CN109631900A - A kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning - Google Patents
A kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning Download PDFInfo
- Publication number
- CN109631900A CN109631900A CN201811583287.XA CN201811583287A CN109631900A CN 109631900 A CN109631900 A CN 109631900A CN 201811583287 A CN201811583287 A CN 201811583287A CN 109631900 A CN109631900 A CN 109631900A
- Authority
- CN
- China
- Prior art keywords
- track
- particle
- node
- algorithm
- cost
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/20—Instruments for performing navigational calculations
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention discloses a kind of no-manned plane three-dimensional track multi-objective particle swarm Global Plannings, comprising the following steps: (1) environmental model is established according to flight environment of vehicle, when handling the flight environment of vehicle of unmanned plane, using numerical map technology；(2) trajectory planning model is established according to the environmental model that step (1) is established；(3) Particle Swarm Optimization is used for step (2) the trajectory planning model, provides a kind of multiple target backbone population overall situation Path Planning of improved few control parameter.The present invention has the advantage that establishing no-manned plane three-dimensional overall situation multi-target traces plan model, it gives track length cost, threaten three kinds of target functions of cost and concealment cost, and demand is constrained accordingly, a kind of multi-objective particle swarm Global Planning of few control parameter is proposed, the practicability of planning path is enhanced.
Description
Technical field
The invention belongs to air vehicle technique fields more particularly to a kind of no-manned plane three-dimensional track multi-objective particle swarm overall situation to advise
The method of drawing.
Background technique
Trajectory planning refers under defined constraint condition, cooks up one and meets aircraft itself from starting point to target point
Performance, least flight path of paying a price.In recent years, as unmanned plane is in the extensive concern of military field, track rule
The performance for the system of drawing is perfect constantly.Specifically, unmanned aerial vehicle flight path planning is primarily referred to as not having the case where manual intervention
Under, unmanned plane is cooked up most according to self performance, external environment disturbing factor and other dynamics or kinematical constraint condition
Excellent flight track, it is ensured that it safely and efficiently completes distributed task.Trajectory planning is important in unmanned plane autonomous system
Component part, good path planning method can not only reduce the fuel oil loss of unmanned plane, and improve its survival ability.
Under normal circumstances, unmanned plane is applied to various landform complicated and changeable, and not only restraining factors are more, but also objective function
Between intercouple, the path quality that trajectory planning systems organization goes out is required high.In order to make unmanned plane in flight course more
Add flexible strain, most trajectory planning systems will do it hierarchical planning, be divided into global trajectory planning and real-time trajectory planning.It is global
Trajectory planning refers to that before unmanned plane takes off, the various known informations of comprehensive improvement utilize the powerful calculating energy of earth station
Power, cooks up the whole path of unmanned plane during flying, and is entered into the airborne computer of unmanned plane.The global boat of unmanned plane
Mark is planned to real-time trajectory planning and provides reference data, so that unmanned plane is when carrying out the adjustment of real-time track, search area subtracts
It is few, rapid Optimum flight track.
The extensive use of unmanned plane, so that its trajectory planning problem becomes research hotspot.Relatively traditional two-dimentional track rule
The problem of drawing, since search space becomes larger, no-manned plane three-dimensional trajectory planning problem becomes relatively difficult.
Summary of the invention
Goal of the invention: in view of the above problems, the present invention proposes a kind of no-manned plane three-dimensional track multi-objective particle swarm overall situation rule
The method of drawing.Present invention research no-manned plane three-dimensional overall situation trajectory planning model and its multi-objective particle swarm derivation algorithm.Firstly, considering
Track length cost threatens cost, concealment cost index and avoiding obstacles constraint, establishes three-dimensional global trajectory planning mould
Type；Then, Particle Swarm Optimization is used to solve above-mentioned model, it is global proposes a kind of improved multiple target backbone population
Path planning algorithm.
Technical solution: to achieve the purpose of the present invention, the technical scheme adopted by the invention is that: a kind of no-manned plane three-dimensional boat
Mark multi-objective particle swarm Global Planning, comprising the following steps:
(1) it establishes environmental model: environmental model is established according to flight environment of vehicle, the flight environment of vehicle of unmanned plane is handled
When, using numerical map technology, the numerical map technology continuously highly carries out landform discrete, is stored in digital form
Among grid, the distance between grid size is according to the division for needing to carry out precision of practical problem；Pass through numerical map technology
Get the center of the physical relief information in unmanned plane during flying regional scope, the threat information by the processing of equivalent landform
And the threat range of deterrent.
(2) it establishes trajectory planning model: establishing trajectory planning model on the basis of step (1) established environmental model, benefit
Fitness function of the unmanned plane in flight course is simulated with cost function is established, for evaluating the adaptive value of particle, to nobody
Machine track performance is evaluated, and considers that minimum flying height and track pass through massif limiting factor, wherein minimum flying height
It is set according to the maximum height of ground object；
(3) the global trajectory planning based on multiple target backbone particle group optimizing: it is based on multiple target backbone Particle Swarm Optimization
Method proposes that particle coding, constraint processing strategie and algorithm execute step, and track plan model in Optimization Steps (2) is sought planning
The global navigation route of unmanned plane；On the basis of multiple target backbone particle swarm optimization algorithm, a kind of multinode particle coding is provided
The problem of continuous domain, is mapped in discrete domain by strategy, and a particle represents a unmanned plane during flying path or track, Yi Tiaolu
Diameter is determined jointly by m node；In trajectory planning, it is necessary to consider constraint present in track, and provide in the algorithm corresponding
Constraint processing strategie select a best trade-off solution to be supplied to decision from the one group of Pareto optimal solution finally generated
Person uses.
(4) analogue simulation is carried out according to step (1) to step (3), by multiple target backbone particle swarm optimization algorithm and NSGA2
The operation result of benchmark algorithm compares, with the global multiple target boat of three-dimensional under MATLAB simulated environment, carrying out unmanned plane
Mark planning.
Preferably, the cost function includes: track length cost, threatens cost and concealment cost.
Preferably, constraint present in the track includes: to prevent track from passing over massif or colliding with barrier.
Preferably, the multiple target backbone particle group optimizing includes: multinode particle coding strategy, constrains processing strategie,
The path production method of few control parameter.
The constraint condition that unmanned plane faces in flight course is numerous, such as fuselage performance, weather conditions, landform mountain peak,
When carrying out trajectory planning, need to consider the various limiting factors in flight course.The present invention is directed to minimum flying height, Yi Jihang
Mark passes through massif and is limited.
Wherein, it in step (2), using fitness function of the cost function simulation unmanned plane in flight course is established, uses
The adaptive value of particle is evaluated, unmanned aerial vehicle flight path performance evaluated, the method is as follows:
(2-1) track length cost is measured to the total length of unmanned aerial vehicle flight path, is referred to meeting other all performances
In the case of target, reduce the distance of flight.The track cost model that the present invention uses is as follows:
Wherein, Pathi=(Pi,1,Pi,1,…,Pi,m) indicate i-th track, Pi,1And Pi,mBe unmanned plane starting point S and
Terminal G；I-th track Path of f1 expression unmanned planeiLength cost；Pi,j=(xi,j,yi,j,zi,j) indicate i-th track
J-th of node, x, y indicate that the floor projection coordinate of track node, z indicate that the corresponding elevation of track node, m are i-th boats
The number of nodes of mark.
(2-2) threatens cost to refer to probability of the unmanned plane by the capture strike of enemy air defences weapon.Under normal conditions, into
When row trajectory planning, reduce the threat cost of unmanned plane.The threat Cost Model that the present invention establishes is as follows:
Wherein ε is a positive number, and when programming takes 0.0001；It is a positive number, when programming takes 0.001；dang
(j-1, j) indicates -1 node of jth to the degree of danger in jth node composition section；A is the folder on unmanned plane during flying direction and ground
Angle；Dis (j-1, j) be between -1 knee level projection coordinate point of jth of track and jth knee level projection coordinate point away from
From；Dis1 (j-1, j) is the distance between -1 node of jth and jth node；Dis1 (j, j+1) is jth node and+1 node of jth
The distance between；DminAnd DmaxIt is the boundary of the valid interval of radar detection object respectively.
(2-3) concealment cost is the height cost of unmanned plane during flying.Unmanned plane during flying it is lower, then can use ground
Clutter come it is hidden oneself, the probability found by other side is lower, but the corresponding probability to crash that hits is increased by, so for most flying at low altitude
Capable height is limited.The concealment Cost Model that the present invention establishes is as follows:
Wherein, yinb (j-1, j) indicates the hidden index of -1 node of jth and the determined segment of jth node, hmaxFor number
Highest elevation value in word hypsographic map；zi,jFor the height value of j-th of node of i-th track；di,jIt is the of i-th article of track
J knee level coordinate corresponds to the height value in digital elevation map；Safth is safe altitude.
Wherein, in step (3), multiple target backbone particle swarm optimization algorithm method is as follows:
Multiple target backbone particle swarm optimization algorithm (BB-MOPSO) is individual optimum point and globe optimum based on particle
A Gaussian distribution model is generated, and with certain probability, the position of particle subsequent time is directly generated by the model, from basic
On eliminate particle position and update dependence to control parameter, it is up to the present the most succinct multi-objective particle swarm optimization
Algorithm.BB-MOPSO algorithm is on the basis of simple backbone particle swarm optimization algorithm (BB-PSO), and progress is various perfect,
Specifically include that first is that, external storage collection is introduced in BB-MOPSO algorithm, is stored in the non-branch generated during algorithm iteration
With collection, guarantee that external storage concentrates the diversity of non-domination solution using the crowding technology in evolution algorithm, so that particle tracks
The alternate boot person of optimization is more reasonable；Second is that introducing mutation operator in BB-MOPSO algorithm, population is disturbed
It is dynamic, to maintain the diversity of population；Third is that changing previous particle more new formula, i.e., formula (4), BB-MOPSO algorithm are given
Go out a kind of new more new formula, i.e. formula (5):
Wherein, Pbi(t) and Gbi(t) be respectively i-th of particle individual extreme point and global extreme point, t indicate algorithm change
Generation number, r3For the random number between [0,1], XiIt (t+1) is the newborn position of i-th of particle, U (0,1) is an obedience normal state
The random number of distribution, N (c, k) are the gauss of distribution function that mean value is c, variance is k.
Different from the individual optimum point of traditional BB-PSO algorithm study particle, in BB-MOPSO algorithm, if random number
Greater than 0.5, then current particle is updated to its corresponding globe optimum.Can more widen in this way particle search field and
Particle search quality.
The process object of multiple target backbone particle swarm optimization algorithm (BB-MOPSO) is abandoned continuous multiple target
Optimization problem.Since in unmanned aerial vehicle flight path planning problem, objective function may be discontinuous, but also by itself mobility
Constraint, therefore, BB-MOPSO algorithm can not be directly used in trajectory planning.For this purpose, the present invention to BB-MOPSO algorithm into
Row improves, and can be applied to no-manned plane three-dimensional overall situation trajectory planning.Main improvement is as follows: firstly, providing a kind of multinode
The problem of continuous domain, is mapped in discrete domain by particle coding strategy；Secondly, in trajectory planning, it is necessary to consider to deposit in track
Constraint, for example prevent track from passing over massif or colliding with barrier, and provide corresponding constraint processing in the algorithm
Strategy.In addition, needing to select a best trade-off for one group of Pareto optimal solution that multi-objective optimization algorithm finally generates
Solution is supplied to policymaker's use.For this purpose, providing a kind of fuzzy decision based on non-domination solution satisfaction in final result output
Scheme is selected than more objective reference track planning path.
Wherein, in step (3), by multiple target backbone particle swarm optimization algorithm, particle coding, constraint processing plan are proposed
Slightly and algorithm executes step, the method is as follows:
(3-1) space planning and particle coding, include the following steps:
(3-1-1) when using particle swarm optimization algorithm processing trajectory planning, need update amplitude to particle position into
Row limitation, i.e., limit the variation range of each node coordinate, in the range of:
{(x,y,z)|xmin<x<xmax,ymin<y<ymax,zmin<z<zmax}
And then constitute the decision space of trajectory planning, wherein xminAnd xmax、yminAnd ymax、zminAnd zmaxIt is every respectively
The boundary of the x, y, z axis of a node.The x-axis value range that each node is arranged in the present invention is [1,101], and y-axis value range is
[1,101], z-axis value range [1,10000], and 100 parts of equal parts are carried out for x, y, constitute the net of the rule of 101*101
Lattice, the corresponding height value of each grid, is the search space of unmanned plane.
The flight path of (3-1-2) unmanned plane is not a continuous smooth curve, is by a series of track section groups
At.There are key point segmentation between track section, the key point of these tracks is used to instruct the flight of aircraft.The present invention proposes
Be unmanned plane Global Planning, therefore, what each particle represented is a complete track, and every track is by m section
Point composition.{ P is indicated with a group node vector to each particle1,P2,…,Pm-1,Pm, wherein P1And PmRespectively fly
Starting point S and target point G, P2,…,Pm-1For the intermediate node of flight path；In the corresponding value range of each node, with
Machine initializes the position of the node, arranged in sequence node produced, and all nodes form a fullpath, that is, generates one
A particle position.
(3-1-3) repeats node initializing and arranged in sequence in step (3-1-2), that is, can produce the grain of defined amount
Son, population needed for forming.
(3-2) bounding algorithm: the optimization of Three-dimensional Track must be carried out in conjunction with constraint when using BB-MOPSO algorithm.This hair
Bright introducing constraint function choice () performs corresponding processing the particle for being unsatisfactory for constraint, includes the following steps:
(3-2-1) in more new particle, according to each component of the node sequencing successively more new node of particle.For
Any one particle judges whether the node feasible using following methods when updating one of node: present node with
N point, number m are taken on segment determined by previous node1,m2,…,mn, compare the practical elevation of each point in set
Value and height value corresponding in numerical map, if there is a point height value lower than digital elevation value, i.e. the path node position
In in barrier, then need to update the position of present node.Continue to judge whether updated node meets according to the method described above
Condition, until just can be carried out the update of next node after meeting condition.By carrying out above-mentioned behaviour to the node for violating constraint
Make, it is possible to reduce the amount of doing over again of algorithm.
(3-2-2) limits minimum track height accordingly, and the present invention is provided with minimum flight altitude safth, leads to
Minimum track height is crossed, the concealment cost in path in step (2-3) is calculated.
(3-3) formulates corresponding track decision scheme: in BB-MOPSO algorithm, needing to formulate corresponding decision strategy
To policymaker one final reference solution.Present invention employs the fuzzy compromise solution of one kind, in multiple non-domination solutions obtained by algorithm
Select a best solution.This method passes through to non-domination solution XkSatisfaction μkIt is calculated, the non-domination solution of Maximum Satisfaction
For final selected flight track, wherein μkCalculation formula it is as follows:
In formula, | Ar | it is the number of non-domination solution；M is the number of objective function；μi kIt is non-domination solution in i-th of target letter
Several satisfactions, membership function are as follows:
The maximum and minimum value of respectively i-th objective function, Fi(Xk) it is non-domination solution XkIn i-th of mesh
The value of scalar functions.
(3-4) algorithm executes step: being based on step (3-1) to (3-3), the improvement multiple target towards unmanned aerial vehicle flight path planning
Steps are as follows for the execution of backbone particle swarm optimization algorithm:
Algorithm relevant parameter, the scale Ns including population, the number of nodes Nt of population, external storage is arranged in (3-4-1)
The capacity integrated is Na, the number of iterations Ts.
(3-4-2) encodes the particle in population, the position of each particle in random initializtion population, setting
The individual extreme point Pbest of each particle is the initial position of the particle.
(3-4-3) evaluates the target function value of each particle, i.e. step (2-1) the track length cost, step (2-2)
The threat cost and step (2-3) the concealment cost return to the non-domination solution in population, non-domination solution are saved in
In external storage set, and assignment is carried out to the global extreme point Gbest of particle.
The Gaussian Profile of (3-4-4) based on Pbest and Gbest carries out location updating to particle, judges that the particle updated is raw
Whether the track section track of production passes through massif, and new position is regenerated if passing through, until the corresponding boat of produced particle
Mark no longer passes through massif.
(3-4-5) evaluates the target function value of each particle, i.e. step (2-1) the track length cost, step (2-2)
The threat cost and step (2-3) the concealment cost carry out more individual extreme point Pbest by dominance relation
Newly.
(3-4-6) carries out the previous generation non-domination solution of the new population non-domination solution generated and external storage concentration
Pareto compares, and is updated to external storage collection.Judge whether external storage collection is greater than the capacity Na of external storage collection, if
It is to be cut to population.
The global extreme point Gbest of (3-4-7) more new particle.
(3-4-8) judges whether the termination condition for reaching algorithm, if it is exports Pareto optimal solution set.If no
Above (3-4-4) is then repeated to (3-4-7) step.
The utility model has the advantages that the invention has the following advantages over the prior art: according to numerical map technology to initial land form
And the equivalent model threatened is merged, and is established no-manned plane three-dimensional overall situation multi-target traces plan model, is given track
Length cost threatens cost and concealment cost target function and constrains demand accordingly；Particle Swarm Optimization is used for
Above-mentioned model is solved, a kind of improved multiple target backbone population overall situation Path Planning is given.The algorithm is not only inherited
Simple, the easy to accomplish feature of the structure of particle swarm optimization algorithm, and no setting is required the ginseng such as inertia weight and Studying factors
Number is a kind of multi-objective particle swarm path planning method of few control parameter.By the success of multi-objective particle swarm Path Planning
For 2 simulated environment；By compared with the global Path Planning based on NSGA2, the results show calculation of the present invention
The superiority and feasibility of method.
Detailed description of the invention
Fig. 1 is algorithm numerical map of the invention；
Fig. 2 is transformation digital terrain of the invention；
Fig. 3 is 1 gained decision track plot of BB-MOPSO Environment Oriented of the invention；
Fig. 4 is 1 gained decision track plot of NSGA2 benchmark algorithm Environment Oriented of the invention；
Fig. 5 is 2 gained decision track plot of BB-MOPSO Environment Oriented of the invention；
Fig. 6 is 2 gained decision track plot of NSGA2 benchmark algorithm Environment Oriented of the invention.
Specific embodiment
Further description of the technical solution of the present invention with reference to the accompanying drawings and examples.
A kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning of the present invention, comprising the following steps:
(1) it establishes environmental model: environmental model is established according to flight environment of vehicle, the flight environment of vehicle of unmanned plane is handled
When, using numerical map technology, the numerical map technology continuously highly carries out landform discrete, is stored in digital form
Among grid, the distance between grid size is according to the division for needing to carry out precision of practical problem；Pass through numerical map technology
Get the center of the physical relief information in unmanned plane during flying regional scope, the threat information by the processing of equivalent landform
And the threat range of deterrent.
(2) it establishes trajectory planning model: establishing trajectory planning model on the basis of step (1) established environmental model, benefit
Fitness function of the unmanned plane in flight course is simulated with cost function is established, for evaluating the adaptive value of particle, to nobody
Machine track performance is evaluated, and considers that minimum flying height and track pass through massif limiting factor, wherein minimum flying height
It is set according to the maximum height of ground object.
(3) the global trajectory planning based on multiple target backbone particle group optimizing: it is based on multiple target backbone Particle Swarm Optimization
Method proposes that particle coding, constraint processing strategie and algorithm execute step, and track plan model in Optimization Steps (2) is sought planning
The global navigation route of unmanned plane；On the basis of multiple target backbone particle swarm optimization algorithm, a kind of multinode particle coding is provided
The problem of continuous domain, is mapped in discrete domain by strategy, and a particle represents a unmanned plane during flying path or track, Yi Tiaolu
Diameter is determined jointly by m node；In trajectory planning, it is necessary to consider constraint present in track, and provide in the algorithm corresponding
Constraint processing strategie select a best trade-off solution to be supplied to decision from the one group of Pareto optimal solution finally generated
Person uses.
(4) analogue simulation is carried out according to step (1) to step (3), by multiple target backbone particle swarm optimization algorithm and NSGA2
The operation result of benchmark algorithm compares, with the global multiple target boat of three-dimensional under MATLAB simulated environment, carrying out unmanned plane
Mark planning.
The cost function includes: track length cost, threatens cost and concealment cost.
Constraint includes: to prevent track from passing over massif or colliding with barrier present in the track.
The multiple target backbone particle group optimizing includes: multinode particle coding strategy, constrains processing strategie, few control ginseng
Several path production methods.
The constraint condition that unmanned plane faces in flight course is numerous, such as fuselage performance, weather conditions, landform mountain peak,
When carrying out trajectory planning, need to consider the various limiting factors in flight course.The present invention is directed to minimum flying height, Yi Jihang
Mark passes through massif and is limited.
Wherein, it in step (2), using fitness function of the cost function simulation unmanned plane in flight course is established, uses
The adaptive value of particle is evaluated, unmanned aerial vehicle flight path performance evaluated, the method is as follows:
(2-1) track length cost is measured to the total length of unmanned aerial vehicle flight path, is referred to meeting other all performances
In the case of target, reduce the distance of flight.The track cost model that the present invention uses is as follows:
Wherein, Pathi=(Pi,1,Pi,1,…,Pi,m) indicate i-th track, Pi,1And Pi,mBe unmanned plane starting point S and
Terminal G；I-th track Path of f1 expression unmanned planeiLength cost；Pi,j=(xi,j,yi,j,zi,j) indicate i-th track
J-th of node, x, y indicate that the floor projection coordinate of track node, z indicate that the corresponding elevation of track node, m are i-th boats
The number of nodes of mark.
(2-2) threatens cost to refer to probability of the unmanned plane by the capture strike of enemy air defences weapon.Under normal conditions, into
When row trajectory planning, reduce the threat cost of unmanned plane.The threat Cost Model that the present invention establishes is as follows:
Wherein, ε is a positive number, and when programming takes 0.0001；It is a positive number, when programming takes 0.001；
Dang (j-1, j) indicates -1 node of jth to the degree of danger in jth node composition section；A is unmanned plane during flying direction and ground
Angle；Dis (j-1, j) be between -1 knee level projection coordinate point of jth of track and jth knee level projection coordinate point away from
From；Dis1 (j-1, j) is the distance between -1 node of jth and jth node；Dis1 (j, j+1) is jth node and+1 node of jth
The distance between；DminAnd DmaxIt is the boundary of the valid interval of radar detection object respectively.
(2-3) concealment cost is the height cost of unmanned plane during flying.Unmanned plane during flying it is lower, then can use ground
Clutter come it is hidden oneself, the probability found by other side is lower, but the corresponding probability to crash that hits is increased by, so for most flying at low altitude
Capable height is limited.The concealment Cost Model that the present invention establishes is as follows:
Wherein, yinb (j-1, j) indicates the hidden index of -1 node of jth and the determined segment of jth node, hmaxFor number
Highest elevation value in word hypsographic map；zi,jFor the height value of j-th of node of i-th track；di,jIt is the of i-th article of track
J knee level coordinate corresponds to the height value in digital elevation map；Safth is safe altitude.
Wherein, in step (3), multiple target backbone particle swarm optimization algorithm (BB-MOPSO) be based on particle individual most
Advantage and globe optimum generate a Gaussian distribution model, and with certain probability, are directly generated one under particle by the model
The position at moment has been inherently eliminated dependence of the particle position update to control parameter, it is up to the present the most succinct
Multi-objective particle.BB-MOPSO algorithm is on the basis of simple backbone particle swarm optimization algorithm (BB-PSO)
On, carry out it is various perfect, specifically include that first is that, external storage collection is introduced in BB-MOPSO algorithm, is stored in algorithm
The non-dominant collection generated in iterative process guarantees that external storage concentrates non-domination solution using the crowding technology in evolution algorithm
Diversity, so that particle tracks, the alternate boot person of optimization is more reasonable；It is calculated second is that introducing variation in BB-MOPSO algorithm
Son disturbs population, to maintain the diversity of population；Third is that changing previous particle more new formula, i.e. formula
(4), BB-MOPSO algorithm gives a kind of new more new formula, i.e. formula (5):
Wherein, Pbi(t) and Gbi(t) be respectively i-th of particle individual extreme point and global extreme point, t indicate algorithm change
Generation number, r3For the random number between [0,1], XiIt (t+1) is the newborn position of i-th of particle, U (0,1) is an obedience normal state
The random number of distribution, N (c, k) are the gauss of distribution function that mean value is c, variance is k.
Different from the individual optimum point of traditional BB-PSO algorithm study particle, in BB-MOPSO algorithm, if random number
Greater than 0.5, then current particle is updated to its corresponding globe optimum.Can more widen in this way particle search field and
Particle search quality.
The process object of multiple target backbone particle swarm optimization algorithm (BB-MOPSO) is abandoned continuous multiple target
Optimization problem.Since in unmanned aerial vehicle flight path planning problem, objective function may be discontinuous, but also by itself mobility
Constraint, therefore, BB-MOPSO algorithm can not be directly used in trajectory planning.For this purpose, the present invention to BB-MOPSO algorithm into
Row improves, and can be applied to no-manned plane three-dimensional overall situation trajectory planning.Main improvement is as follows: firstly, providing a kind of multinode
The problem of continuous domain, is mapped in discrete domain by particle coding strategy；Secondly, in trajectory planning, it is necessary to consider to deposit in track
Constraint, for example prevent track from passing over massif or colliding with barrier, and provide corresponding constraint processing in the algorithm
Strategy.In addition, needing to select a best trade-off for one group of Pareto optimal solution that multi-objective optimization algorithm finally generates
Solution is supplied to policymaker's use.For this purpose, providing a kind of fuzzy decision based on non-domination solution satisfaction in final result output
Scheme is selected than more objective reference track planning path.
Wherein, in step (3), by multiple target backbone particle swarm optimization algorithm, particle coding, constraint processing plan are proposed
Slightly and algorithm executes step, the method is as follows:
(3-1) space planning and particle coding, include the following steps:
(3-1-1) when using particle swarm optimization algorithm processing trajectory planning, need update amplitude to particle position into
Row limitation, i.e., limit the variation range of each node coordinate, in the range of:
{(x,y,z)|xmin<x<xmax,ymin<y<ymax,zmin<z<zmax}
And then constitute the decision space of trajectory planning, wherein xminAnd xmax、yminAnd ymax、zminAnd zmaxIt is every respectively
The boundary of the x, y, z axis of a node.The x-axis value range that each node is arranged in the present invention is [1,101], and y-axis value range is
[1,101], z-axis value range [1,10000], and 100 parts of equal parts are carried out for x, y, constitute the net of the rule of 101*101
Lattice, the corresponding height value of each grid, is the search space of unmanned plane.
The flight path of (3-1-2) unmanned plane is not a continuous smooth curve, is by a series of track section groups
At.There are key point segmentation between track section, the key point of these tracks is used to instruct the flight of aircraft.The present invention proposes
Be unmanned plane Global Planning, therefore, what each particle represented is a complete track, and every track is by m section
Point composition.{ P is indicated with a group node vector to each particle1,P2,…,Pm-1,Pm, wherein P1And PmRespectively fly
Starting point S and target point G, P2,…,Pm-1For the intermediate node of flight path；In the corresponding value range of each node, with
Machine initializes the position of the node, arranged in sequence node produced, and all nodes form a fullpath, that is, generates one
A particle position.
(3-1-3) repeats node initializing and arranged in sequence in step (3-1-2), that is, can produce the grain of defined amount
Son, population needed for forming.
(3-2) bounding algorithm: the optimization of Three-dimensional Track must be carried out in conjunction with constraint when using BB-MOPSO algorithm.This hair
Bright introducing constraint function choice () performs corresponding processing the particle for being unsatisfactory for constraint, includes the following steps:
(3-2-1) in more new particle, according to each component of the node sequencing successively more new node of particle.For
Any one particle judges whether the node feasible using following methods when updating one of node: present node with
N point, number m are taken on segment determined by previous node1,m2,…,mn, compare the practical elevation of each point in set
Value and height value corresponding in numerical map, if there is a point height value lower than digital elevation value, i.e. the path node position
In in barrier, then need to update the position of present node.Continue to judge whether updated node meets according to the method described above
Condition, until just can be carried out the update of next node after meeting condition.By carrying out above-mentioned behaviour to the node for violating constraint
Make, it is possible to reduce the amount of doing over again of algorithm.
(3-2-2) limits minimum track height accordingly, and the present invention is provided with minimum flight altitude safth, leads to
Minimum track height is crossed, the concealment cost in path in step (2-3) is calculated.
(3-3) formulates corresponding track decision scheme: in BB-MOPSO algorithm, needing to formulate corresponding decision strategy
To policymaker one final reference solution.Present invention employs the fuzzy compromise solution of one kind, in multiple non-domination solutions obtained by algorithm
Select a best solution.This method passes through to non-domination solution XkSatisfaction μkIt is calculated, the non-domination solution of Maximum Satisfaction
For final selected flight track, wherein μkCalculation formula it is as follows:
In formula, | Ar | it is the number of non-domination solution；M is the number of objective function；μi kIt is non-domination solution in i-th of target letter
Several satisfactions, membership function are as follows:
The maximum and minimum value of respectively i-th objective function, Fi(Xk) it is non-domination solution XkIn i-th of mesh
The value of scalar functions.
(3-4) algorithm executes step: being based on step (3-1) to (3-3), the improvement multiple target towards unmanned aerial vehicle flight path planning
Steps are as follows for the execution of backbone particle swarm optimization algorithm:
Algorithm relevant parameter, the scale Ns including population, the number of nodes Nt of population, external storage is arranged in (3-4-1)
The capacity integrated is Na, the number of iterations Ts.
(3-4-2) encodes the particle in population, the position of each particle in random initializtion population, setting
The individual extreme point Pbest of each particle is the initial position of the particle.
(3-4-3) evaluates the target function value of each particle, i.e. step (2-1) the track length cost, step (2-2)
The threat cost and step (2-3) the concealment cost return to the non-domination solution in population, non-domination solution are saved in
In external storage set, and assignment is carried out to the global extreme point Gbest of particle.
The Gaussian Profile of (3-4-4) based on Pbest and Gbest carries out location updating to particle, judges that the particle updated is raw
Whether the track section track of production passes through massif, if passed through, regenerates new position, until the corresponding boat of produced particle
Mark no longer passes through massif.
(3-4-5) evaluates the target function value of each particle, i.e. step (2-1) the track length cost, step (2-2)
The threat cost and step (2-3) the concealment cost carry out more individual extreme point Pbest by dominance relation
Newly.
(3-4-6) carries out the previous generation non-domination solution of the new population non-domination solution generated and external storage concentration
Pareto compares, and is updated to external storage collection.Judge whether external storage collection is greater than the capacity Na of external storage collection, if
It is then to be cut to population.
The global extreme point Gbest of (3-4-7) more new particle.
(3-4-8) judges whether the termination condition for reaching algorithm, if it is, output Pareto optimal solution set.If not yet
Have, then repeats above (3-4-4) to (3-4-7) step.
Wherein, in step (4), it is separately operable two kinds of algorithms of BB-MOPSO and NSGA2 in environment 1 and environment 2 and carries out
Emulation, steps are as follows for analogue simulation:
(4-1) algorithm parameter prepares
BB-MOPSO algorithm and NSGA2 algorithm population parameter setting: population scale is 60, and the dimension of particle is 70, outside
The size of portion's storage collection is 10, and the maximum number of iterations of population is 100, and the starting point of unmanned plane and terminal are all in algorithm
[1,1,150], [101,101,150], for x, the limitation of y, z are respectively 1≤x≤101,1≤y≤101, z >=150.Nobody
Such as Fig. 1 and Fig. 2 respectively in machine trajectory planning space, size is all the topographic map of 101km*101km.Fig. 1 and Fig. 2 are in addition to ground
Looks it is not identical, the distribution of one of radar is also different, Fig. 1 three at radar distribution horizontal centre projection be respectively
[10,20], [40,60], [60,40], Fig. 2 three at radar distribution horizontal centre projection not Wei [20,10], [60,40],
[40,60]。
The selection of (4-2) multi-objective optimization algorithm evaluation index
The optimization of multi-objective problem is different from single-object problem, and multi-objective optimization question solution is often uncertain,
Therefore, from theoretical side, it is difficult to make and objectively evaluate to the optimizing performance of algorithm itself.For this purpose, being calculated by comparing Different Optimization
The operation result of method, can be with the superiority of quantitative analysis algorithm.The purpose of multi-objective optimization algorithm is to be quickly found out optimization problem
The forward position Pareto, so the distributivity and convergence by the Pareto optimal solution to multi-objective optimization algorithm are analyzed,
Carry out the performance quality of evaluation algorithms.The present invention has selected distributivity sp metric and superspace to estimate H metric, to the two property
It can be carried out qualitative more analysis.Method is as follows:
(4-2-1) distributivity sp indicates the space distribution situation of final non-domination solution obtained by algorithm.Its calculation formula is as follows:
In formula, | Q | indicate the number of non-domination solution；diIndicate that the i non-domination solution concentrates all solutions to non-domination solution
The shortest distance；Indicate the mean value of all shortest distances, N indicates the element number that storage is left concentratedly.For sp performance indicator
For, value is smaller, indicates that the optimal angle distribution of the resulting Pareto of algorithm must be more uniform, then the effect of algorithm is also better.
It is a kind of comprehensive evaluation index that (4-2-2) superspace, which estimates H, is dominated for non-dominant disaggregation obtained by evaluation algorithms
The size in region, the comprehensive distributivity and constringency performance for measuring Pareto optimal solution.When using the performance indicator, it is necessary first to
Reference point appropriate is chosen, it must be dominated by all non-domination solutions.For superspace estimates H, resulting value is bigger,
The performance of corresponding algorithm is better, shows the more uniform of the distributivity of non-domination solution, closer apart from true Pareto.BB-MOPSO
The selection of the reference point of algorithm and NSGA2 algorithm is all [10000,10000,10000000], and guarantee can be by each non-dominant
Collection dominates.
(4-3) algorithm statistical result compares
(4-3-1) in environment 1, BB-MOPSO algorithm and NSGA2 algorithm are separately operable ten times, measured distributivity sp
It is as follows that metric and superspace estimate H metric detailed results:
Table 1 illustrates BB-MOPSO algorithm and NSGA2 algorithm is separately operable ten times in environment 1, gained distributivity sp degree
Magnitude and superspace estimate the mean value of H metric.
Table 1
In table 1, analysis distribution sp metric is it is found that BB-MOPSO algorithm runs the entirety of resulting Pareto optimal solution
Distribution performance will be substantially better than NSGA2 algorithm；Analysis superspace estimates H metric it is found that Pareto obtained by BB-MOPSO algorithm
Domination range between optimal solution and reference point will be significantly greater than optimal dominate of Pareto obtained by NSGA2 algorithm and solve and reference point
Dominate range；And it is a kind of composite target that superspace, which is estimated, therefore the distributivity of BB-MOPSO algorithm and convergence will
Better than NSGA2 algorithm.
(4-3-2) in environment 2, BB-MOPSO algorithm and NSGA2 algorithm are separately operable ten times, measured distributivity sp
It is as follows that metric and superspace estimate H metric detailed results:
Table 2 illustrates BB-MOPSO algorithm and NSGA2 algorithm is separately operable ten times in environment 2, gained distributivity sp degree
Magnitude and superspace estimate the mean value of H metric.
Table 2
In table 2, for analysis distribution sp metric it is recognized that while obvious not as good as environment 1, the operation of BB-MOPSO algorithm is resulting
The distribution performance of Pareto optimal solution is better than NSGA2 algorithm；Analysis superspace estimates H metric it is found that BB-MOPSO algorithm
Domination range between gained Pareto optimal solution and reference point is also significantly greater than the optimal domination solution of Pareto obtained by NSGA2 algorithm
With the domination range of reference point；And it is a kind of composite target that superspace, which is estimated, thus the distributivity of BB-MOPSO algorithm and
Convergence will be better than NSGA2 algorithm.
Therefore, by the statistical result of environment 1 and environment 2 it is found that the distributivity of BB-MOPSO algorithm Pareto non-domination solution
It is all to be substantially better than NSGA2 algorithm with convergence.
(4-4) result of decision compares
In algorithm execution, each operation all selects optimal one by the fuzzy decision of non-domination solution satisfaction
Solution, then, carries out final decision solution in ten optimal solutions selected.Operation result in two kinds of environment is as follows.
(4-4-1) in environment 1, the final decision track plot that BB-MOPSO algorithm and NSGA2 algorithm generate is respectively such as Fig. 3
And Fig. 4, further, table 3 gives the corresponding performance index value of two tracks.
Table 3
1 interpretation of result of environment: from track length cost value, threaten the truthful data of cost value and concealment cost value can
To find out, performance of the resulting final track of BB-MOPSO algorithm in each target is all substantially better than what NSGA2 algorithmic rule went out
Decision track, this safety to unmanned plane in enemy's flight range more have guarantee.From the actual emulation figure of decision track
From the point of view of, the final convergence region of BB-MOPSO algorithm is close with NSGA2 algorithm.The optimization track of BB-MOPSO algorithm is to a certain degree
On it is more smooth, the mobility of unmanned plane is required smaller, is more suitable flight.
(4-4-2) in environment 2, the final decision track plot that BB-MOPSO algorithm and NSGA2 algorithm generate is respectively such as Fig. 5
And Fig. 6, further, table 4 gives the corresponding performance index value of two tracks.
Table 4
In conclusion threaten and how to be distributed regardless of environment is changed, the decision that BB-MOPSO algorithm search arrives is navigated
Mark is more guaranteed for the flight reappearance of unmanned plane.
The above description is only a preferred embodiment of the present invention, is not intended to restrict the invention, for those skilled in the art
For member, the invention may be variously modified and varied.All within the spirits and principles of the present invention, it is made it is any modification,
Equivalent replacement, improvement etc., are all included in the scope of protection of the present invention.
Claims (10)
1. a kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning, it is characterised in that: the following steps are included:
(1) environmental model is established according to flight environment of vehicle；
(2) trajectory planning model is established according to the environmental model that step (1) is established, is existed using cost function simulation unmanned plane is established
Fitness function in flight course evaluates unmanned aerial vehicle flight path performance；
(3) it is based on multiple target backbone particle swarm optimization algorithm, proposes that particle coding, constraint processing strategie and algorithm execute step,
Track plan model in Optimization Steps (2) plans the global navigation route of unmanned plane；
(4) analogue simulation is carried out according to step (1) to step (3), carries out the global multi-target traces planning of three-dimensional of unmanned plane.
2. a kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning according to claim 1, feature exist
In: in step (1), when handling the flight environment of vehicle of unmanned plane, using numerical map technology, the numerical map technology
Landform is continuously highly carried out discrete, is stored among grid in digital form, the distance between grid size is according to reality
Need to carry out the division of precision；Got by numerical map technology physical relief information in unmanned plane during flying regional scope,
The center of threat information by the processing of equivalent landform and the threat range of deterrent.
3. a kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning according to claim 1, feature exist
In: in step (2), when establishing trajectory planning model, consider that minimum flying height and track pass through massif limiting factor, wherein
Minimum flying height is set according to the maximum height of ground object；The cost function includes: track length cost, threatens cost
And concealment cost.
4. a kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning according to claim 1, feature exist
In: in step (3), particle coding refers to a kind of multinode particle coding strategy, by the problem of continuous domain be mapped to from
It dissipates on domain, a particle represents a unmanned plane during flying path or track, and a paths are determined jointly by m node；It is described about
Beam processing strategie, which refers to, considers constraint present in track in trajectory planning, and provides corresponding constraint processing plan in the algorithm
Slightly, from the one group of Pareto optimal solution finally generated, a best trade-off solution is selected to be supplied to policymaker's use；The boat
Constraint includes: to prevent track from passing over massif or colliding with barrier present in mark.
5. a kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning according to claim 1 to 4,
It is characterized in that: in step (2), using fitness function of the cost function simulation unmanned plane in flight course is established, to nobody
Machine track performance is evaluated, the method is as follows:
(2-1) establishes track length Cost Model by formula (1), and formula (1) is expressed as follows:
Wherein, Pathi=(Pi,1,Pi,1,…,Pi,m) indicate i-th track, Pi,1And Pi,mIt is the starting point S and terminal of unmanned plane
G；I-th track Path of f1 expression unmanned planeiLength cost；Pi,j=(xi,j,yi,j,zi,j) indicate i-th track jth
A node, x, y indicate that the floor projection coordinate of track node, z indicate the corresponding elevation of track node, and m is the section of i-th track
Points；
(2-2) is established by formula (2) and is threatened Cost Model, and formula (2) is expressed as follows:
Wherein, ε andIt is positive number, dang (j-1, j) indicates -1 node of jth to the degree of danger in jth node composition section；A is nothing
The angle of man-machine heading and ground；Dis (j-1, j) is -1 knee level projection coordinate point of jth and jth node water of track
The distance between flat projection coordinate's point；Dis1 (j-1, j) is the distance between -1 node of jth and jth node；dis1(j,j+1)
For the distance between jth node and+1 node of jth；DminAnd DmaxIt is the boundary of the valid interval of radar detection object respectively；
(2-3) establishes concealment Cost Model by formula (3), and formula (3) is expressed as follows:
Wherein, yinb (j-1, j) indicates the hidden index of -1 node of jth and the determined segment of jth node, hmaxIt is digital high
Highest elevation value in journey map；zi,jFor the height value of j-th of node of i-th track；di,jIt is j-th of i-th track
Knee level coordinate corresponds to the height value in digital elevation map；Safth is safe altitude.
6. a kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning according to claim 5, feature exist
In: in step (3), it is based on multiple target backbone particle swarm optimization algorithm, proposes that particle coding, constraint processing strategie and algorithm are held
Row step, the method is as follows:
(3-1) constitutes the decision space and required population, completion space planning and particle coding of trajectory planning；
Bounding algorithm combination BB-MOPSO algorithm is carried out the optimization of Three-dimensional Track by (3-2), is carried out to the particle for being unsatisfactory for constraint
Corresponding processing；
(3-3) formulates corresponding track decision scheme, is supplied to policymaker one final reference solution；
(3-4) is based on step (3-1) to (3-3), proposes that the improvement multiple target backbone population planned towards unmanned aerial vehicle flight path is excellent
Change the execution step of algorithm.
7. a kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning according to claim 6, feature exist
In: in step (3-1), the decision space and required population, completion space planning and particle coding of trajectory planning are constituted,
Include the following steps:
(3-1-1) limits the update amplitude of particle position, i.e., limits the variation range of each node coordinate, if
Set particle position value range are as follows: and (x, y, z) | xmin<x<xmax,ymin<y<ymax,zmin<z<zmax, and then constitute track rule
The decision space drawn, wherein xminAnd xmax、yminAnd ymax、zminAnd zmaxIt is the boundary of the x, y, z axis of each node respectively；
(3-1-2) indicates { P with a group node vector to each particle1,P2,…,Pm-1,Pm, wherein P1And PmRespectively fly
Capable starting point S and target point G, P2,…,Pm-1For the intermediate node of flight path；In the corresponding value range of each node
In, the position of the random initializtion node, arranged in sequence node produced, all nodes form a fullpath, that is, produce
A raw particle position；
(3-1-3) repeats step (3-1-2), generates the particle of defined amount, population needed for forming.
8. a kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning according to claim 6, feature exist
In: in step (3-2), by bounding algorithm combination BB-MOPSO algorithm carry out Three-dimensional Track optimization, to be unsatisfactory for constraint
Particle performs corresponding processing, and includes the following steps:
(3-2-1) in more new particle, according to each component of the node sequencing successively more new node of particle；For any
One particle judges whether the node feasible using following methods when updating one of node: present node with it is previous
N point, number m are taken on segment determined by a node1,m2,…,mn, compare set in each point practical height value with
Corresponding height value in numerical map is lower than digital elevation value if there is the height value of a point, then needs to update present node
Position；Continue to judge whether updated node meets condition according to the method described above, until just can be carried out down after meeting condition
The update of one node；
(3-2-2) limits minimum track height accordingly, by minimum track height, calculates path in step (2-3)
Concealment cost.
9. a kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning according to claim 6, feature exist
In: in step (3-3), corresponding track decision scheme is formulated, is supplied to policymaker one final reference solution, method is such as
Under:
By to non-domination solution XkSatisfaction μkIt is calculated, the non-domination solution of Maximum Satisfaction is final selected flight boat
Mark, wherein μkCalculation formula it is as follows:
In formula, | Ar | it is the number of non-domination solution；M is the number of objective function；μi kIt is non-domination solution in i-th objective function
Satisfaction, membership function are as follows:
The maximum and minimum value of respectively i-th objective function, Fi(Xk) it is non-domination solution XkIn i-th of target letter
Several values.
10. a kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning according to claim 6, feature exist
In: in step (3-4), the execution step of the improvement multiple target backbone particle swarm optimization algorithm towards unmanned aerial vehicle flight path planning is such as
Under:
(3-4-1) setting algorithm relevant parameter, the scale Ns including population, the number of nodes Nt of population, external storage collection
Capacity is Na, the number of iterations Ts；
(3-4-2) encodes the particle in population, the position of each particle in random initializtion population, and setting is each
The individual extreme point Pbest of particle is the initial position of the particle；
(3-4-3) evaluates the target function value of each particle, i.e. step (2-1) the track length cost, step (2-2) is described
Cost and step (2-3) the concealment cost are threatened, the non-domination solution in population is returned, non-domination solution is saved in outside
In storage set, and assignment is carried out to the global extreme point Gbest of particle；
The Gaussian Profile of (3-4-4) based on Pbest and Gbest carries out location updating to particle, judges the particle updated production
Whether track section track passes through massif, and new position is regenerated if passing through, until the corresponding track of produced particle not
Massif is passed through again；
(3-4-5) evaluates the target function value of each particle, i.e. step (2-1) the track length cost, step (2-2) is described
Cost and step (2-3) the concealment cost are threatened, individual extreme point Pbest is updated by dominance relation；
(3-4-6) carries out Pareto ratio for the previous generation non-domination solution of the new population non-domination solution generated and external storage concentration
Compared with being updated to external storage collection, judge whether external storage collection is greater than the capacity Na of external storage collection, if it is, right
Population is cut；
The global extreme point Gbest of (3-4-7) more new particle；
(3-4-8) judges whether the termination condition for reaching algorithm, Pareto optimal solution set is if it is exported, if weighing without if
Multiple above (3-4-4) to (3-4-7) step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811583287.XA CN109631900B (en) | 2018-12-24 | 2018-12-24 | Unmanned aerial vehicle three-dimensional flight path multi-target particle swarm global planning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811583287.XA CN109631900B (en) | 2018-12-24 | 2018-12-24 | Unmanned aerial vehicle three-dimensional flight path multi-target particle swarm global planning method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109631900A true CN109631900A (en) | 2019-04-16 |
| CN109631900B CN109631900B (en) | 2022-08-26 |
Family
ID=66076902
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811583287.XA Active CN109631900B (en) | 2018-12-24 | 2018-12-24 | Unmanned aerial vehicle three-dimensional flight path multi-target particle swarm global planning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109631900B (en) |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110187716A (en) * | 2019-06-17 | 2019-08-30 | 中国地质大学（北京） | Geological exploration UAV Flight Control method and apparatus |
| CN110181508A (en) * | 2019-05-09 | 2019-08-30 | 中国农业大学 | Underwater robot three-dimensional Route planner and system |
| CN110544296A (en) * | 2019-07-31 | 2019-12-06 | 中国矿业大学 | intelligent planning method for three-dimensional global flight path of unmanned aerial vehicle in environment with uncertain enemy threat |
| CN110706519A (en) * | 2019-10-11 | 2020-01-17 | 中国人民解放军63629部队 | Real-time planning method and device for aircraft route and computer equipment |
| CN111024085A (en) * | 2019-12-18 | 2020-04-17 | 四川大学 | Unmanned aerial vehicle track planning method with end point direction and time constraints |
| CN111580556A (en) * | 2020-05-22 | 2020-08-25 | 中国人民解放军国防科技大学 | Multi-unmanned aerial vehicle collaborative path planning and guidance method under space-time constraint |
| CN111580564A (en) * | 2020-04-30 | 2020-08-25 | 湖南科技大学 | Parallel multi-target search cooperative operation method for three-dimensional group unmanned aerial vehicle |
| CN112051750A (en) * | 2020-08-07 | 2020-12-08 | 广东工业大学 | Four rotor unmanned aerial vehicle digital twin system |
| CN112070628A (en) * | 2020-09-02 | 2020-12-11 | 沈阳工程学院 | Intelligent power grid multi-target economic dispatching method considering environmental factors |
| CN112230678A (en) * | 2020-10-29 | 2021-01-15 | 皖江工学院 | Three-dimensional unmanned aerial vehicle path planning method and planning system based on particle swarm optimization |
| CN112733251A (en) * | 2020-12-24 | 2021-04-30 | 中国人民解放军空军工程大学 | Multi-unmanned aerial vehicle collaborative track planning method |
| CN112783207A (en) * | 2020-12-31 | 2021-05-11 | 哈尔滨工程大学 | Unmanned aerial vehicle flight path planning method based on multi-objective particle swarm optimization |
| CN112880688A (en) * | 2021-01-27 | 2021-06-01 | 广州大学 | Unmanned aerial vehicle three-dimensional flight path planning method based on chaotic self-adaptive sparrow search algorithm |
| CN113190041A (en) * | 2021-04-30 | 2021-07-30 | 哈尔滨工业大学 | Unmanned aerial vehicle cluster online target distribution method based on constraint relaxation technology |
| CN113253760A (en) * | 2021-06-08 | 2021-08-13 | 北京远度互联科技有限公司 | Path planning method and device, movable carrier and storage medium |
| CN113485419A (en) * | 2021-07-01 | 2021-10-08 | 广东省建筑工程监理有限公司 | Unmanned aerial vehicle initial flight path planning method for soil pollution boundary survey |
| CN113671985A (en) * | 2021-07-28 | 2021-11-19 | 中国人民解放军32146部队 | Staged multi-base unmanned aerial vehicle task allocation and flight path planning method |
| CN113759958A (en) * | 2021-07-07 | 2021-12-07 | 哈尔滨工程大学 | Unmanned aerial vehicle formation flight path planning method based on positioning precision |
| CN113987718A (en) * | 2021-11-17 | 2022-01-28 | 中国矿业大学 | Product change scheme design method based on multi-objective particle swarm optimization |
| CN116126032A (en) * | 2023-04-17 | 2023-05-16 | 华南农业大学 | Unmanned aerial vehicle group path planning method based on improved multi-objective evolutionary algorithm |
| CN116518982A (en) * | 2023-07-03 | 2023-08-01 | 江西啄木蜂科技有限公司 | Low-altitude forestry monitoring remote sensing unmanned aerial vehicle path multi-target planning method |
| CN116822362A (en) * | 2023-06-29 | 2023-09-29 | 中国人民解放军93209部队 | Unmanned aerial vehicle conflict-free four-dimensional flight path planning method based on particle swarm optimization |
| CN112733251B (en) * | 2020-12-24 | 2024-04-19 | 中国人民解放军空军工程大学 | Collaborative flight path planning method for multiple unmanned aerial vehicles |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060031004A1 (en) * | 2003-10-13 | 2006-02-09 | Kristian Lundberg | Method and device for planning a trajectory |
| CN105700549A (en) * | 2016-01-21 | 2016-06-22 | 北京理工大学 | Unmanned plane multi-track planning method based on sequence ecological niche PSO (particle swarm optimization) algorithm |
| CN105841702A (en) * | 2016-03-10 | 2016-08-10 | 赛度科技（北京）有限责任公司 | Method for planning routes of multi-unmanned aerial vehicles based on particle swarm optimization algorithm |
| CN107356254A (en) * | 2017-06-07 | 2017-11-17 | 东南大学 | Suitable for the particle group optimizing method of geomagnetic auxiliary navigation trajectory planning |
| CN107622327A (en) * | 2017-09-15 | 2018-01-23 | 哈尔滨工程大学 | Multiple no-manned plane path planning method based on cultural ant colony search mechanisms |
| CN108680168A (en) * | 2018-04-19 | 2018-10-19 | 哈尔滨工程大学 | A kind of hidden paths planning method of underwater hiding-machine based on transparency and spring layer data |
-
2018
- 2018-12-24 CN CN201811583287.XA patent/CN109631900B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060031004A1 (en) * | 2003-10-13 | 2006-02-09 | Kristian Lundberg | Method and device for planning a trajectory |
| CN105700549A (en) * | 2016-01-21 | 2016-06-22 | 北京理工大学 | Unmanned plane multi-track planning method based on sequence ecological niche PSO (particle swarm optimization) algorithm |
| CN105841702A (en) * | 2016-03-10 | 2016-08-10 | 赛度科技（北京）有限责任公司 | Method for planning routes of multi-unmanned aerial vehicles based on particle swarm optimization algorithm |
| CN107356254A (en) * | 2017-06-07 | 2017-11-17 | 东南大学 | Suitable for the particle group optimizing method of geomagnetic auxiliary navigation trajectory planning |
| CN107622327A (en) * | 2017-09-15 | 2018-01-23 | 哈尔滨工程大学 | Multiple no-manned plane path planning method based on cultural ant colony search mechanisms |
| CN108680168A (en) * | 2018-04-19 | 2018-10-19 | 哈尔滨工程大学 | A kind of hidden paths planning method of underwater hiding-machine based on transparency and spring layer data |
Cited By (39)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110181508B (en) * | 2019-05-09 | 2021-01-12 | 中国农业大学 | Three-dimensional route planning method and system for underwater robot |
| CN110181508A (en) * | 2019-05-09 | 2019-08-30 | 中国农业大学 | Underwater robot three-dimensional Route planner and system |
| CN110187716A (en) * | 2019-06-17 | 2019-08-30 | 中国地质大学（北京） | Geological exploration UAV Flight Control method and apparatus |
| CN110544296A (en) * | 2019-07-31 | 2019-12-06 | 中国矿业大学 | intelligent planning method for three-dimensional global flight path of unmanned aerial vehicle in environment with uncertain enemy threat |
| CN110544296B (en) * | 2019-07-31 | 2023-05-30 | 中国矿业大学 | Intelligent planning method for three-dimensional global track of unmanned aerial vehicle in uncertain enemy threat environment |
| CN110706519A (en) * | 2019-10-11 | 2020-01-17 | 中国人民解放军63629部队 | Real-time planning method and device for aircraft route and computer equipment |
| CN111024085A (en) * | 2019-12-18 | 2020-04-17 | 四川大学 | Unmanned aerial vehicle track planning method with end point direction and time constraints |
| CN111024085B (en) * | 2019-12-18 | 2020-10-16 | 四川大学 | Unmanned aerial vehicle track planning method with end point direction and time constraints |
| CN111580564B (en) * | 2020-04-30 | 2022-06-10 | 湖南科技大学 | Parallel multi-target search cooperative operation method for three-dimensional group unmanned aerial vehicle |
| CN111580564A (en) * | 2020-04-30 | 2020-08-25 | 湖南科技大学 | Parallel multi-target search cooperative operation method for three-dimensional group unmanned aerial vehicle |
| CN111580556A (en) * | 2020-05-22 | 2020-08-25 | 中国人民解放军国防科技大学 | Multi-unmanned aerial vehicle collaborative path planning and guidance method under space-time constraint |
| CN111580556B (en) * | 2020-05-22 | 2023-04-07 | 中国人民解放军国防科技大学 | Multi-unmanned aerial vehicle collaborative path planning and guidance method under space-time constraint |
| CN112051750B (en) * | 2020-08-07 | 2023-10-03 | 广东工业大学 | Digital twin system of four-rotor unmanned aerial vehicle |
| CN112051750A (en) * | 2020-08-07 | 2020-12-08 | 广东工业大学 | Four rotor unmanned aerial vehicle digital twin system |
| CN112070628B (en) * | 2020-09-02 | 2023-07-11 | 沈阳工程学院 | Multi-target economic dispatching method for smart power grid considering environmental factors |
| CN112070628A (en) * | 2020-09-02 | 2020-12-11 | 沈阳工程学院 | Intelligent power grid multi-target economic dispatching method considering environmental factors |
| CN112230678A (en) * | 2020-10-29 | 2021-01-15 | 皖江工学院 | Three-dimensional unmanned aerial vehicle path planning method and planning system based on particle swarm optimization |
| CN112230678B (en) * | 2020-10-29 | 2023-07-14 | 皖江工学院 | Three-dimensional unmanned aerial vehicle path planning method and system based on particle swarm optimization |
| CN112733251A (en) * | 2020-12-24 | 2021-04-30 | 中国人民解放军空军工程大学 | Multi-unmanned aerial vehicle collaborative track planning method |
| CN112733251B (en) * | 2020-12-24 | 2024-04-19 | 中国人民解放军空军工程大学 | Collaborative flight path planning method for multiple unmanned aerial vehicles |
| CN112783207A (en) * | 2020-12-31 | 2021-05-11 | 哈尔滨工程大学 | Unmanned aerial vehicle flight path planning method based on multi-objective particle swarm optimization |
| CN112783207B (en) * | 2020-12-31 | 2022-07-15 | 哈尔滨工程大学 | Unmanned aerial vehicle flight path planning method based on multi-objective particle swarm optimization |
| CN112880688B (en) * | 2021-01-27 | 2023-05-23 | 广州大学 | Unmanned aerial vehicle three-dimensional track planning method based on chaotic self-adaptive sparrow search algorithm |
| CN112880688A (en) * | 2021-01-27 | 2021-06-01 | 广州大学 | Unmanned aerial vehicle three-dimensional flight path planning method based on chaotic self-adaptive sparrow search algorithm |
| CN113190041A (en) * | 2021-04-30 | 2021-07-30 | 哈尔滨工业大学 | Unmanned aerial vehicle cluster online target distribution method based on constraint relaxation technology |
| CN113190041B (en) * | 2021-04-30 | 2022-05-10 | 哈尔滨工业大学 | Unmanned aerial vehicle cluster online target distribution method based on constraint relaxation technology |
| CN113253760A (en) * | 2021-06-08 | 2021-08-13 | 北京远度互联科技有限公司 | Path planning method and device, movable carrier and storage medium |
| CN113253760B (en) * | 2021-06-08 | 2021-11-09 | 北京远度互联科技有限公司 | Path planning method and device, movable carrier and storage medium |
| CN113485419A (en) * | 2021-07-01 | 2021-10-08 | 广东省建筑工程监理有限公司 | Unmanned aerial vehicle initial flight path planning method for soil pollution boundary survey |
| CN113759958A (en) * | 2021-07-07 | 2021-12-07 | 哈尔滨工程大学 | Unmanned aerial vehicle formation flight path planning method based on positioning precision |
| CN113759958B (en) * | 2021-07-07 | 2023-08-01 | 哈尔滨工程大学 | Unmanned aerial vehicle formation track planning method based on positioning precision |
| CN113671985A (en) * | 2021-07-28 | 2021-11-19 | 中国人民解放军32146部队 | Staged multi-base unmanned aerial vehicle task allocation and flight path planning method |
| CN113987718A (en) * | 2021-11-17 | 2022-01-28 | 中国矿业大学 | Product change scheme design method based on multi-objective particle swarm optimization |
| CN116126032B (en) * | 2023-04-17 | 2023-07-14 | 华南农业大学 | Unmanned aerial vehicle group path planning method based on improved multi-objective evolutionary algorithm |
| CN116126032A (en) * | 2023-04-17 | 2023-05-16 | 华南农业大学 | Unmanned aerial vehicle group path planning method based on improved multi-objective evolutionary algorithm |
| CN116822362A (en) * | 2023-06-29 | 2023-09-29 | 中国人民解放军93209部队 | Unmanned aerial vehicle conflict-free four-dimensional flight path planning method based on particle swarm optimization |
| CN116822362B (en) * | 2023-06-29 | 2024-01-30 | 中国人民解放军93209部队 | Unmanned aerial vehicle conflict-free four-dimensional flight path planning method based on particle swarm optimization |
| CN116518982A (en) * | 2023-07-03 | 2023-08-01 | 江西啄木蜂科技有限公司 | Low-altitude forestry monitoring remote sensing unmanned aerial vehicle path multi-target planning method |
| CN116518982B (en) * | 2023-07-03 | 2023-09-19 | 江西啄木蜂科技有限公司 | Low-altitude forestry monitoring remote sensing unmanned aerial vehicle path multi-target planning method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109631900B (en) | 2022-08-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109631900A (en) | A kind of no-manned plane three-dimensional track multi-objective particle swarm Global Planning | |
| CN102880186B (en) | flight path planning method based on sparse A* algorithm and genetic algorithm | |
| CN107169608B (en) | Distribution method and device for multiple unmanned aerial vehicles to execute multiple tasks | |
| CN107807665B (en) | Unmanned aerial vehicle formation detection task cooperative allocation method and device | |
| CN112082552A (en) | Unmanned aerial vehicle flight path planning method based on improved hybrid particle swarm optimization algorithm | |
| CN111240353B (en) | Unmanned aerial vehicle collaborative air combat decision method based on genetic fuzzy tree | |
| CN105426970B (en) | A kind of meteorological intimidation estimating method based on discrete dynamic Bayesian network | |
| Yu et al. | A knee-guided differential evolution algorithm for unmanned aerial vehicle path planning in disaster management | |
| CN106873599A (en) | Unmanned bicycle paths planning method based on ant group algorithm and polar coordinate transform | |
| CN107063255A (en) | A kind of three-dimensional Route planner based on improvement drosophila optimized algorithm | |
| CN110926477A (en) | Unmanned aerial vehicle route planning and obstacle avoidance method | |
| CN102901500A (en) | Aircraft optimal path determination method based on mixed probability A star and agent | |
| CN110181508A (en) | Underwater robot three-dimensional Route planner and system | |
| CN114840020A (en) | Unmanned aerial vehicle flight path planning method based on improved whale algorithm | |
| CN112666981B (en) | Unmanned aerial vehicle cluster dynamic route planning method based on dynamic group learning of original pigeon group | |
| Wang et al. | Monitoring trajectory optimization for unmanned surface vessel in sailboat race | |
| CN113848919A (en) | Ant colony algorithm-based indoor AGV path planning method | |
| CN110986960A (en) | Unmanned aerial vehicle track planning method based on improved clustering algorithm | |
| CN112414405A (en) | Unmanned aerial vehicle cluster flight path planning method considering emergency task of DSM | |
| CN113268087A (en) | Flight path planning method for cooperative work of multiple unmanned aerial vehicles based on improved ant colony algorithm in multi-constraint complex environment | |
| CN116225066A (en) | Unmanned aerial vehicle path optimization method based on chaotic mapping pely optimization algorithm | |
| CN111121784A (en) | Unmanned reconnaissance aircraft route planning method | |
| CN104729509A (en) | Route planning method based on non-dominated sorting genetic algorithm II | |
| Ma et al. | A Fast path re-planning method for UAV based on improved A* algorithm | |
| Shan | Study on submarine path planning based on modified ant colony optimization algorithm |
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 |