CN113203417B - Unmanned aerial vehicle inspection path planning method based on brain storm optimization algorithm - Google Patents
Unmanned aerial vehicle inspection path planning method based on brain storm optimization algorithm Download PDFInfo
- Publication number
- CN113203417B CN113203417B CN202110368928.5A CN202110368928A CN113203417B CN 113203417 B CN113203417 B CN 113203417B CN 202110368928 A CN202110368928 A CN 202110368928A CN 113203417 B CN113203417 B CN 113203417B
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- aerial vehicle
- optimization algorithm
- path planning
- flight
- 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.)
- Active
Links
Images
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/901—Indexing; Data structures therefor; Storage structures
- G06F16/9024—Graphs; Linked lists
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/906—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Automation & Control Theory (AREA)
- Software Systems (AREA)
- Probability & Statistics with Applications (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The application discloses an unmanned aerial vehicle routing inspection path planning method based on a brainstorming optimization algorithm, which comprises the following steps: establishing a three-dimensional environment for routing and planning of the unmanned aerial vehicles, and determining the number of the unmanned aerial vehicles, and a starting point and an acquisition point of each unmanned aerial vehicle; establishing an unmanned aerial vehicle inspection path planning model with corner energy consumption according to the energy consumption constraint parameters; and acquiring the optimal path of the routing plan of the unmanned aerial vehicle by adopting a target space clustering discrete head storm optimization algorithm according to the routing plan model of the unmanned aerial vehicle. According to the unmanned aerial vehicle routing inspection path planning method based on the brainstorming optimization algorithm, the unmanned aerial vehicle routing inspection path planning model with corner energy consumption can be established according to the energy consumption constraint parameters, the optimal path of the unmanned aerial vehicle routing inspection path planning is obtained by adopting the target space clustering discrete brainstorming optimization algorithm, and the problem of unmanned aerial vehicle routing inspection scheduling is effectively solved.
Description
Technical Field
The application relates to the technical field of unmanned aerial vehicle routing inspection scheduling, in particular to an unmanned aerial vehicle routing inspection path planning method based on a target space clustering discrete brainstorming optimization algorithm.
Background
With the development of mobile communication technology and internet of things technology, sensor networks with short-distance transmission capability are increasingly widely applied to various field power cable monitoring. The power cable has large span and wide distribution range and is often remote. In order to acquire sensor data, workers are often required to go to various places in the field to acquire data, manpower and material resources are consumed, and the instantaneity is low.
As a new emerging technology, the unmanned aerial vehicle technology has the characteristics of high speed, convenience and flexibility in use, low cost and the like, and is more suitable for collecting scattered sensor data. In the unmanned aerial vehicle routing inspection scheduling, the fact that the capacity of a battery carried by the unmanned aerial vehicle is limited is firstly considered, flight energy consumption required by an unmanned aerial vehicle task needs to be optimized, the unmanned aerial vehicle energy consumption under the conditions of different turning angles, flight speeds and the like is calculated, and the path planning of the unmanned aerial vehicle is carried out based on an energy consumption model. Secondly, under the conditions that a plurality of data acquisition points are arranged and the distance is long, a plurality of unmanned aerial vehicles need to cooperatively finish the data acquisition task of the sensor within a limited time.
However, in the unmanned aerial vehicle with corner energy consumption planning path model in the prior art, while traversing all acquisition points, each unmanned aerial vehicle is difficult to solve by applying an accurate algorithm on the premise of satisfying energy consumption constraint and minimizing total flight time, and a more appropriate routing inspection route is designed.
Disclosure of Invention
An object of the application is to provide a new technical scheme of an unmanned aerial vehicle routing inspection path planning method based on a target space clustering discrete head storm optimization algorithm, which can solve the problem that in the prior art, a routing model with corner energy consumption unmanned aerial vehicle cannot be designed to be more suitable routing inspection routes on the premise of meeting energy consumption constraints and the shortest total flight time.
The application provides an unmanned aerial vehicle routing inspection path planning method based on a brainstorming optimization algorithm, which comprises the following steps:
establishing a three-dimensional environment for routing and planning of the unmanned aerial vehicles, and determining the number of the unmanned aerial vehicles, and a starting point and an acquisition point of each unmanned aerial vehicle;
establishing an unmanned aerial vehicle routing inspection path planning model with corner energy consumption according to the energy consumption constraint parameters;
according to the unmanned aerial vehicle routing inspection path planning model, acquiring an optimal path of the unmanned aerial vehicle routing inspection path planning by adopting a target space clustering discrete head storm optimization algorithm;
the energy consumption constraint parameters comprise unmanned aerial vehicle acceleration flight time, unmanned aerial vehicle uniform speed flight time, unmanned aerial vehicle deceleration flight time, acceleration flight energy consumption between path nodes, uniform speed flight energy consumption, deceleration flight energy consumption and corner energy consumption at path intermediate nodes;
the unmanned aerial vehicle routing inspection path planning model is represented by an undirected graph G (V, E), wherein V is {0,1, …, n } and is a vertex set, wherein 0 is an unmanned aerial vehicle departure point, a collection point set C is V \ 0}, E is { (i, j) | i, j ∈ V } and is a set of edges, all unmanned aerial vehicles are of the same type, the number of the unmanned aerial vehicles is K, and the battery capacity of the unmanned aerial vehicles is limited to Q;
the target space clustering dispersion head storm optimization algorithm is directly applied to the unmanned aerial vehicle inspection path planning model, the algorithm adopts individual space integer coding, a staged greedy decoding strategy with 2-OPT is adopted to realize conversion from an individual space to a solution space, and discretization definition is carried out on a disturbance operator and an individual update operator, so that the optimal path of the unmanned aerial vehicle inspection path planning is obtained.
Further, in the unmanned aerial vehicle routing inspection path planning model, it is assumed that path nodes i, j, and l are not on a straight line, wherein an included angle between an extension line of ij and a line segment jl is a flight rotation angle of the unmanned aerial vehicle, and is recorded as θ ijl And recording the corner energy consumption of the unmanned aerial vehicle flying as EA ijl ，EA ijl Is theta ijl As shown in equation (1):
EA ijl ＝f EA (θ ijl ) (1)
assuming that L is the range of the complete path of the drone, the time of flight FT and the energy consumed EN of the drone are functions of the range L, as shown in equations (2) and (3):
wherein l Acc For accelerating the course of flight L,/ Dec Flight path, L, of a decelerated flight path, L Const The flight distance is the flight distance of the uniform speed flight process of the flight distance L.
Further, the unmanned aerial vehicle inspection path planning model is as shown in formula (4) -formula (8):
wherein, c ij Is the distance from point i to point j, x ijk Is a decision variable and is 1 when the unmanned plane k travels from node i to j, otherwise is 0; y is ik Is a decision variable, and is 1 when the unmanned plane k serves the node i, otherwise is 0; the formula (4) is an objective function, and the sum of the flight times of all the paths of the unmanned aerial vehicles is minimized; formula (5) indicates that the acquisition point can only be executed by one unmanned aerial vehicle; formula (6) shows that the unmanned aerial vehicle which starts from the center is not more than K stations; the formula (7) ensures that the number of the unmanned aerial vehicles arriving and leaving at each node is balanced; and the formula (8) is that the battery capacity of the unmanned aerial vehicle is greater than the sum of the voyage energy consumption and the corner energy consumption.
Further, the unmanned aerial vehicle inspection path planning model further comprises an equation (9) and an equation (11):
y ik ∈{0,1},
x ijk ∈{0,1},
wherein, the formula (9) is used for avoiding the sub-loop phenomenon; equations (10) and (11) represent the decision variables x ijk And y ik Is a variable from 0 to 1.
Further, the target spatial clustering discrete brainstorming optimization algorithm comprises the following steps:
s1: randomly initializing scheme clusters X with the number of POP _ SIZE, and initializing the current iteration number to be 1;
s2: calculating the fitness of all individuals in the scheme cluster X;
s3: if the current iteration times do not reach MaxIter, jumping to the next step, otherwise, jumping to the step S9;
s4: sorting the scheme groups according to the fitness, and clustering by adopting a convergence operator;
s5: performing a perturbation operator operation;
s6: carrying out individual updating on each individual in the scheme group by adopting an individual updating operator;
s7: adding 1 to the current iteration times;
s8: jumping to step S3;
s9: returning to the optimal solution;
wherein, POP _ SIZE represents the number of original population of the scheme, and MaxIter represents the maximum iteration number.
Further, in the target spatial clustering discrete brainstorming optimization algorithm, each scheme individual in the individual spatial integer coding is an n-dimensional integer vector, a single unmanned aerial vehicle route is represented by a route representation method, and acquisition point sequences in cruise routes of multiple unmanned aerial vehicles are connected to obtain the scheme individual coding.
Further, in the target space clustering discrete brain storm optimization algorithm, decoding is performed according to an acceleration flight phase, a constant speed flight phase and a deceleration flight phase of the unmanned aerial vehicle flight, and 2-OPT local optimization is performed on a decoding result.
Further, in the target space clustering discrete brainstorming optimization algorithm, the perturbation operator adopts a random individual clustering center replacement strategy to carry out discretization definition.
Further, in the target space clustering discrete brainstorming optimization algorithm, the individual updating operator adopts a random inversion transformation strategy and a partial matching transformation generation strategy to carry out discretization definition on individual generating operation, and adopts an integral substitution updating strategy to carry out discretization definition on the individual updating operation.
Further, the fitness function of the target spatial clustering discrete head storm optimization algorithm is as shown in formula (12):
wherein f is FT A function representing the time of flight and range L of the drone, c ij Is the distance from point i to point j, x ijk Representing the decision variables.
According to the unmanned aerial vehicle routing inspection path planning method based on the brainstorming optimization algorithm, the unmanned aerial vehicle routing inspection path planning model with corner energy consumption can be established according to the energy consumption constraint parameters, the optimal path of the unmanned aerial vehicle routing inspection path planning is obtained by adopting the target space clustering discrete brainstorming optimization algorithm, and the problem of unmanned aerial vehicle routing inspection scheduling is effectively solved.
Further features of the present application and advantages thereof will become apparent from the following detailed description of exemplary embodiments thereof, which is to be read in connection with the accompanying drawings.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the application and together with the description, serve to explain the principles of the application.
FIG. 1 is a flow chart of an unmanned aerial vehicle routing inspection path planning method based on a brainstorming optimization algorithm;
FIG. 2 is a schematic view of the corner of the present invention;
FIG. 3 is a flow chart of a greedy by stage approach with 2-OPT in the algorithm of the present invention;
FIG. 4 is a schematic diagram of the random inversion transformation in the algorithm of the present invention;
FIG. 5 is a schematic diagram of a partial match transform in the algorithm of the present invention;
FIG. 6 is a flow chart of individual updates in the algorithm of the present invention.
Detailed Description
Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. It should be noted that: the relative arrangement of the components and steps, the numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present application unless specifically stated otherwise.
The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the application, its application, or uses.
Techniques, methods, and apparatus known to those of ordinary skill in the relevant art may not be discussed in detail but are intended to be part of the specification where appropriate.
In all examples shown and discussed herein, any particular value should be construed as exemplary only and not as limiting. Thus, other examples of the exemplary embodiments may have different values.
It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be discussed further in subsequent figures.
The following describes an unmanned aerial vehicle inspection path planning method based on a brainstorming optimization algorithm according to an embodiment of the invention with reference to the accompanying drawings.
As shown in fig. 1, the unmanned aerial vehicle inspection path planning method based on the brainstorming optimization algorithm according to the embodiment of the invention includes the following steps:
s1, establishing a three-dimensional environment for routing and planning of unmanned aerial vehicles, and determining the number of unmanned aerial vehicles, and the starting point and the acquisition point of each unmanned aerial vehicle;
s2, establishing an unmanned aerial vehicle inspection path planning model with corner energy consumption according to the energy consumption constraint parameters;
and S3, according to the unmanned aerial vehicle routing inspection path planning model, acquiring the optimal path of the unmanned aerial vehicle routing inspection path planning by adopting a target space clustering discrete brainstorming optimization algorithm.
Specifically, referring to fig. 1, in the unmanned aerial vehicle inspection path planning method based on the brainstorm optimization algorithm according to the embodiment of the present invention, first, a three-dimensional environment for unmanned aerial vehicle inspection path planning may be established, and the number of unmanned aerial vehicles participating in the unmanned aerial vehicle inspection path planning, the departure point and the acquisition point of each unmanned aerial vehicle, may be determined. In the Multi-UAV path planning problem (MUPP-AEC) with corner energy consumption, the Multi-UAV path planning and Multi-UAV path planning system comprises a plurality of path nodes, wherein one node is an unmanned aerial vehicle departure point, and the rest path nodes are unmanned aerial vehicle data acquisition points.
And then, establishing an unmanned aerial vehicle inspection path planning model with corner energy consumption according to the energy consumption constraint parameters. The energy consumption constraint parameters mainly comprise unmanned aerial vehicle acceleration flight time, unmanned aerial vehicle constant-speed flight time, unmanned aerial vehicle deceleration flight time, acceleration flight energy consumption between path nodes, constant-speed flight energy consumption, deceleration flight energy consumption and corner energy consumption at the path middle node. That is, unmanned aerial vehicle's departure point has a limited number of unmanned aerial vehicle, and unmanned aerial vehicle itself is equipped with rechargeable battery, has the restriction of maximum battery capacity. The unmanned aerial vehicle needs to fly along a path formed by a plurality of data acquisition points so as to execute a data acquisition task. The inter-node flight path may be represented in two-dimensional euclidean distances.
The unmanned aerial vehicle has fixed at the uniform velocity flying speed, acceleration when starting and acceleration when arriving and slowing down in the acquisition process, and the process of cruising includes starting from unmanned aerial vehicle departure point, accelerates to the speed of cruising. And then collecting the sensor data of the internet of things through airborne equipment when flying at a constant speed and arriving at each cruising place. Under the condition that the energy is enough, the unmanned aerial vehicle keeps at a constant speed and flies to the next cruising place at a corner, otherwise, the corner flies back to the starting point of the unmanned aerial vehicle and decelerates to zero when approaching the terminal point. The unmanned aerial vehicle cruising time comprises the time required by accelerated flight in the initial stage, the time required by uniform flight in the intermediate stage and the time required by decelerated flight in the final stage, and the energy consumption of the unmanned aerial vehicle comprises the energy consumption of accelerated flight among path nodes, the energy consumption of uniform flight, the energy consumption of decelerated flight and the corner energy consumption at the path intermediate nodes.
The unmanned aerial vehicle routing inspection path planning model (MUPP-AEC) can be represented by an undirected graph G ═ V, E, where V ═ {0,1, …, n } is a vertex set, where 0 is an unmanned aerial vehicle departure point, an acquisition point set C ═ V \ {0}, E { (i, j) | i, j ∈ V } is a set of edges, all unmanned aerial vehicles are of the same model, the number of unmanned aerial vehicles is K, and the unmanned aerial vehicle battery capacity is limited to Q.
And finally, acquiring the optimal path of the routing plan of the unmanned aerial vehicle by adopting a target space clustering discrete head storm optimization algorithm according to the routing plan model of the unmanned aerial vehicle. The aim of the path planning problem of the multiple unmanned aerial vehicles with the corner energy consumption is that the total cruising time of all the unmanned aerial vehicles is shortest under the conditions that the number of the unmanned aerial vehicles is limited, the energy consumption constraint of the unmanned aerial vehicles is met, and all the data acquisition points can be accessed by one unmanned aerial vehicle.
The target Space clustering Discrete head Storm Optimization Algorithm (DBSO-OS) can be directly applied to an unmanned aerial vehicle routing inspection path planning model, and the problem that the existing head Storm Optimization Algorithm cannot be directly applied to MUPP-AEC is solved. The target space clustering discrete brainstorming optimization algorithm improves the existing algorithm, adopts individual space integer coding, adopts a grading greedy decoding strategy with 2-OPT (binary optimization) to realize the conversion from an individual space to a solution space, and carries out discretization definition on a disturbance operator and an individual update operator to obtain the optimal path of the routing plan of the unmanned aerial vehicle routing inspection. The target space clustering discrete brainstorming optimization algorithm reduces the calculation load, has good convergence speed and solving precision, and can effectively solve the problem of routing inspection and scheduling of the unmanned aerial vehicle.
Therefore, according to the unmanned aerial vehicle routing inspection path planning method based on the brainstorm optimization algorithm, the unmanned aerial vehicle routing inspection path planning model with corner energy consumption can be established according to the energy consumption constraint parameters, the optimal path of the unmanned aerial vehicle routing inspection path planning is obtained by adopting the target space clustering discrete brainstorm optimization algorithm, and the problem of unmanned aerial vehicle routing inspection scheduling is effectively solved.
According to an embodiment of the invention, in the routing model for routing inspection of the unmanned aerial vehicle, the path nodes i, j and l are not assumed to be on a straight line, wherein, as shown in fig. 2, the included angle between the extension line of ij and the line segment jl is the flight rotation angle of the unmanned aerial vehicle and is marked as theta ijl EA is recorded to the corner energy consumption of unmanned aerial vehicle flight ijl ，EA ijl Is theta ijl As shown in equation (1):
EA ijl ＝f EA (θ ijl ) (1)
assuming that L is the range of the complete path of the drone, the time of flight FT and the energy consumed EN of the drone are functions of the range L, as shown in equations (2) and (3):
wherein l Acc For accelerating the course of flight L,/ Dec Course, L, of the deceleration flight course being course L Const The flight distance is the flight distance of the uniform speed flight process of the flight distance L.
In the present application, a decision variable x is defined ijk And is 1 when drone k travels from node i to j, and is 0 otherwise. Defining a decision variable y ik And is 1 when drone k serves node i, otherwise is 0. The method comprises the following steps of establishing a mathematical model of the routing planning problem of the multi-unmanned aerial vehicle with corner energy consumption as shown in formula (4) to formula (8):
wherein, c ij Is the distance from point i to point j, x ijk Is a decision variable and is 1 when the unmanned plane k travels from node i to j, otherwise is 0; y is ik Is a decision variable, and is 1 when the unmanned plane k serves the node i, otherwise is 0; the formula (4) is an objective function, and the sum of the flight times of all the paths of the unmanned aerial vehicles is minimized; formula (5) indicates that the acquisition point can only be executed by one unmanned aerial vehicle; formula (6) shows that the unmanned aerial vehicle which starts from the center is not more than K stations; the formula (7) ensures that the number of the unmanned aerial vehicles arriving and leaving at each node is balanced; and the formula (8) is that the battery capacity of the unmanned aerial vehicle is greater than the sum of the voyage energy consumption and the corner energy consumption.
And in this application, unmanned aerial vehicle patrols and examines the route planning model and still includes equation (9) -equation (11):
y ik ∈{0,1},
x ijk ∈{0,1},
wherein, the formula (9) is used for avoiding the occurrence of the sub-loop phenomenon; equations (10) and (11) represent the decision variable x ijk And y ik Is a variable of 0 to 1。
In some embodiments of the present invention, the target spatial clustering discrete brainstorming optimization algorithm comprises the following steps:
s1: randomly initializing scheme clusters X with the number of POP _ SIZE, and initializing the current iteration number to be 1;
s2: calculating the fitness of all individuals in the scheme cluster X;
s3: if the current iteration times do not reach MaxIter, jumping to the next step, otherwise, jumping to step S9;
s4: sorting the scheme groups according to the fitness, and clustering by adopting a convergence operator;
s5: executing a perturbation operator operation;
s6: carrying out individual updating on each individual in the scheme group by adopting an individual updating operator;
s7: adding 1 to the current iteration times;
s8: jumping to step S3;
s9: and returning the optimal solution.
Wherein, POP _ SIZE represents the number of original population of the scheme, and MaxIter represents the maximum iteration number.
According to one embodiment of the invention, in the target space clustering discrete brain storm optimization algorithm, each scheme individual can be defined as an n-dimensional integer vector in individual space integer coding, a single unmanned aerial vehicle route is represented by a route representation method, and acquisition point sequences in cruise routes of a plurality of unmanned aerial vehicles are connected to obtain scheme individual coding. For example, in scenario i, two drones perform patrol, and if the routes are R1 ═ 0, 4, 1, 5, 2, 8, 0 and R2 ═ 0, 7, 3, 9, 6, 0, then scenario individuals x are listed as i ＝(4，1，5，2，8，7，3，9，6)。
Optionally, in the target spatial clustering discrete brainstorming optimization algorithm, decoding is performed according to an acceleration flight phase, a constant speed flight phase and a deceleration flight phase of the unmanned aerial vehicle flight, and 2-OPT local optimization is performed on a decoding result.
Specifically, in the target space clustering discrete brainstorming optimization algorithm, decoding is to convert n-dimensional integer vectors into cruise paths of multiple unmanned aerial vehicles. In the MUPP-AEC, each drone has a battery capacity constraint, and the drone requires distance energy consumption and corner energy consumption for cruising. The distance energy consumption comprises acceleration energy consumption of the unmanned aerial vehicle starting from the unmanned aerial vehicle starting point, middle constant-speed energy consumption and deceleration energy consumption of the unmanned aerial vehicle returning to the unmanned aerial vehicle starting point. The unmanned aerial vehicle routing inspection path planned based on the brainstorm optimization algorithm needs to ensure that the unmanned aerial vehicle leaves enough battery energy capacity to return to the center after finishing cruising of all acquisition points.
As shown in fig. 3, the encoding of an individual is n-dimensional sequence (x) 1 ，x 2 ，…，x n ) Node 0 represents the command center, which contains all the acquisition points. The distance matrix is a two-dimensional square matrix DT with elements (i, j) representing the distances from i to j, the angle matrix is a three-dimensional matrix AN with elements (i, j, l) representing θ ijl And j is a vertex. At uniform velocity SP c At uniform power PW c . Unmanned plane from 0 to SP c Has a full acceleration time of T a The total acceleration distance is D a Total acceleration energy consumption is E a . Slave SP c The complete deceleration time to 0 is T d The total deceleration distance is D d With total retarding energy consumption of E d 。EA ijl Represents an angle theta ijl And (4) required corner energy consumption. Current battery power consumption is P bat The elapsed time is T.
The result of the decoding is a plurality of cruise routes, each assigned to a drone. Setting a path matrix R for storing a plurality of cruise routes, wherein the row number of R is equal to the available number of the unmanned aerial vehicles, the column number is n +1, the current row index of the matrix is set to be R, and the current column index of the matrix is p. The phased greedy decoding with 2-opt is divided into an acceleration phase, a uniform velocity phase and a deceleration phase, and referring to fig. 3, the detailed steps are described as follows:
step 1: initializing flight time, making flight distance zero and current battery energy consumption P bat ＝Q；
Step 2: unmanned aerial vehicle accelerates from 0 to SP c The flight path of which is added to D a Time of flight plus T a Unmanned plane P bat Subtract E a (ii) a If the middle passes through the node, P bat Subtracting the corner energy consumption passing through the node; after the unmanned aerial vehicle accelerates, calculating the distance from the unmanned aerial vehicle to the next node as the residual constant-speed distance;
and step 3: calculating the time required by the remaining uniform distance and the required energy consumption, the flight time plus the required time, P bat The required energy consumption is reduced; setting the next node as the current node;
and 4, step 4: calculating the safe total energy consumption of the next node of the current node as the corner energy consumption of the current node, the constant-speed energy consumption of the voyage from the current node to the next node, the energy consumption from the next node to the terminal point and the corner energy consumption of the next node; judgment of P bat Whether it is greater than the safe total energy consumption, if P bat More than or equal to the total safety energy consumption, the flight distance plus the range from the current node to the next node, the flight time plus the uniform speed time from the current node to the next node, P bat Subtracting the current node corner energy consumption and the current point to next node voyage constant-speed energy consumption, and skipping to the next step; if P bat If the total energy consumption is less than the safe total energy consumption, jumping to step 6 to execute the unmanned aerial vehicle return stroke step;
and 5: setting the next node as the current node; judging whether the current node is the last node or not; if the node is the last node, jumping to step 6 to execute the unmanned aerial vehicle return trip step, otherwise, jumping to step 4;
and 6: calculating the residual uniform speed distance as the distance from the current node to the terminal minus the complete deceleration distance D d Calculating the residual uniform velocity time and the residual uniform velocity energy consumption of the residual uniform velocity distance, adding the flight distance to the current node to the terminal flight distance, adding the flight time to the residual uniform velocity distance and the T d Sum of P bat Subtracting the residual constant energy consumption and E d Summing; and saving the current cruising route.
And 7: judging whether the current node is the last node or not, and outputting all paths if the current node is the last node; otherwise, adding 1 to the number of the unmanned aerial vehicles, and skipping to the step 1.
According to one embodiment of the invention, in the target space clustering discrete brainstorming optimization algorithm, a perturbation operator adopts a random individual clustering center replacement strategy for discretization definition. And the individual updating operator carries out discretization definition on the individual generating operation by adopting a random inversion transformation strategy and a partial matching transformation generating strategy, and carries out discretization definition on the individual updating operation by adopting an integral substitution updating strategy.
Specifically, the target space clustering discrete brainstorm optimization algorithm can simulate the process of convergence and divergence of a preferred scheme obtained by people by adopting a brainstorm method, wherein convergence refers to the fact that an organizer clusters all solutions and assigns a center in each class, and divergence refers to the fact that a plurality of experts generate new solutions for problems according to professional backgrounds of the experts. The target space clustering discrete brainstorming optimization algorithm carries out clustering according to the target space fitness of each solution, and main operators can comprise convergence operators, disturbance operators and individual updating operators.
In a target spatial clustering Discrete brainstorm Optimization Algorithm (DBSO-OS) of the application, the ELITE _ PERC of the proportion of essence in a scheme population is set, the Algorithm performs all individual sorting according to all individual fitness values by adopting a quick sorting method, the individuals of the front ELITE _ PERC are classified into the ELITE individuals, and the rest individuals are classified into the common individuals. And defining a clustering center for each of the two types of individuals, wherein the individual with the best fitness among the elite type individuals is defined as the elite type center, and the rest are members of the elite type.
In order to adapt to discretization operation, a DBSO-OS perturbation operator adopts a random individual clustering center replacement strategy. Setting the probability of disturbance P d Generating a random number in each iteration, if the random number is less than P d Then the perturbation is performed. And randomly generating a new individual according to a coding rule during disturbance, decoding the new individual to obtain path fitness, randomly selecting a clustering center between the elite type center and the common type center, comparing the new individual fitness with the individual fitness of the selected clustering center, and replacing the new individual with the clustering center if the new individual is better than the selected clustering center.
The DBSO-OS individual updating operator adopts a random inversion transformation, partial matching transformation generation strategy and an integral substitution updating strategy, and discretization definition is carried out on individual generation operation and individual updating operation. New individuals are generated from one to two selected individuals who are both elite or both common. In contrast, an individually generated discretization improvement is employed. The DBSO-OS first determines whether to base one or both selected individuals, and the second determines whether to generate a new individual based on an elite class or a normal class. Algorithm sets individual probabilities P one Deciding whether to generate a new individual, elite individual probability P, based on one selected individual or two selected individuals e For deciding whether to create new individuals by elite individuals or by normal individuals.
When a random selection of a class center or class member is used, a new individual is generated by a random inversion transformation on the selected single individual as shown in fig. 4. To produce [1, n]Two different random integers of the interval, the smaller of which is the inversion starting position p start Larger as the inversion termination position p end Number of individuals is p start To p end The sequence in between is reversed. Suppose an individual x i (5, 1, 9, 6, 8, 3, 7, 2, 4), p is randomly obtained start ＝3，p end Transform result x, 7 i new ＝(5,1,7,3,8,6,9,2,4)。
When two class centers or class members are selected, as shown in fig. 5, a partial matching transformation operation is performed on the two selected individuals to generate new individuals. Let two individuals be x respectively 1 、x 2 Generating [1, n ]]Two different random integers of the interval, the smaller of which is used as the starting position p of the partial matching transformation start Larger as the partial matching transformation end position p end ，x 1 、x 2 Number p start To p end Match elements between them one by one, each matching forms a number pair<n 1 ，n 2 >At x 2 To pair<n 1 ，n 2 >Pairs of exchange positions, let x 2 P of (a) start To p end X and 1 the same is true. Suppose an individual x 1 ＝(5，1，9，6，8，3，7，2，4)，x 2 (8, 5, 4, 3, 7, 9, 6, 1, 2), p is randomly obtained start ＝4，p end Transform result x 6 2 ＝(7，5，4，6，8，3，9，1，2)。
As shown in fig. 6, the individual updating operation adopts an overall replacement updating mode, for each individual in the scheme group, a new individual obtained by the individual generating operation is adopted, the new individual is decoded to obtain the path fitness, the new individual fitness and the current position individual fitness are compared, and if the new individual is better than the current position individual, the current position individual is replaced by the new individual.
According to one embodiment of the invention, the fitness function of the target space clustering discrete head storm optimization algorithm is as shown in formula (12):
wherein f is FT A function representing the time of flight and range L of the drone, c ij Is the distance from point i to point j, x ijk Representing the decision variables.
That is to say, the target spatial clustering discrete brainstorming optimization algorithm adopts the corresponding fitness function to evaluate the scheme group, so that a more optimal individual is selected. From the fitness function formula (12), it can be seen that the smaller the cost of Multi-UAV path planning with angular energy consumption Multi-UAV routing with angular energy consumption path planning (MUPP-AEC), the better the fitness, and the scheme has higher probability to replace the original scheme in the iterative process of the algorithm.
In summary, according to the unmanned aerial vehicle routing inspection path planning method based on the brainstorm optimization algorithm, an unmanned aerial vehicle routing inspection path planning model with corner energy consumption can be established according to the energy consumption constraint parameters, the optimal path of the unmanned aerial vehicle routing inspection path planning is obtained by adopting the target space clustering discrete brainstorm optimization algorithm, and the problem of unmanned aerial vehicle routing inspection scheduling is effectively solved.
Although some specific embodiments of the present application have been described in detail by way of example, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present application. It will be appreciated by those skilled in the art that modifications can be made to the above embodiments without departing from the scope and spirit of the present application. The scope of the application is defined by the appended claims.
Claims (8)
1. An unmanned aerial vehicle routing inspection path planning method based on a brainstorming optimization algorithm is characterized by comprising the following steps:
establishing a three-dimensional environment for routing and planning of the unmanned aerial vehicles, and determining the number of the unmanned aerial vehicles, and a departure point and a collection point of each unmanned aerial vehicle;
establishing an unmanned aerial vehicle routing inspection path planning model with corner energy consumption according to the energy consumption constraint parameters;
according to the unmanned aerial vehicle routing inspection path planning model, acquiring an optimal path of the unmanned aerial vehicle routing inspection path planning by adopting a target space clustering discrete head storm optimization algorithm;
the energy consumption constraint parameters comprise unmanned aerial vehicle accelerated flight time, unmanned aerial vehicle constant-speed flight time, unmanned aerial vehicle decelerated flight time, accelerated flight energy consumption among path nodes, constant-speed flight energy consumption, decelerated flight energy consumption and corner energy consumption at path intermediate nodes;
the unmanned aerial vehicle routing inspection path planning model is represented by an undirected graph G (V, E), wherein V is {0,1, …, n } and is a vertex set, wherein 0 is an unmanned aerial vehicle departure point, a collection point set C is V \ 0}, E is { (i, j) | i, j ∈ V } and is a set of edges, all unmanned aerial vehicles are of the same type, the number of the unmanned aerial vehicles is K, and the battery capacity of the unmanned aerial vehicles is limited to Q;
the target space clustering dispersion head storm optimization algorithm is directly applied to the unmanned aerial vehicle routing inspection path planning model, the algorithm adopts individual space integer coding, a staged greedy decoding strategy with 2-OPT is adopted to realize conversion from an individual space to a solution space, and discretization definition is carried out on a disturbance operator and an individual update operator, so that an optimal path of the unmanned aerial vehicle routing inspection path planning is obtained;
in the unmanned aerial vehicle routing inspection path planning model, path nodes i, j and l are not assumed to be on a straight line, wherein the included angle between the extension line of ij and the line segment jl is the flight rotation angle of the unmanned aerial vehicle and is recorded as theta ijl And recording the corner energy consumption of the unmanned aerial vehicle flying as EA ijl ，EA ijl Is theta ijl As shown in equation (1):
EA ijl ＝f EA (θ ijl ) (1)
assuming that L is the range of the complete path of the drone, the time of flight FT and the energy consumed EN of the drone are functions of the range L, as shown in equations (2) and (3):
wherein l Acc The flight path of the accelerated flight path L Dec Course, L, of the deceleration flight course being course L Const The voyage is the uniform speed flight process of the voyage L;
the unmanned aerial vehicle inspection path planning model is as shown in the formula (4) to the formula (8):
wherein, c ij Is the distance from point i to point j, x ijk Is a decision variable and is 1 when the unmanned plane k travels from node i to j, otherwise is 0; y is ik Is a decision variable, and is 1 when the unmanned plane k serves the node i, otherwise is 0; equation (4) is an objective function, so that the sum of the flight times of all unmanned aerial vehicle paths is minimum; formula (5) indicates that the acquisition point can only be executed by one unmanned aerial vehicle; formula (6) shows that the unmanned aerial vehicles departing from the center are not more than K; the equation (7) ensures that the number of the unmanned aerial vehicles arriving and leaving at each node is balanced; and the formula (8) is that the battery capacity of the unmanned aerial vehicle is greater than the sum of the voyage energy consumption and the corner energy consumption.
2. The unmanned aerial vehicle inspection path planning method based on the brainstorm optimization algorithm according to claim 1, wherein the unmanned aerial vehicle inspection path planning model further comprises formula (9) -formula (11):
wherein, the formula (9)For avoiding sub-loop phenomenon; equations (10) and (11) represent the decision variables x ijk And y ik Is a variable from 0 to 1.
3. The unmanned aerial vehicle inspection path planning method based on the brainstorming optimization algorithm of claim 1, wherein the target space clustering discrete brainstorming optimization algorithm comprises the following steps:
s1: randomly initializing scheme clusters X with the number of POP _ SIZE, and initializing the current iteration number to be 1;
s2: calculating the fitness of all individuals in the scheme cluster X;
s3: if the current iteration times do not reach MaxIter, jumping to the next step, otherwise, jumping to step S9;
s4: sorting the scheme groups according to the fitness, and clustering by adopting a convergence operator;
s5: performing a perturbation operator operation;
s6: carrying out individual updating on each individual in the scheme group by adopting an individual updating operator;
s7: adding 1 to the current iteration times;
s8: jumping to step S3;
s9: returning to an optimal solution;
wherein, POP _ SIZE represents the number of original population of the scheme, and MaxIter represents the maximum iteration number.
4. The unmanned aerial vehicle inspection path planning method based on the brainstorm optimization algorithm according to claim 2, wherein in the target space clustering discrete brainstorm optimization algorithm, each scheme individual in the individual space integer coding is an n-dimensional integer vector, a single unmanned aerial vehicle route is represented by a path representation method, and acquisition point sequences in cruise paths of a plurality of unmanned aerial vehicles are connected to obtain scheme individual codes.
5. The unmanned aerial vehicle inspection path planning method based on the brainstorming optimization algorithm of claim 3, wherein in the target space clustering dispersion brainstorming optimization algorithm, decoding is performed according to an acceleration flight phase, a constant speed flight phase and a deceleration flight phase of the unmanned aerial vehicle flight, and 2-OPT local optimization is performed on a decoding result.
6. The unmanned aerial vehicle inspection path planning method based on the brainstorming optimization algorithm of claim 4, wherein in the target space clustering dispersion brainstorming optimization algorithm, the perturbation operator performs discretization definition by adopting a random individual clustering center replacement strategy.
7. The unmanned aerial vehicle inspection path planning method based on the brainstorming optimization algorithm of claim 5, wherein in the target space clustering dispersion brainstorming optimization algorithm, a random inversion transformation strategy and a partial matching transformation generation strategy are adopted in the individual updating operator to carry out discretization definition on individual generation operation, and an integral substitution updating strategy is adopted to carry out discretization definition on the individual updating operation.
8. The unmanned aerial vehicle inspection path planning method based on the brainstorming optimization algorithm of claim 6, wherein a fitness function of the target space clustering discrete brainstorming optimization algorithm is as shown in formula (12):
wherein f is FT A function representing the time of flight and range L of the drone, c ij Is the distance from point i to point j, x ijk Representing the decision variables.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110368928.5A CN113203417B (en) | 2021-04-06 | 2021-04-06 | Unmanned aerial vehicle inspection path planning method based on brain storm optimization algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110368928.5A CN113203417B (en) | 2021-04-06 | 2021-04-06 | Unmanned aerial vehicle inspection path planning method based on brain storm optimization algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113203417A CN113203417A (en) | 2021-08-03 |
| CN113203417B true CN113203417B (en) | 2022-08-19 |
Family
ID=77026301
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110368928.5A Active CN113203417B (en) | 2021-04-06 | 2021-04-06 | Unmanned aerial vehicle inspection path planning method based on brain storm optimization algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113203417B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115099675B (en) * | 2022-07-14 | 2022-11-15 | 海门市三德体育用品有限公司 | Layout planning method for fitness place |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107238388B (en) * | 2017-05-27 | 2018-02-23 | 合肥工业大学 | Multiple no-manned plane task is distributed and trajectory planning combined optimization method and device |
| CN109520504B (en) * | 2018-11-27 | 2022-07-22 | 北京航空航天大学 | Grid discretization-based unmanned aerial vehicle patrol route optimization method |
| CN109682380B (en) * | 2019-01-16 | 2020-01-10 | 北京邮电大学 | Communication unmanned aerial vehicle path optimization method and device |
| CN109933089B (en) * | 2019-03-25 | 2020-10-23 | 北京邮电大学 | Multi-unmanned aerial vehicle task planning method and device based on minimum maximum energy consumption |
| CN109917815B (en) * | 2019-04-23 | 2022-02-22 | 中原工学院 | Unmanned aerial vehicle three-dimensional path design method based on global optimal brainstorming algorithm |
| FR3098336B1 (en) * | 2019-07-01 | 2022-08-12 | Uavia | Method for determining the path of an unmanned aerial device and other associated methods |
| CN111091286B (en) * | 2019-12-13 | 2022-08-16 | 河海大学常州校区 | Public bicycle scheduling method |
| CN111121784B (en) * | 2019-12-24 | 2023-03-14 | 中国航空工业集团公司西安飞机设计研究所 | Unmanned reconnaissance aircraft route planning method |
| CN111536979B (en) * | 2020-07-08 | 2020-10-30 | 浙江浙能天然气运行有限公司 | Unmanned aerial vehicle routing inspection path planning method based on random optimization |
-
2021
- 2021-04-06 CN CN202110368928.5A patent/CN113203417B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113203417A (en) | 2021-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jia et al. | Cooperative multiple task assignment problem with stochastic velocities and time windows for heterogeneous unmanned aerial vehicles using a genetic algorithm | |
| Zhu et al. | Multi-UAV reconnaissance task allocation for heterogeneous targets using an opposition-based genetic algorithm with double-chromosome encoding | |
| Chen et al. | Multi-UAV task assignment with parameter and time-sensitive uncertainties using modified two-part wolf pack search algorithm | |
| Deng et al. | Cooperative task assignment of multiple heterogeneous unmanned aerial vehicles using a modified genetic algorithm with multi-type genes | |
| Shima et al. | UAV cooperative multiple task assignments using genetic algorithms | |
| Fu et al. | Route planning for unmanned aerial vehicle (UAV) on the sea using hybrid differential evolution and quantum-behaved particle swarm optimization | |
| CN112683276B (en) | Unmanned aerial vehicle routing inspection cable path planning method based on mixed discrete wolf algorithm | |
| CN111813144B (en) | Multi-unmanned aerial vehicle collaborative route planning method based on improved flocks of sheep algorithm | |
| CN112733251B (en) | Collaborative flight path planning method for multiple unmanned aerial vehicles | |
| Xu et al. | Deep reinforcement learning approach for joint trajectory design in multi-UAV IoT networks | |
| Jin et al. | Optimal scheduling for refueling multiple autonomous aerial vehicles | |
| CN112437502A (en) | Hierarchical clustering network topology structure generation method based on multitask unmanned aerial vehicle cluster information interaction | |
| CN113203417B (en) | Unmanned aerial vehicle inspection path planning method based on brain storm optimization algorithm | |
| CN111121784A (en) | Unmanned reconnaissance aircraft route planning method | |
| Rahili et al. | Optimal routing for autonomous taxis using distributed reinforcement learning | |
| CN117055619A (en) | Unmanned aerial vehicle scheduling method based on multi-agent reinforcement learning | |
| Xiang et al. | An effective memetic algorithm for UAV routing and orientation under uncertain navigation environments | |
| Gao et al. | Cooperative multi-aircraft conflict resolution based on co-evolution | |
| Shen | Bionic communication network and binary pigeon-inspired optimization for multiagent cooperative task allocation | |
| Asghar et al. | Risk-aware recharging rendezvous for a collaborative team of uavs and ugvs | |
| Tao et al. | Decentralized Multi⁃ agent Task Planning for Heterogeneous UAV Swarm. | |
| CN114518772B (en) | Unmanned aerial vehicle swarm self-organization method in rejection environment | |
| CN115826614A (en) | Multi-unmanned aerial vehicle energy support task allocation method | |
| Wu et al. | A time-efficient and attention-aware deployment strategy for UAV networks driven by deep reinforcement learning | |
| CN114885379A (en) | Large-scale unmanned aerial vehicle cluster self-adaptive clustering networking method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |