-changing woodland unmanned aerial vehicle path multi-objective planning method according to claim 2, wherein: in S2.3, the number of the steps is,，g i (X) Is the firstiThe number of constraint functions is a function of the number of constraints,a i is the firstiThe control coefficients of the individual constraint functions,i=1~nthe constraint function includes obstacles encountered by the drone as it travels along the path.
4. The user preference-based color-changing woodland unmanned aerial vehicle path multi-objective planning method according to claim 2, wherein: in S2.4, the design variable space Ω= {ω 1 ,ω 2 ,…,ω k And } wherein,ω k =[minx k ,maxx k ]，ω k is thatx k Value interval of (min)x k 、maxx k Respectively isx k Lower and upper limits of (2).
5. The user preference-based color-changing woodland unmanned aerial vehicle path multi-objective planning method according to claim 1, wherein: and S3, obtaining a sample point by using a Latin hypercube sampling method.
6. The user preference-based color-changing woodland unmanned aerial vehicle path multi-objective planning method according to claim 1, wherein: s4, decision spaceWherein->Is the firstNIndividual objective functionsf N （X) Value interval of (min)f N （X）、maxf N （X) Respectively isf N （X) Lower and upper limits of (2).
7. The user preference-based color-changing woodland unmanned aerial vehicle path multi-objective planning method according to claim 1, wherein: and S8, the differential evolution of the sample points is that for each initial reference vector, the sample points governed by the initial reference vector are put into a blank evolution pool, then the adjacent initial reference vectors of the initial reference vectors are found, the sample points governed by the adjacent initial reference vectors are added into the evolution pool, the differential evolution is adopted for the evolution pool, the sample points with the same scale as the evolution pool are obtained, and the sample points are stored into a standby offspring sample point set.
8. The user preference-based color-changing woodland unmanned aerial vehicle path multi-objective planning method according to claim 1, wherein: in S9, the iteration of the population is specifically that if the numerical value of the offspring is small, the offspring is used for replacing the parent in the population, otherwise, the parent is reserved.
9. The user preference-based color-changing woodland unmanned aerial vehicle path multi-objective planning method according to claim 1, wherein: s5 comprises the steps of S5.1-S5.2;
s5.1, calculating the included angle value from the multi-objective function value of each sample point to all initial reference vectors in a decision space II, finding the initial reference vector corresponding to the minimum included angle value, and adjusting the sample point to be governed by the initial reference vector;
s5.2, after the step S5.1 is completed, if the following conditions exist, the step is adjusted again;
case 1: when 3 isk+2≥If the number of the sample points governed by the initial reference vector is 0, adjusting any sample point to the initial reference vector from the initial reference vector governed by a plurality of sample points;
case 2: when 3 isk+2<If the number of the sample points governed by the initial reference vector exceeds 1, optionally reserving one sample point, and randomly dividing the rest sample points under the initial reference vector governed by zero sample points.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311707190.6A CN117387634B (en) | 2023-12-13 | 2023-12-13 | Color-changing wood forest zone unmanned aerial vehicle path multi-target planning method based on user preference |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311707190.6A CN117387634B (en) | 2023-12-13 | 2023-12-13 | Color-changing wood forest zone unmanned aerial vehicle path multi-target planning method based on user preference |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117387634A true CN117387634A (en) | 2024-01-12 |
| CN117387634B CN117387634B (en) | 2024-02-27 |
Family
ID=89468866
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311707190.6A Active CN117387634B (en) | 2023-12-13 | 2023-12-13 | Color-changing wood forest zone unmanned aerial vehicle path multi-target planning method based on user preference |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117387634B (en) |
Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0631273A1 (en) * | 1993-06-25 | 1994-12-28 | Koninklijke KPN N.V. | Method for detecting the best path through a stochastic network, in particular for speech or image recognition |
| CN102129249A (en) * | 2011-01-10 | 2011-07-20 | 中国矿业大学 | Method for planning global path of robot under risk source environment |
| US20130245940A1 (en) * | 2012-03-15 | 2013-09-19 | Fujitsu Limited | Path searching method and path search device |
| CN105513400A (en) * | 2015-12-03 | 2016-04-20 | 四川长虹电器股份有限公司 | Method for dynamically planning travel route |
| CN105929848A (en) * | 2016-06-28 | 2016-09-07 | 南京邮电大学 | Track planning method for multi-unmanned plane system in three-dimensional environment |
| CN111461383A (en) * | 2019-03-14 | 2020-07-28 | 北京京东尚科信息技术有限公司 | Method and device for planning distribution path |
| WO2020191692A1 (en) * | 2019-03-28 | 2020-10-01 | 电子科技大学 | Maklink graph-based pathfinding method |
| CN213120575U (en) * | 2020-08-31 | 2021-05-04 | 国家电网有限公司 | Indoor positioning navigation system and indoor power distribution equipment monitoring control system |
| CN113420583A (en) * | 2021-01-29 | 2021-09-21 | 山东农业大学 | Color-changing wood monitoring system and method based on edge computing platform |
| CN113435497A (en) * | 2021-06-24 | 2021-09-24 | 桂林电子科技大学 | Coal-rock interface multi-sensing information fusion preference identification method based on hardness characteristics |
| KR102312413B1 (en) * | 2021-04-07 | 2021-10-13 | 이승희 | artificial neural network-based space interior curating system and method therefor |
| CN113885535A (en) * | 2021-11-25 | 2022-01-04 | 长春工业大学 | Impact-constrained robot obstacle avoidance and time optimal trajectory planning method |
| CN114740860A (en) * | 2022-04-21 | 2022-07-12 | 浙江科技学院 | Multi-target path planning method based on Gaussian reference point self-adaptive strategy |
| US20220351626A1 (en) * | 2021-04-29 | 2022-11-03 | Skygrid, Llc | Multi-objective mission planning and execution for an unmanned aerial vehicle |
| CN115903896A (en) * | 2022-11-22 | 2023-04-04 | 中国人民解放军国防科技大学 | Multi-unmanned aerial vehicle path planning method based on proxy model optimization |
| CN116258971A (en) * | 2023-05-15 | 2023-06-13 | 江西啄木蜂科技有限公司 | Multi-source fused forestry remote sensing image intelligent interpretation method |
| CN116518982A (en) * | 2023-07-03 | 2023-08-01 | 江西啄木蜂科技有限公司 | Low-altitude forestry monitoring remote sensing unmanned aerial vehicle path multi-target planning method |
| CN116518979A (en) * | 2023-06-26 | 2023-08-01 | 深圳市森歌数据技术有限公司 | Unmanned plane path planning method, unmanned plane path planning system, electronic equipment and medium |
| CN116863342A (en) * | 2023-09-04 | 2023-10-10 | 江西啄木蜂科技有限公司 | Large-scale remote sensing image-based pine wood nematode dead wood extraction method |
-
2023
- 2023-12-13 CN CN202311707190.6A patent/CN117387634B/en active Active
Patent Citations (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0631273A1 (en) * | 1993-06-25 | 1994-12-28 | Koninklijke KPN N.V. | Method for detecting the best path through a stochastic network, in particular for speech or image recognition |
| CN102129249A (en) * | 2011-01-10 | 2011-07-20 | 中国矿业大学 | Method for planning global path of robot under risk source environment |
| US20130245940A1 (en) * | 2012-03-15 | 2013-09-19 | Fujitsu Limited | Path searching method and path search device |
| CN105513400A (en) * | 2015-12-03 | 2016-04-20 | 四川长虹电器股份有限公司 | Method for dynamically planning travel route |
| CN105929848A (en) * | 2016-06-28 | 2016-09-07 | 南京邮电大学 | Track planning method for multi-unmanned plane system in three-dimensional environment |
| CN111461383A (en) * | 2019-03-14 | 2020-07-28 | 北京京东尚科信息技术有限公司 | Method and device for planning distribution path |
| WO2020191692A1 (en) * | 2019-03-28 | 2020-10-01 | 电子科技大学 | Maklink graph-based pathfinding method |
| CN213120575U (en) * | 2020-08-31 | 2021-05-04 | 国家电网有限公司 | Indoor positioning navigation system and indoor power distribution equipment monitoring control system |
| CN113420583A (en) * | 2021-01-29 | 2021-09-21 | 山东农业大学 | Color-changing wood monitoring system and method based on edge computing platform |
| KR102312413B1 (en) * | 2021-04-07 | 2021-10-13 | 이승희 | artificial neural network-based space interior curating system and method therefor |
| US20220351626A1 (en) * | 2021-04-29 | 2022-11-03 | Skygrid, Llc | Multi-objective mission planning and execution for an unmanned aerial vehicle |
| CN113435497A (en) * | 2021-06-24 | 2021-09-24 | 桂林电子科技大学 | Coal-rock interface multi-sensing information fusion preference identification method based on hardness characteristics |
| CN113885535A (en) * | 2021-11-25 | 2022-01-04 | 长春工业大学 | Impact-constrained robot obstacle avoidance and time optimal trajectory planning method |
| CN114740860A (en) * | 2022-04-21 | 2022-07-12 | 浙江科技学院 | Multi-target path planning method based on Gaussian reference point self-adaptive strategy |
| CN115903896A (en) * | 2022-11-22 | 2023-04-04 | 中国人民解放军国防科技大学 | Multi-unmanned aerial vehicle path planning method based on proxy model optimization |
| CN116258971A (en) * | 2023-05-15 | 2023-06-13 | 江西啄木蜂科技有限公司 | Multi-source fused forestry remote sensing image intelligent interpretation method |
| CN116518979A (en) * | 2023-06-26 | 2023-08-01 | 深圳市森歌数据技术有限公司 | Unmanned plane path planning method, unmanned plane path planning system, electronic equipment and medium |
| CN116518982A (en) * | 2023-07-03 | 2023-08-01 | 江西啄木蜂科技有限公司 | Low-altitude forestry monitoring remote sensing unmanned aerial vehicle path multi-target planning method |
| CN116863342A (en) * | 2023-09-04 | 2023-10-10 | 江西啄木蜂科技有限公司 | Large-scale remote sensing image-based pine wood nematode dead wood extraction method |
Non-Patent Citations (1)
| Title |
|---|
| 陈先亮等: "基于多模态多目标进化算法的无人机三维路径规划", 火力与指挥控制, vol. 48, no. 11, 30 November 2023 (2023-11-30), pages 32 - 39 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117387634B (en) | 2024-02-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113110592B (en) | Unmanned aerial vehicle obstacle avoidance and path planning method | |
| CN112733421B (en) | Task planning method for cooperation of unmanned aerial vehicle with ground fight | |
| CN110928189B (en) | Robust control method based on reinforcement learning and Lyapunov function | |
| CN106441308A (en) | Unmanned aerial vehicle path planning method based on adaptive weight pigeon flock algorithm | |
| CN109871032A (en) | A kind of multiple no-manned plane formation cooperative control method based on Model Predictive Control | |
| CN110531786B (en) | Unmanned aerial vehicle maneuvering strategy autonomous generation method based on DQN | |
| CN112230678A (en) | Three-dimensional unmanned aerial vehicle path planning method and planning system based on particle swarm optimization | |
| CN108549233A (en) | A kind of unmanned plane air combat maneuvering game method with intuitionistic fuzzy information | |
| CN110926477A (en) | Unmanned aerial vehicle route planning and obstacle avoidance method | |
| CN113848919A (en) | Ant colony algorithm-based indoor AGV path planning method | |
| CN109144062B (en) | Path planning method for dangerous rescue robot | |
| CN112947541B (en) | Unmanned aerial vehicle intention track prediction method based on deep reinforcement learning | |
| CN113190037A (en) | Unmanned aerial vehicle optimal path searching method based on improved fluid disturbance and sparrow algorithm | |
| CN115033022A (en) | DDPG unmanned aerial vehicle landing method based on expert experience and oriented to mobile platform | |
| CN116048071A (en) | Mobile robot path planning method based on particle swarm and differential evolution algorithm | |
| CN113268074A (en) | Unmanned aerial vehicle flight path planning method based on joint optimization | |
| CN113625569B (en) | Small unmanned aerial vehicle prevention and control decision method and system based on hybrid decision model | |
| CN117556979B (en) | Unmanned plane platform and load integrated design method based on group intelligent search | |
| CN114895707A (en) | Agricultural unmanned aerial vehicle path planning method and system based on variable-frequency bat algorithm | |
| CN115060263A (en) | Flight path planning method considering low-altitude wind and energy consumption of unmanned aerial vehicle | |
| CN116069045A (en) | Radiation environment detection method and system based on mobile robot | |
| Cui et al. | UAV path planning method for data collection of fixed-point equipment in complex forest environment | |
| CN117387634B (en) | Color-changing wood forest zone unmanned aerial vehicle path multi-target planning method based on user preference | |
| CN117193004A (en) | Unmanned aerial vehicle three-dimensional path planning method based on improved symbiotic particle swarm algorithm | |
| CN114580301A (en) | Hypersonic aircraft trajectory online prediction method based on parameter extrapolation |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Ye Shaoze Inventor after: Tian Yao Inventor after: Xi Yong Inventor before: Ye Shaoze Inventor before: Tian Yao Inventor before: Xi Yong |