CN116301031A - Multi-unmanned aerial vehicle collaborative track planning method based on self-adaptive differential evolution algorithm - Google Patents

Multi-unmanned aerial vehicle collaborative track planning method based on self-adaptive differential evolution algorithm Download PDF

Info

Publication number
CN116301031A
CN116301031A CN202310140529.2A CN202310140529A CN116301031A CN 116301031 A CN116301031 A CN 116301031A CN 202310140529 A CN202310140529 A CN 202310140529A CN 116301031 A CN116301031 A CN 116301031A
Authority
CN
China
Prior art keywords
unmanned aerial
aerial vehicle
constraint
points
representing
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.)
Pending
Application number
CN202310140529.2A
Other languages
Chinese (zh)
Inventor
胡敏
黄刚
杨学颖
林鹏
陈韬鸣
齐晶
黄飞耀
郭雯
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Peoples Liberation Army Strategic Support Force Aerospace Engineering University
Original Assignee
Peoples Liberation Army Strategic Support Force Aerospace Engineering University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Peoples Liberation Army Strategic Support Force Aerospace Engineering University filed Critical Peoples Liberation Army Strategic Support Force Aerospace Engineering University
Priority to CN202310140529.2A priority Critical patent/CN116301031A/en
Publication of CN116301031A publication Critical patent/CN116301031A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T19/00—Manipulating 3D models or images for computer graphics
    • G06T19/003—Navigation within 3D models or images
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/10—Simultaneous control of position or course in three dimensions
    • G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft
    • G05D1/104—Simultaneous control of position or course in three dimensions specially adapted for aircraft involving a plurality of aircrafts, e.g. formation flying
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T19/00—Manipulating 3D models or images for computer graphics
    • G06T19/006—Mixed reality
    • 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)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Hardware Design (AREA)
  • Computer Graphics (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Theoretical Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Automation & Control Theory (AREA)
  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

The application discloses a multi-unmanned aerial vehicle collaborative track planning method based on a self-adaptive differential evolution algorithm, which comprises the following steps: constructing a three-dimensional simulation space of a mountain type and a radar scanning area, setting auxiliary track points, and calculating the fitness of key track points; constructing single-machine constraint of the unmanned aerial vehicle and cooperative constraint conditions of the unmanned aerial vehicle, and establishing a multi-machine cooperative track planning objective function; constructing a self-adaptive rule of path planning of multiple unmanned aerial vehicles; and selecting a variation strategy according to the self-adaptive degree value, deleting or supplementing a task point sequence meeting the conditions by adopting a reasonable coding correction rule, taking the departure points, auxiliary track points, key track points and task points of each unmanned aerial vehicle as control points of spline curves, and approximating to obtain a fitted smooth track of the unmanned aerial vehicle. The invention utilizes the three-dimensional space feasible domain to reduce the complexity of the search space; the blindness of algorithm searching is reduced by utilizing the cooperative constraint of multiple unmanned aerial vehicles; by generating the key track points, the algorithm searching efficiency is improved.

Description

Multi-unmanned aerial vehicle collaborative track planning method based on self-adaptive differential evolution algorithm
Technical Field
The application relates to a multi-unmanned aerial vehicle collaborative track planning method, belongs to the technical field of track planning, and particularly relates to a multi-unmanned aerial vehicle collaborative track planning method based on a self-adaptive differential evolution algorithm.
Background
The multi-unmanned aerial vehicle collaborative track planning is to plan a plurality of safe, reasonable and collision-free flyable tracks according to the space positions of unmanned aerial vehicles, task points and threat obstacles in a complex planning environment. Due to limited maneuver performance and payload, single-frame unmanned aerial vehicle track planning is difficult to accomplish complex military tasks. Compared with the prior art, the collaborative flight path planning of the multiple unmanned aerial vehicles has the advantages of stronger flexibility, higher task execution efficiency and the like, and gradually replaces a single unmanned aerial vehicle to execute complex tasks. However, multi-drone collaborative track planning also increases the difficulty of the combat system, for example: in the calculation process, a large amount of forbidden flight data is generated in the complex mountain region and the radar detection region, and the combat system is difficult to analyze the data in a short time; in the collaborative planning process, constraint conditions of the heterogeneous unmanned aerial vehicles and mutual constraints among the multiple unmanned aerial vehicles may cause execution relation conflict between each unmanned aerial vehicle and task points, and finally cause task failure and the like.
At present, in order to improve the efficiency of multi-unmanned aerial vehicle collaborative track planning, research on solving the problem of multi-unmanned aerial vehicle collaborative track planning at home and abroad is mainly focused on collaborative track planning based on intelligent optimization algorithm, collaborative track planning based on reinforcement machine learning and collaborative track planning based on spline interpolation algorithm. Liang Xu et al propose a multi-machine collaborative flight path planning based on task demand constraints, and the algorithm proposes a new dynamic particle swarm algorithm and a comprehensive learning particle swarm algorithm to improve the execution speed of the multi-machine collaborative flight path planning. Liu Yi et al propose to predict flight path by using a closed loop model of the unmanned aerial vehicle 6 degree of freedom nonlinear model, and simultaneously use a path mapping network based on deep learning to improve planning calculation speed and prediction accuracy. Zong Xuepeng et al propose a track planning framework in a three-dimensional dynamic environment, design an unmanned aerial vehicle trajectory predictor using a least squares method, and then represent the generated track using a segmented Bezier curve.
Disclosure of Invention
In order to solve the defects in the prior art, the application provides a low-complexity and high-efficiency multi-unmanned-plane collaborative track planning method, and the collaborative performance of unmanned planes is improved.
The multi-unmanned aerial vehicle collaborative track planning method based on the self-adaptive differential evolution algorithm comprises the following steps:
constructing a three-dimensional simulation space comprising an unmanned plane, task points and barriers;
constructing single-machine constraint of the unmanned aerial vehicle and cooperative constraint conditions of the unmanned aerial vehicle, and establishing a multi-machine cooperative track planning objective function;
respectively constructing vertical tangent planes between each unmanned aerial vehicle and each task point according to the positions of the unmanned aerial vehicle and the task points in the three-dimensional simulation space, expanding a three-dimensional space feasible region according to the feasible region parameters, setting auxiliary track points, and generating a key track point set according to the adaptability values of candidate key track points in the feasible region;
constructing a self-adaptive rule of multi-unmanned aerial vehicle track planning, wherein each individual automatically selects a variation strategy meeting the conditions according to the self-adaptive degree value, and deleting or supplementing a task point sequence meeting the conditions by adopting a coding correction rule;
and taking the departure points, auxiliary track points, key track points and task points of each unmanned aerial vehicle as control points of spline curves, and approximating to obtain a fitted smooth track of the unmanned aerial vehicle.
Preferably, the constructing a three-dimensional simulation space includes: and constructing a radar scanning area and fusing the barriers.
Preferably, the constructed radar scan area is represented by the following formula:
Figure BDA0004088771170000021
Figure BDA0004088771170000022
wherein, (x) i ,y i ,z i ) Representing the position of the auxiliary track point in space, (x) r ,y r ,z r ) Representing the position of the radar in space, d r Representing the distance between the auxiliary track point and the radar, R r Represents the scanning radius of the radar, k represents the threat level of the radar, F Radar Representing threat costs of radar;
preferably, the method for acquiring the obstacle comprises the following steps: the USGS24kdem is adopted to read the real USGS 1:24 000 digital elevation map file, and the file is used as a Terrain obstacle Terrain.
Preferably, the unmanned aerial vehicle stand-alone constraint includes: unmanned aerial vehicle and mission point decision variable constraint, unmanned aerial vehicle maximum voyage constraint, unmanned aerial vehicle minimum/maximum speed constraint, maximum voyage time constraint and unmanned aerial vehicle maximum loading constraint.
Preferably, the unmanned aerial vehicle cooperative constraint includes: and (3) time sequence constraint among task points and waiting time constraint of each unmanned aerial vehicle, and determining that each constraint violates penalty.
Preferably, each unmanned aerial vehicle must execute a task point, each task point must be executed by one unmanned aerial vehicle, and the unmanned aerial vehicle and task point decision variable constraints are constructed according to the following formula:
Figure BDA0004088771170000031
wherein N represents the number of unmanned aerial vehicles, M represents the number of task points, and D (i,j) Representing the corresponding relation of the execution task points j of the unmanned aerial vehicle i;
preferably, the maximum range constraint of the unmanned aerial vehicle refers to the sum of range costs of the execution task points of each unmanned aerial vehicle, and the formula is as follows:
Figure BDA0004088771170000032
Figure BDA0004088771170000033
wherein d (i,j) Representing voyage cost, D i Representing a defined maximum range distance, C, for each unmanned aerial vehicle i 1violation Indicating that the unmanned aerial vehicle i punishs when exceeding the limited maximum range distance;
preferably, the unmanned aerial vehicle minimum/maximum speed constraint is as follows:
V (i) ＝[V (i)min ,V (i)max ]
Figure BDA0004088771170000034
wherein V is (i)min Representing the minimum speed of unmanned plane i, V (i)max Represents the maximum speed of unmanned plane i, C 2violation The unmanned plane i is punished by exceeding the defined minimum/maximum speed;
preferably, the maximum voyage time constraint T max The following formula is shown:
Figure BDA0004088771170000035
Figure BDA0004088771170000036
wherein t is (i,j) Representing the time cost of the unmanned plane i reaching the task point j, D (i,j) Representing corresponding relation of execution task points j of unmanned plane i, T max Representing the maximum travel time constraint of unmanned plane i, v i Indicating the set sailing speed of the unmanned plane i, T i Representing the actual unmanned plane iCourse time, C 3violation The unmanned aerial vehicle i is punished when exceeding the limited maximum navigation time;
preferably, the unmanned aerial vehicle maximum-load constraint U missile The following formula is shown:
U missile (i)≤U max (i)
Figure BDA0004088771170000041
wherein U is max Representing the maximum loading capacity of the unmanned plane i, C 4violation Indicating that unmanned plane i penalizes beyond its maximum loading.
Preferably, the timing constraints T between task points sort The following formula is shown:
Figure BDA0004088771170000042
Figure BDA0004088771170000043
wherein T is sort (j) Representing the time sequence of the task point j, wherein tau is a positive integer, the task point j must be executed later than the task point j+tau, C 5violation Indicating that the task point is not beaten and punished according to the set time sequence;
preferably, the unmanned aerial vehicle latency constraint T wait The following formula is shown:
T wait (i)≤T max (i)
Figure BDA0004088771170000044
wherein T is max (i) Representing maximum waiting time of unmanned plane i, C 6violation Indicating that the unmanned aerial vehicle i has exceeded the maximum waiting time for punishment.
Preferably, the objective function for establishing the multi-machine collaborative track planning is as follows:
Figure BDA0004088771170000045
wherein, minf (x) represents an objective function of multi-unmanned aerial vehicle collaborative track planning, D (i,j) Representing a decision variable, d, between the drone i and the task point j (i,j) Representing the voyage cost of unmanned plane i reaching task point j, t (i,j) Representing the time cost of the unmanned aerial vehicle i reaching the task point j, c l Is a penalty function corresponding to the constraint, μ, λ, and τ are scaling factors of the voyage cost, the time cost, and the penalty function, respectively, and the three are kept in the same order.
Preferably, the auxiliary track points are arranged according to the maneuvering performance of the fixed wing unmanned aerial vehicle, and the flying included angle between the auxiliary track points is 120 degrees or more and 180 degrees or less.
Preferably, the constructing the adaptive rule of the multi-unmanned aerial vehicle track planning, each individual self-selects a variation strategy meeting the conditions according to the self-adaptive degree value, and the method comprises the following steps:
randomly dividing the whole population into a plurality of populations with the same number of individuals, calculating the fitness value of each parent population individual and carrying out preliminary classification;
constructing a mutation strategy archive to store three mutation strategies with different characteristics:
mutation strategy I: DE/rand/1, which is beneficial to improving individual exploratory property;
mutation strategy II: DE/target to best/2, which is beneficial to balancing exploratory and developability of population;
mutation strategy III: DE/target to best/2, which is beneficial to enhancing individual developability;
and calculating the fitness value of each individual, and adaptively selecting a mutation strategy.
The coding correction rule is used for correcting invalid discrete values, and can directly map invalid sequences to obtain new discrete sequences.
Preferably, the coding correction law satisfies the following condition:
in the correction process, the mapped discrete sequence must be unique;
the mapped invalid discrete sequences no longer participate in subsequent variant strategy calculations.
Preferably, the spline curve is a cubic B-spline curve.
The beneficial effects that this application can produce include:
1) The method and the device utilize the three-dimensional space feasible domain, greatly reduce the complexity of the search space, and play a key role in solving candidate key track points;
2) According to the method, the constraint of a single unmanned aerial vehicle and the cooperative constraint of multiple unmanned aerial vehicles are considered, the blindness of algorithm searching is reduced, the constraint violation amount (punishment function) in the planning process is increased, and the objective function can display the cooperative performance of each unmanned aerial vehicle;
3) According to the method, the distribution scheme of the unmanned aerial vehicle and the task points is increased, the self-adaptive rule and the coding correction rule are established to share the information such as the fitness value and the like to generate the key track points, the individual exploratory property and the development property are balanced, and the algorithm searching efficiency is improved.
Drawings
FIG. 1 is a flow chart of a method for solving a collaborative flight path planning party of a plurality of unmanned aerial vehicles based on a feasible region space self-adaptive differential evolution algorithm in one embodiment of the application;
FIG. 2 is a simplified space structure method diagram of utilizing a flying feasibility domain for a complex three-dimensional space in one embodiment of the present application;
FIG. 3 is a diagram of a strategy for adaptively selecting variation by calculating fitness values for each individual in a population in one embodiment of the present application;
fig. 4 is a sequence diagram of task points deleted or supplemented with a reasonable code correction rule according to an embodiment of the present application.
Detailed Description
The present application is described in detail below with reference to examples, but the present application is not limited to these examples.
A multi-unmanned aerial vehicle collaborative track planning method based on a self-adaptive differential evolution algorithm is shown in fig. 1, and specifically comprises the following steps:
step 1: the USGS24kdem is adopted to read a USGS 1:24 000 Digital Elevation Map (DEM) file in a standard format as a terrain obstacle, a radar scanning area is constructed, different obstacles are fused, and the principle formula of the radar scanning area is as follows:
constructing a radar scanning area:
Figure BDA0004088771170000061
Figure BDA0004088771170000062
wherein, (x) i ,y i ,z i ) Representing the position of the auxiliary track point in space, (x) r ,y r ,z r ) Representing the position of the radar in space, d r Representing the distance between the auxiliary track point and the radar, R r Represents the scanning radius of the radar, k represents the threat level of the radar, F Radar Representing the threat cost of radar.
Step 2: and constructing decision variables of unmanned aerial vehicles and task points, constructing course constraints, minimum/large speed constraints, time constraints and maximum bullet load self constraints of all unmanned aerial vehicles, constructing cooperative constraints of task execution time sequences and waiting time of all unmanned aerial vehicles, and constructing a multi-unmanned aerial vehicle cooperative track planning objective function with course cost, time cost and violation of the constraints.
(1) Constructing a single constraint condition of the unmanned aerial vehicle:
1) Unmanned aerial vehicle and task point decision variable constraint: each unmanned aerial vehicle must perform a task point, and each task point must be performed by one unmanned aerial vehicle. N represents the number of unmanned aerial vehicles, and M represents the number of task points.
Figure BDA0004088771170000071
2) Maximum voyage constraint: the maximum range constraint refers to the sum of the range costs of each unmanned aerial vehicle executing the task point.
Figure BDA0004088771170000072
Figure BDA0004088771170000073
d (i,j) Representing voyage cost, D i Representing a defined maximum range distance, C, for each unmanned aerial vehicle i 1violation Indicating that the unmanned aerial vehicle i punishs when exceeding the limited maximum range distance;
3) Unmanned plane imum/maximum speed constraint V (km/h):
V (i) ＝[V (i)min ,V (i)max ]
Figure BDA0004088771170000074
V (i)min representing the minimum speed of unmanned plane i, V (i)max Represents the maximum speed of unmanned plane i, C 2violation The unmanned plane i is punished by exceeding the defined minimum/maximum speed;
4) Maximum travel time constraint T max : each unmanned aerial vehicle i performs a maximum of time consuming task.
Figure BDA0004088771170000075
Figure BDA0004088771170000076
t (i,j) Representing the time cost of the unmanned plane i reaching the task point j, D (i,j) Representing corresponding relation of execution task points j of unmanned plane i, T max Representing the maximum travel time constraint of unmanned plane i, v i Indicating the set sailing speed of the unmanned plane i, T i Representing the actual voyage time of the unmanned plane i, C 3violation The unmanned aerial vehicle i is punished when exceeding the limited maximum navigation time;
5) Unmanned plane i maximum loading constraint U missile : the method is used for reflecting the maximum loading capacity of the unmanned aerial vehicle i.
U missile (i)≤U max (i)
Figure BDA0004088771170000077
U max Representing the maximum loading capacity of the unmanned plane i, C 4violation Indicating that unmanned plane i penalizes beyond its maximum loading.
(2) Constructing a multi-unmanned aerial vehicle cooperative constraint condition:
1) Time sequence constraint T between task points sort : the method is used for reflecting the importance degree of the task points, the important task points are executed preferentially, and other task points are executed sequentially according to the constraint of the important task points;
Figure BDA0004088771170000081
Figure BDA0004088771170000082
T sort (j) Representing the time sequence of the task point j, wherein tau is a positive integer, the task point j must be executed later than the task point j+tau, C 5violation Indicating that the task point is not beaten and punished according to the set time sequence;
2) Latency constraint T wait : in order to ensure that the unmanned aerial vehicle i reaches a specified target, part of unmanned aerial vehicles i are allowed to wait for a period of time before starting.
T wait (i)≤T max (i)
Figure BDA0004088771170000083
T max (i) Representing unmanned plane iMaximum waiting time, C 6violation Indicating that the unmanned aerial vehicle i has exceeded the maximum waiting time for punishment.
(3) Constructing a multi-unmanned aerial vehicle collaborative track planning objective function:
Figure BDA0004088771170000084
D (i,j) representing a decision variable, d, between the drone i and the task point j (i,j) Representing the voyage cost of unmanned plane i reaching task point j, t (i,j) Representing the time cost of the unmanned aerial vehicle i reaching the task point j, c l Is a penalty function corresponding to the constraint, μ, λ, and τ are scaling factors of the voyage cost, the time cost, and the penalty function, respectively, and the three are kept in the same order.
Step 3: according to the positions of each unmanned aerial vehicle and task points in a simulation space, respectively constructing vertical tangential planes between the unmanned aerial vehicle and the task points, expanding a three-dimensional space feasible region according to a feasible region parameter, setting auxiliary track points, and only calculating candidate key track point evaluation indexes in the feasible region as a key track point set, wherein the method is shown in figure 2:
in fig. 2, the black dots indicate the positions U (x i ,y i ,z i ) The black triangle indicates the task point j position T (x j ,y j ,z j ). When the unmanned plane U and the task point T execute relation determination, a vertical tangent plane between the unmanned plane U and the task point T is established and expanded into a feasible region space. Determining each feasible region parameter Pa= [ delta x, delta y, delta z according to prior knowledge]The method comprises the steps of carrying out a first treatment on the surface of the The positions of auxiliary track points in each execution relation are set in a feasible region space, and the pentagram represents key track points k= [ k ] 1 ,k 2 ,k 3 ,k 4 ,k 5 ]Gray dotted lines represent the execution relationship between the unmanned aerial vehicle and the task point, hemispherical represents the radar scan area, v= [ x ] i +Δx,y i +Δy,z i +Δz]For the feasible region space of the unmanned aerial vehicle, each candidate key track point will be generated in V. Meanwhile, according to the maneuvering performance of the prior fixed wing unmanned aerial vehicle, the critical track points must meet the angle range of 120 degrees-180 degrees, wherein the angle range is provided withAnd the too large folding angle of the track point connection is effectively prevented. The space setting of the feasible region and the generation angle of the key track points are integrated, so that the algorithm can be ensured to keep effective track point information in the searching process, the searching blindness is reduced, and the specific implementation is as shown in the algorithm 1:
Figure BDA0004088771170000091
Figure BDA0004088771170000101
step 4: aiming at the fact that the variation strategy in the key track point algorithm is single, an adaptive rule of multi-unmanned aerial vehicle track planning is built, and each individual can select the variation strategy according to the self-adaptive degree value, as shown in fig. 3; then, the eligible task point sequence is deleted or supplemented using a reasonable code correction algorithm, as shown in FIG. 4.
Firstly, randomly dividing the whole population into a plurality of populations with the same number of individuals, calculating the fitness value of each parent population individual and carrying out preliminary classification; then, constructing a mutation strategy archive to store three mutation strategies with different characteristics: mutation strategy I: "DE/rand/1" is beneficial to improving individual exploratory properties; mutation strategy II: "DE/target to best/2" is useful for balancing exploratory and developability of populations; mutation strategy III: "DE/best/1" is beneficial to enhancing individual developability. Meanwhile, it should be noted that by calculating the fitness value of the individual, the better individual can only select the mutation strategy III, the moderate individual can select the mutation strategy II and the mutation strategy IIII, the worse individual can select the mutation strategies I, II and the mutation strategy III at will, and the method can improve the proportion of the better individual and the moderate individual in the exploratory individual.
Figure BDA0004088771170000102
Figure BDA0004088771170000111
The candidate key track points of the collaborative track planning of the multiple unmanned aerial vehicles are all integer discrete sequences, if the sequence values are subjected to variation result rounding, a large number of invalid values can appear in the solution after variation, and if the invalid values continue to participate in evolutionary computation, algorithm stagnation can be caused; of course, through a large number of iterations, a very small portion of individuals may still be present within a reasonable range, but may result in wasted computing resources. Therefore, the invalid discrete values must be corrected, the coding correction rule can directly map the invalid sequence to obtain a new discrete sequence, and in order to meet the requirement that the alternative key track points meet the planning requirement, the coding correction rule must meet the following conditions:
rule 1: mapping uniqueness; that is, during the correction process, the mapped discrete sequence must be unique; as shown in fig. 4, the invalid sequence individuals are 0, 9 and 5, when the differential individual 1 is selected as the newly added individual gene mapping, the position of the sequence where the original differential individual 1 is located needs to be deleted, and other differential values are mapped in sequence, so that the method can ensure that the mapping has uniqueness;
rule 2: mapping validity; that is, the mapped invalid discrete sequence does not participate in the subsequent mutation policy calculation, and although the mapped sequence is covered in rule 1, mutation policy information is still stored in the individual, so that the invalid mapping matrix needs to be deleted; meanwhile, the mapped individual sequence participates in the subsequent mutation strategy calculation, and the algorithm executable can be improved by the method.
Step 5: smoothing the track by adopting a cubic B spline to smooth the track L ij ＝(U i ,Auxi i ,k i1 ,k i2 ,...,k ik ,T j ) All points in the model (a) are used as control points of a B spline curve, and the smooth flyable track can be directly obtained by carrying out curve fitting in a three-dimensional space through iteration;
when n=3, the basis function of the B-spline can be expressed as:
Figure BDA0004088771170000112
thus, the parametric equation of the three-dimensional cubic B-spline curve can be obtained:
Figure BDA0004088771170000121
cubic B spline matrix form:
Figure BDA0004088771170000122
the matrix equation can approximate and obtain the fitted smooth flight path of the unmanned aerial vehicle by taking discrete unmanned aerial vehicle issuing points, auxiliary flight path points, key flight path points and task points as control points of spline curves.
The method aims at overcoming the influence of a complex three-dimensional simulation space structure on the solution of key track points, effectively simplifying the three-dimensional simulation space, establishing a single-machine constraint condition and a multi-machine cooperative constraint condition of the unmanned aerial vehicle, constructing a multi-unmanned aerial vehicle cooperative track planning objective function, sharing fitness value information according to a self-adaptive rule and a coding correction rule, improving algorithm searching efficiency, and further obtaining better track combination when the system is operated.
The foregoing description is only a few examples of the present application and is not intended to limit the present application in any way, and although the present application is disclosed in the preferred examples, it is not intended to limit the present application, and any person skilled in the art may make some changes or modifications to the disclosed technology without departing from the scope of the technical solution of the present application, and the technical solution is equivalent to the equivalent embodiments.

Claims (10)

1. A multi-unmanned aerial vehicle collaborative track planning method based on a self-adaptive differential evolution algorithm is characterized by comprising the following steps:
constructing a three-dimensional simulation space comprising an unmanned plane, task points and barriers;
constructing single-machine constraint of the unmanned aerial vehicle and cooperative constraint conditions of the unmanned aerial vehicle, and establishing a multi-machine cooperative track planning objective function;
respectively constructing vertical tangent planes between each unmanned aerial vehicle and each task point according to the positions of the unmanned aerial vehicle and the task points in the three-dimensional simulation space, expanding a three-dimensional space feasible region according to the feasible region parameters, setting auxiliary track points, and generating a key track point set according to the adaptability values of candidate key track points in the feasible region;
constructing a self-adaptive rule of multi-unmanned aerial vehicle track planning, wherein each individual automatically selects a variation strategy meeting the conditions according to the self-adaptive degree value, and deleting or supplementing a task point sequence meeting the conditions by adopting a coding correction rule;
and taking the departure points, auxiliary track points, key track points and task points of each unmanned aerial vehicle as control points of spline curves, and approximating to obtain a fitted smooth track of the unmanned aerial vehicle.
2. The method for planning a collaborative flight path of a plurality of unmanned aerial vehicles based on an adaptive differential evolution algorithm according to claim 1, wherein the constructing a three-dimensional simulation space comprises: constructing a radar scanning area and fusing obstacles;
preferably, the constructed radar scan area is represented by the following formula:
Figure FDA0004088771160000011
Figure FDA0004088771160000012
wherein, (x) i ,y i ,z i ) Representing the position of the auxiliary track point in space, (x) r ,y r ,z r ) Representing the position of the radar in space, d r Representing the distance between the auxiliary track point and the radar, R r Represents the scanning radius of the radar, k represents the threat level of the radar, F Radar Representing threat costs of radar;
preferably, the method for acquiring the obstacle comprises the following steps: the USGS24kdem is adopted to read the real USGS 1:24 000 digital elevation map file, and the file is used as a Terrain obstacle Terrain.
3. The adaptive differential evolutionary algorithm-based multi-unmanned aerial vehicle collaborative track planning method according to claim 1, wherein the unmanned aerial vehicle single-machine constraints comprise: unmanned aerial vehicle and mission point decision variable constraint, unmanned aerial vehicle maximum voyage constraint, unmanned aerial vehicle minimum/maximum speed constraint, maximum voyage time constraint and unmanned aerial vehicle maximum loading constraint.
4. The adaptive differential evolution algorithm-based multi-unmanned aerial vehicle collaborative track planning method according to claim 1, wherein the unmanned aerial vehicle collaborative constraint comprises: and (3) time sequence constraint among task points and waiting time constraint of each unmanned aerial vehicle, and determining that each constraint violates penalty.
5. A multi-unmanned aerial vehicle collaborative flight path planning method according to claim 3, wherein each unmanned aerial vehicle must execute a task point, each task point must be executed by one unmanned aerial vehicle, the unmanned aerial vehicle and task point decision variable constraints are constructed according to the following formula:
Figure FDA0004088771160000021
wherein N represents the number of unmanned aerial vehicles, M represents the number of task points, and D (i,j) Representing the corresponding relation of the execution task points j of the unmanned aerial vehicle i;
preferably, the maximum range constraint of the unmanned aerial vehicle refers to the sum of range costs of the execution task points of each unmanned aerial vehicle, and the formula is as follows:
Figure FDA0004088771160000022
Figure FDA0004088771160000023
wherein d (i,j) Representing voyage cost, D i Representing a defined maximum range distance, C, for each unmanned aerial vehicle i 1violation Indicating that the unmanned aerial vehicle i punishs when exceeding the limited maximum range distance;
preferably, the unmanned aerial vehicle minimum/maximum speed constraint is as follows:
V (i) ＝[V (i)min ,V (i)max ]
Figure FDA0004088771160000024
wherein V is (i)min Representing the minimum speed of unmanned plane i, V (i)max Represents the maximum speed of unmanned plane i, C 2violation The unmanned plane i is punished by exceeding the defined minimum/maximum speed;
preferably, the maximum voyage time constraint T max The following formula is shown:
Figure FDA0004088771160000031
Figure FDA0004088771160000032
wherein t is (i,j) Representing the time cost of the unmanned plane i reaching the task point j, D (i,j) Representing corresponding relation of execution task points j of unmanned plane i, T max Representing the maximum travel time constraint of unmanned plane i, v i Indicating the set sailing speed of the unmanned plane i, T i Representing the actual voyage time of the unmanned plane i, C 3violation The unmanned aerial vehicle i is punished when exceeding the limited maximum navigation time;
preferably, the unmanned aerial vehicle is the mostLarge-load constraint U missile The following formula is shown:
U missile (i)≤U max (i)
Figure FDA0004088771160000033
wherein U is max Representing the maximum loading capacity of the unmanned plane i, C 4violation Indicating that unmanned plane i penalizes beyond its maximum loading.
6. The multi-unmanned aerial vehicle collaborative flight path planning method based on the adaptive differential evolution algorithm according to claim 4, wherein the time sequence constraint T between task points is characterized in that sort The following formula is shown:
Figure FDA0004088771160000034
Figure FDA0004088771160000035
wherein T is sort (j) Representing the time sequence of the task point j, wherein tau is a positive integer, the task point j must be executed later than the task point j+tau, C 5violation Indicating that the task point is not beaten and punished according to the set time sequence;
preferably, the unmanned aerial vehicle latency constraint T wait The following formula is shown:
T wait (i)≤T max (i)
Figure FDA0004088771160000036
wherein T is max (i) Representing maximum waiting time of unmanned plane i, C 6violation Indicating that the unmanned aerial vehicle i has exceeded the maximum waiting time for punishment.
7. The multi-unmanned aerial vehicle collaborative track planning method based on the adaptive differential evolution algorithm according to claim 1, wherein the objective function minf (x) for establishing the multi-unmanned aerial vehicle collaborative track planning is as follows:
Figure FDA0004088771160000041
wherein D is (i,j) Representing a decision variable, d, between the drone i and the task point j (i,j) Representing the voyage cost of unmanned plane i reaching task point j, t (i,j) Representing the time cost of the unmanned aerial vehicle i reaching the task point j, c l Is a penalty function corresponding to the constraint, μ, λ, and τ are scaling factors of the voyage cost, the time cost, and the penalty function, respectively, and the three are kept in the same order.
8. The multi-unmanned aerial vehicle collaborative flight path planning method based on the self-adaptive differential evolution algorithm according to claim 1, wherein the auxiliary flight path points are arranged according to the maneuvering performance of the fixed wing unmanned aerial vehicle, and the flying included angle between the auxiliary flight path points is 120 degrees less than or equal to alpha less than or equal to 180 degrees.
9. The method for multi-unmanned aerial vehicle collaborative flight path planning based on the adaptive differential evolution algorithm according to claim 1, wherein the constructing the adaptive algorithm for multi-unmanned aerial vehicle flight path planning, each individual self-selects a variation strategy according to the self-adaptive degree value, comprises:
randomly dividing the whole population into a plurality of populations with the same number of individuals, calculating the fitness value of each parent population individual and carrying out preliminary classification;
constructing a mutation strategy archive to store three mutation strategies with different characteristics: the mutation strategy I is DE/rand/1; the mutation strategy II is DE/target to best/2; the mutation strategy III is DE/target to best/2;
and calculating the fitness value of each individual, and adaptively selecting a mutation strategy.
10. The adaptive differential evolution algorithm-based multi-unmanned aerial vehicle collaborative track planning method according to claim 1, wherein the code correction algorithm satisfies the following conditions:
in the correction process, the mapped discrete sequence must be unique;
the mapped invalid discrete sequence no longer participates in the subsequent mutation policy calculation;
preferably, the spline curve is a cubic B-spline curve.
CN202310140529.2A 2023-02-13 2023-02-13 Multi-unmanned aerial vehicle collaborative track planning method based on self-adaptive differential evolution algorithm Pending CN116301031A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310140529.2A CN116301031A (en) 2023-02-13 2023-02-13 Multi-unmanned aerial vehicle collaborative track planning method based on self-adaptive differential evolution algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310140529.2A CN116301031A (en) 2023-02-13 2023-02-13 Multi-unmanned aerial vehicle collaborative track planning method based on self-adaptive differential evolution algorithm

Publications (1)

Publication Number Publication Date
CN116301031A true CN116301031A (en) 2023-06-23

Family

ID=86823228

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310140529.2A Pending CN116301031A (en) 2023-02-13 2023-02-13 Multi-unmanned aerial vehicle collaborative track planning method based on self-adaptive differential evolution algorithm

Country Status (1)

Country Link
CN (1) CN116301031A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117008641A (en) * 2023-10-07 2023-11-07 中国人民解放军战略支援部队航天工程大学 Distribution method and device for cooperative low-altitude burst prevention of multiple heterogeneous unmanned aerial vehicles

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110134007A (en) * 2019-05-22 2019-08-16 南昌航空大学 Multiple no-manned plane cooperates with target assignment method

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110134007A (en) * 2019-05-22 2019-08-16 南昌航空大学 Multiple no-manned plane cooperates with target assignment method

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
GANG HUANG 等: "Multi-UAV Cooperative Trajectory Planning Based on FDS-ADEA in Complex Environments", 《DRONES》, vol. 7, no. 1, 12 January 2023 (2023-01-12), pages 1 - 15 *
黄刚 等: "基于AC-DSDE进化算法多UAVs协同目标分配", 《自动化学报》, vol. 47, no. 1, 31 January 2021 (2021-01-31), pages 174 - 179 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117008641A (en) * 2023-10-07 2023-11-07 中国人民解放军战略支援部队航天工程大学 Distribution method and device for cooperative low-altitude burst prevention of multiple heterogeneous unmanned aerial vehicles
CN117008641B (en) * 2023-10-07 2024-01-16 中国人民解放军战略支援部队航天工程大学 Distribution method and device for cooperative low-altitude burst prevention of multiple heterogeneous unmanned aerial vehicles

Similar Documents

Publication Publication Date Title
CN111780777B (en) Unmanned vehicle route planning method based on improved A-star algorithm and deep reinforcement learning
Wu et al. A hybrid algorithm of particle swarm optimization, metropolis criterion and RTS smoother for path planning of UAVs
Liu et al. Multi-UAV path planning based on fusion of sparrow search algorithm and improved bioinspired neural network
US20220317695A1 (en) Multi-agv motion planning method, device and system
Li et al. Trajectory planning for UAV based on improved ACO algorithm
CN103267528A (en) Multi-unmanned aerial vehicle cooperative area search method under non-flight zone limitation
CN109764886A (en) A kind of paths planning method
Abousleiman et al. Electric vehicle modelling and energy‐efficient routing using particle swarm optimisation
Yan et al. Task allocation and route planning of multiple UAVs in a marine environment based on an improved particle swarm optimization algorithm
CN112230678A (en) Three-dimensional unmanned aerial vehicle path planning method and planning system based on particle swarm optimization
CN114281104B (en) Multi-unmanned aerial vehicle cooperative regulation and control method based on improved ant colony algorithm
CN112733251B (en) Collaborative flight path planning method for multiple unmanned aerial vehicles
CN114879745B (en) Multi-unmanned aerial vehicle collaborative task planning method based on preplanning and tabu search algorithm
CN116301031A (en) Multi-unmanned aerial vehicle collaborative track planning method based on self-adaptive differential evolution algorithm
Li et al. A Kriging-based bi-objective constrained optimization method for fuel economy of hydrogen fuel cell vehicle
CN112000126B (en) Whale algorithm-based multi-unmanned-aerial-vehicle collaborative searching multi-dynamic-target method
CN112947541A (en) Unmanned aerial vehicle intention track prediction method based on deep reinforcement learning
CN114839968A (en) Unmanned surface vehicle path planning method
Ke et al. Cooperative path planning for air–sea heterogeneous unmanned vehicles using search-and-tracking mission
Huang et al. An efficient self-evolution method of autonomous driving for any given algorithm
Yang et al. Fast marine route planning for UAV using improved sparse A* algorithm
Liu et al. Multi-uavs cooperative coverage reconnaissance with neural network and genetic algorithm
CN116400718A (en) Unmanned aerial vehicle short-distance air combat maneuver autonomous decision-making method, system, equipment and terminal
CN114326821B (en) Unmanned aerial vehicle autonomous obstacle avoidance system and method based on deep reinforcement learning
CN115933633A (en) Boat-engine cooperative task allocation method based on quantum silk optimization mechanism

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