CN106020230B - A kind of multiple no-manned plane method for allocating tasks under power consumption constraint - Google Patents
A kind of multiple no-manned plane method for allocating tasks under power consumption constraint Download PDFInfo
- Publication number
- CN106020230B CN106020230B CN201610341065.1A CN201610341065A CN106020230B CN 106020230 B CN106020230 B CN 106020230B CN 201610341065 A CN201610341065 A CN 201610341065A CN 106020230 B CN106020230 B CN 106020230B
- Authority
- CN
- China
- Prior art keywords
- task
- point
- engine
- air route
- unmanned plane
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
Abstract
The present invention relates to the multiple no-manned plane method for allocating tasks under power consumption constraint, unmanned plane form into columns by two frame task engines and a frame service organization at, each task point is distributed to each task engine using Revised genetic algorithum by host computer, and introduce three exchange crossover operators, the unmanned plane formation air route made is minimum, and the air route of each task engine is balanced as far as possible；Simultaneously, consider the power consumption constraint problem of task engine, to keep the energy of task engine consuming few as far as possible, the air route point of task engine is determined on the basis of the ideal air route of task engine, and energy recharge is provided in point by server, the distribution of all task points is completed in the complete air route for finally obtaining unmanned plane.Compared to traditional multiple no-manned plane Task Allocation Problem, the present invention improves the efficiency that multiple no-manned plane system completes task, keeps the road air route of multiple tasks machine access task point as balanced as possible；The setting method for introducing energy recharge point, for guaranteeing that smoothly completing for unmanned plane task distribution has very great meaning.
Description
Technical field
The invention patent relates to multiple no-manned plane tasks to distribute field, for solving aerial more unmanned vehicle limited energies
Task Allocation Problem.
Background technique
With to robot technology further investigation and system capability be obviously improved, robot system applies
Rapid development is arrived.Currently, robot can be applied to wider field, robot replaces the mankind so that the mankind from it is heavy,
It frees in dangerous labour, especially some risk are higher or the unapproachable special occasions of the mankind, such as core is let out
The application demands such as leakage, the monitoring of aerial unmanned plane, Mars exploration, fire rescue and military anti-terrorism.For complicated cumbersome
Job task, the time spent by individual machine people's completion task is longer, and the factors such as low efficiency cause single robot that can not expire
Foot completes the demand of certain tasks, and therefore, multiple-mobile-robot system can be applied to industrial and agricultural production, military and national defense by the mankind
It human labour intensity is reduced, improves working efficiency with being realized in the fields such as scientific research, avoid casualties and the extension mankind
The purpose of activity space.
Replace soldier to execute dangerous task using robot in modern war, can reduce to the maximum extent ground forces and
The injures and deaths of non-belligerent personnel robot cooperated concurrently completed different sons and appointed due to the decomposability of these tasks with multiple
Business will obviously be than single robot much faster.For multi-robot system, task distribution is effective use multiple mobile robot system
Unite resource, to give full play to the important foundation of system effectiveness advantage, importance degree with DBMS member function difference and
The increase of the structural complexity of task and increase.The genetic algorithm that professor Holland of Michigan university of the U.S. proposes
(Genetic Algorithm, GA) is to solve for the effective ways of complicated combinatorial optimization problem.It is distributed in multi-robotic task and is
In system, the model of Task Allocation Problem can be analyzed to obtain by multiple traveling salesmen problem.The Mathematical Planning of earliest traveling salesman problem
It is to be proposed by Dantzig (1959) et al., multiple traveling salesmen problem (Multiple Traveling Salesman Problem, after
Abbreviation MTSP) refer to M travelling salesman from the same city or different cities, an itinerary is walked respectively, except city of setting out
Outside city, so that one and only one travelling salesman of each city passes through, it is eventually returned to city of setting out, and total distance is most short.
In task assignment procedure, each robot is equivalent to each travelling salesman, and each city of travelling salesman's access is equivalent to machine
Each task point of device people access, the difference is that the process of task distribution is class travelling salesman model, as travelling salesman is not necessarily to
Return to the close copy of starting point.
Since the above conventional method does not consider the problem of each task engine air route equilibrium in task assignment procedure, Yi Jiren
The problem of business machine power consumption constraint, and for unmanned aerial vehicle (UAV) control field, task engine power consumption constraint, which is the key that again one, to be solved
Problem.
Summary of the invention
The technical problem to be solved by the present invention is to, in view of the above technical defects, provide under a kind of power consumption constraint mostly nobody
The method of machine task distribution, can front task to multiple no-manned plane carry out reasonable distribution and carry out air route optimization；Meanwhile it solving
The supply problem of multiple no-manned plane energy consumption；The distribution of other air taskings such as the following air fighting battlefield and air transport is adapted it to ask
Topic.
In order to solve the above technical problems, the present invention adopts the following technical scheme:
Multiple no-manned plane method for allocating tasks under a kind of power consumption constraint, it is characterised in that form into columns in given unmanned plane, each
Business point and under the consumption constraint of task function, minimizes air route as Target Assignment task simultaneously using balance optimizing air route, to make times
The energy that business machine expends is few as far as possible, the air route point of task engine is determined on the basis of the ideal air route of task engine, and by servicing
Machine provides energy recharge in point, finally obtains the complete air route of unmanned plane, completes the distribution of all task points；It specifically includes
Following steps:
Step 1: the initialization of task distribution system；At the beginning of mainly including host computer initialization, ground control cabinet initialization, clock
Beginningization, unmanned plane, which are formed into columns, to be initialized；The unmanned plane formation has at least three frame unmanned planes, and respectively task engine A, task engine B are
The server C of the task engine offer energy；There is multiple tasks point in setting front, and task point number is more than task engine number, nobody
Machine is formed into columns before receiving out and saying the word, and is awaited orders positioned at starting point O；
Step 2: establish the communication between ground control cabinet, position location satellite and unmanned plane formation: when executing front task,
Obtain the location information of each task point, unmanned plane in real time by satellite positioning tech；
Step 3: the location information of each task point of unmanned plane and front is returned to ground control cabinet, ground in real time by position location satellite
Face console real-time reception, analysis simultaneously handle these location informations；
Step 4: host computer introduces three exchange crossover operators for each task point and distributes to each by Revised genetic algorithum
Business machine obtains ideal air route l0, i.e., task engine is not by the air route of power consumption constraint；Establishing task point distribution model is class MTSP mould
The class MTSP model of type, as travelling salesman without return；
Step 5: obtain the position of energy recharge point P: host computer combination constraint condition simulates task engine in preliminary ideal
Air route l0Upper flight obtains first energy recharge point P0Position；
Step 6: setting unmanned plane, which is formed into columns, reaches first energy recharge point P0Afterwards, task distribution system is to remaining task
Point makees task distribution again, while seeking next energy recharge point, i.e. circular flow step 2, step 3, step 4, step always
Rapid 5, until completing the distribution of all task points；
Step 7: after all task points are assigned, console issues control command, and unmanned plane, which is formed into columns, receives order, from
O point sets out, and goes to each task point to execute distributing for task along specified planning air route.
In above-mentioned technical proposal, task engine A and task engine B are isomorphism unmanned plane, server C and task engine be isomery nobody
Machine, and the limitation of server C energy consumption is enough to ensure that and completes all tasks, task engine A and task engine B limited energy need to be in flight courses
Middle progress energy supplement.
In above-mentioned technical proposal, the energy recharge point P be task engine depleted of energy point and server C with
The point of task engine A, task engine B；And disregard and converge the time spent by the moment, the flight route of server C is follow
Point successively goes to each energy recharge point；Judge whether each task engine needs to carry out energy recharge, energy in each iteration cycle
Source supply point P is located at the line midpoint of a certain moment task engine A and B, when a team, which forms into columns, reaches energy recharge point P, task
Machine A, task engine B the energy exhausted, the flight path of server C is by starting point O to the straight line path of energy recharge point P
Diameter.
In above-mentioned technical proposal, when the air route length of individual task machine is up to air route limiting value, which is gone to
Energy recharge point.
In above-mentioned technical proposal, after setting task engine formation has accessed all task points in step 4, the last one is rested on
Task point, without returning to starting point, obtain task engine it is preliminary not by the air route l of power consumption constraint0, namely ideal air route l0。
Revised genetic algorithum of the present invention distributes each task point to each task engine, and introducing three exchange crossover operators makes
The total air route of task engine is most short, each task engine air route is substantially balanced.Task Allocation Problem can be described as: there are multiple tasks in front
Point, and task point number is more than task engine number, the aerial unmanned plane formation of a team is just awaited orders in starting point, and host computer waiting passes through
The location information of task point and unmanned plane that satellite positioning tech is passed back, when the method for salary distribution of each task point in front battlefield determines
Afterwards, control command is issued by ground control cabinet, task engine is connected to new order and goes to investigate the point of some determining positions, reaches specified
Location point go to execute a certain task.
In the present invention, bevel edge principle is less than by triangular right-angle side, point P is set at the line midpoint of task engine A, B
Place, can guarantee the efficient utilization rate of the energy, make task engine during deviateing ideal air route and reaching energy recharge point P, institute
The energy of consumption minimizes, the final position for determining energy recharge point P；The time for disregarding the consuming of energy recharge point place, then exist
After the completion of non-renewable energy supply, unmanned plane formation restarts to be allocated remaining task point, while seeking next energy
Source supply point, until completing the distribution of all task points.
During unmanned plane task of the invention is formed into columns, task engine A and task engine B are isomorphism, and server C is different with task engine
Structure unmanned plane, isomery unmanned plane have different hardware configuration and body structure, execute investigation for different type task point, visit
The tasks such as ask, hit；Server as energy recharge is high-performance, high-cost unmanned plane, and flying speed can be more than 3 horses
It is conspicuous that (flight Mach number of unmanned plane is the ratio of velocity of sound in the flying speed and its flying height of unmanned plane, such as mach one or 1M
As one times of velocity of sound), meanwhile, the engine thrust-weight ratio of the unmanned plane in this method is set as 9, and (thrust ratio of unmanned plane refers to hair
Motivation thrust and the ratio between engine weight or body weight, it indicates thrust caused by engine or body Unit Weight, mesh
Generally between 7.5-9.0,11.0) U.S.'s this numerical value of F-119 engine can reach the thrust ratio of the advanced engine of former world,
It follows that the cost of server is accordingly higher, therefore the above formation mode can be efficiently complete with lesser operation cost
At combat duty.Task engine A and task engine B during flying to each task point, energy consumption be it is restricted, server C is
High Performance Unmanned Aerial Vehicle, therefore suffered energy consumption restriction is weaker, during execution task, the energy is enough to ensure that completion is appointed
Business.
In conclusion the present invention relates to a kind of methods of solution multiple no-manned plane limited energy task distribution.The unmanned plane
It forms into columns by two frame task engines and a frame service organization at suitable for the following air fighting battlefield and other Task Allocation Problems.
By technologies such as satellite positionings, the location information of unmanned plane and task point is obtained in real time；Ground control cabinet receives position location satellite and passes
The location information returned；Each task point is distributed to each task engine using Revised genetic algorithum by host computer, is calculated in traditional heredity
Three exchange crossover operators are introduced in method, the unmanned plane formation air route made is minimum, and the air route of each task engine is balanced as far as possible；
Meanwhile considering the power consumption constraint problem of task engine, to keep the energy of task engine consuming few as far as possible, in the ideal air route base of task engine
The air route point of task engine is determined on plinth, and provides energy recharge in point by server.Finally obtain the complete of unmanned plane
The distribution of all task points is completed in whole air route.Compared to traditional multiple no-manned plane Task Allocation Problem, the invention proposes one kind
Low cost, the simple unmanned plane formation mode of structure；Secondly, the present invention introduces three exchanges in traditional genetic algorithm intersects calculation
Son, the unmanned plane formation air route made is small as far as possible, improves the efficiency that multiple no-manned plane system completes task, limits each task
Machine accesses the deadline of task point, keeps the road air route of multiple tasks machine access task point as balanced as possible, takes full advantage of more
UAV system；Again, the present invention forms into columns for the unmanned plane that one group of energy will exhaust, and introduces the setting of energy recharge point
Method, for guaranteeing that smoothly completing for unmanned plane task distribution has very great meaning.
Detailed description of the invention
Fig. 1 is the multiple no-manned plane method for allocating tasks communication scheme under power consumption constraint of the present invention；
Fig. 2 is the multiple no-manned plane task allocation process diagram under power consumption constraint of the present invention；
Fig. 3 is task engine ideal of the present invention air route schematic diagram；
Fig. 4 is the multiple no-manned plane task distribution air route schematic diagram under power consumption constraint of the present invention.
Specific implementation measure
Technical solution in order to further illustrate the present invention, the present invention will be described in detail for control attached drawing.
Fig. 1 is the communication scheme between ground control cabinet of the present invention, position location satellite and unmanned plane are formed into columns.
Obtain the location information of each unmanned plane, task point by position location satellite, position location satellite is by each task point, unmanned seat in the plane
Confidence breath is returned to ground control cabinet, and ground control cabinet receives, analyzes and handle these location informations, is calculated by improved heredity
Method obtains the task distribution path of task engine, to obtain the planning air route of task distribution.
Fig. 2 is the flow chart of the multiple no-manned plane method for allocating tasks under power consumption constraint of the present invention, is mainly comprised the steps that
Step 1: the initialization of task distribution system, at the beginning of mainly including host computer initialization, ground control cabinet initialization, clock
Beginningization, unmanned plane, which are formed into columns, to be initialized；
Step 1-1: setting unmanned plane is formed into columns: A, B, C.Wherein A, B are task engine, and C is server.Unmanned plane formation is connecing
It receives out before saying the word, is located at starting point O；
Step 1-2: unmanned plane initialization specifically includes that gyroscope initialization, accelerometer initialization, magnetometer are initial
Change, barometer initialization, GPS initialization, camera initialization, single-chip microcontroller initialization；
Step 2: establishing the communication between ground control cabinet, position location satellite and unmanned plane formation；
Step 3: carrying out information transmitting between ground control cabinet, position location satellite and unmanned plane formation.Ground as shown in Figure 1
Communication scheme between face console, position location satellite and unmanned plane formation；
Step 3-1: when executing the air mission in front battlefield, each task is obtained in real time by technologies such as satellite positionings
The location information of point, unmanned plane；
Step 3-2: each task point, unmanned plane location information are returned to ground control cabinet by position location satellite；
Step 3-3: ground control cabinet receives, analyzes and handle the location information passed back；
Step 4: each task point is distributed to two frame task engines by Revised genetic algorithum by host computer, obtains task mechanism
Think air route l0。
Step 4-1: task engine ideal air route as shown in Figure 3 schematic diagram.Task point distribution model is established, mathematical model is such as
Under:
Wherein, xijk、ykiFor 0-1 decision variable: working as xijkWhen=1, indicate that task engine k passes through arc (i, j), otherwise xijk=
0；Work as ykiWhen=1, indicate that task engine k accesses task point i, otherwise yki=0；The starting point O for indicating unmanned plane formation with 0, referred to as
Initial point；Point 1 ..., n indicates the task point that 2 frame task engines need to access；cijIt is traversed by task engine by corresponding segmental arc (i, j)
Distance；tkThe time consumed after having flown for k-th of task engine；vkFor the speed of k-th of task engine, in which:
v1=vA=1M；
v2=vB=1M；
v3=vC=3M；
Wherein, 1M or mach one are 1 times of velocity of sound；
Objective function are as follows:
T=min (max (t1,t2)) (1)
Wherein:
Constraint condition are as follows:
X=(xijk)∈S (6)
In the model, formula (1) indicates in two frame task engines, keeps the access task point time maximum task engine used time minimum
Change；Formula (2) indicates that each task engine has accessed the time consumed after all task points；Formula (3) indicate from starting point 0 (namely go out
Hair point O) it sets out, all task points are only strictly accessed by some task engine primary；The terminal task of formula (4) expression any bar arc
Point only has a starting point task point and is attached thereto；Formula (5) indicates that the starting point task point of any bar arc only has a terminal task point
It is attached thereto；Formula (6) indicates that cancellation constitutes the solution of imperfect route, and S is that branch eliminates constraint, i.e. cancellation constitutes imperfect route
Solution.
Step 4-2: the optimal conditions that each task presses illuminated (1) are distributed into task engine A using Revised genetic algorithum
And B: being intersected using three exchanges, and chromosome coding and matrix coding/decoding method repeatedly select population, intersect, make a variation
Equal genetic manipulations, constantly produce the population of new generation that environment is more adapted to than parent.
Step 4-2-1: genetic coding is carried out to task.Task engine A, B carry out task distribution, share N+1 task point, use
Point 0 indicates the starting point of all unmanned planes, and N indicates the task point that 2 frame task engines need to access to point 1 ....Additional 1 virtual symbol
Number, indicate 1 virtual task point.There is virtual sign and mean that appoint each time in they and the coordinate having the same of starting point 0
Business machine accesses starting point 0, but in the case where completing for task access, does not return to starting point 0 when there is virtual sign.It is each
Route indicates the access air route of some task engine.To avoid the occurrence of meaningless subpath, it will be assumed that from task point 0 to
The distance of task point 0 is infinitely great.For example, working as N=6, that is, when having 6 task points, chromosome code is 0~7, wherein " 7 " are
The virtual sign of addition, the Task Allocation Problem of 2 frame task engines, its item chromosome coding is:
| 1 | 2 | 5 | 7 | 4 | 3 | 6 |
Then the air route of task engine A, B respectively indicates are as follows:
0-1-2-5 0-4-3-6
During algorithm operating, it is possible that virtual sign is the both ends of item chromosome the case where:
| 1 | 2 | 5 | 6 | 4 | 3 | 7 |
Then the air route of task engine A, B indicates at this time are as follows:
1-2-5-6-4-3 0-0
Distance from task point 0 to task point 0 is infinitely great, so the distance maximum value in task engine at present is infinite
Greatly.Again since Prescribed Properties air route maximum value minimizes, so such chromosome can be eliminated.
Step 4-2-2: population size selection.Suitable population size is of great significance to the convergence of genetic algorithm.Group
Body is too small to be difficult to acquire satisfied as a result, group then calculates greatly very much complexity.Rule of thumb, population size generally takes 10~200.
Step 4-2-3: fitness function.Since optimization aim is to minimize the deadline, objective function is enabled to make exponential transform
Obtain fitness function:
F=a exp (- b × T) (7)
Wherein: a, b are positive real number.
Step 4-2-4: selection.Using the method for roulette selection, the probability that each individual is selected is equal to its
In adaptive value and entire population individual fitness and ratio.If some individual i, just when for fi, then its probability selected
It indicates are as follows:
Step 4-2-5: crossover operation.Introducing three exchange crossover operators keeps the maximum value of each task engine air route length minimum
Change, to realize the air route length of balanced each task engine.Three exchange crossover operators occur 1 using 3 parent chromosome hybridization
Filial generation retains and is exaggerated the characteristic that parent excellent genes are inherited in filial generation this method increase the quantity of parent chromosome, accelerates
The search speed of algorithm.In case where 6 task points, to describe the crossover process of three exchange crossover operators in detail:
The non symmetrical distance of 6 task points is as follows, distributes the method still apart from symmetrical task for task point
It is applicable in.
The distance between 6 task points
| 0 | 1 | 2 | 3 | 4 | 5 | 6 | |
| 0 | ∞ | 1 | 7 | 6 | 4 | 2 | 5 |
| 1 | 2 | 0 | 6 | 5 | 5 | 4 | 4 |
| 2 | 1 | 7 | 0 | 3 | 1 | 2 | 3 |
| 3 | 4 | 3 | 5 | 0 | 5 | 6 | 7 |
| 4 | 4 | 1 | 5 | 3 | 0 | 2 | 7 |
| 5 | 5 | 3 | 4 | 2 | 1 | 0 | 7 |
| 6 | 3 | 2 | 7 | 6 | 5 | 4 | 0 |
It randomly chooses three individuals and carries out three exchange intersections:
α=1 257436
β=2 734165
γ=3 645721
The distance in the air route α and be 23, the maximum distance in individual task machine is 14；
The distance in the air route β and be 27, the maximum distance in individual task machine is 20；
The distance in the air route γ and be 34, the maximum distance in individual task machine is 20.Randomly choosing starting point is 1, makes 1 one-tenth
For first position of three male parents.
α=1 257436
β=1 652734
γ=1 364572
L=1 * * * * * *
Because of d (1,2)=6, d (1,6)=4, d (1,3)=5, so d (1,2) > d (1,3) > d (1,6), so
α=1 625743
β=1 652734
γ=1 645723
6 * * * * * of L=1
By parity of reasoning, and the offspring finally obtained is:
L=1 652743
The distance of L and be 20, the maximum distance in individual task machine is 13, so the total distance of L and and individual task
The maximum value in machine air route is respectively less than α, β and γ.
Step 4-2-6: mutation operation.The chromosomal variation of class MTSP multi-robotic task distribution system uses in the present invention
Exchange mutation exchanges the gene on two random sites.
Step 4-2-7: matrix decoding.In the operation situation of 6 task points, item chromosome therein is:
α=1 257436
Then the air route of 2 frame task engines is respectively:
0-1-2-5 0-4-3-6
I.e. the air route of task engine A is 0-1-2-5；The air route of task engine B is 0-4-3-6.
The task point Distance matrix D of a 6*6 is established, as follows:
According to the access path for having each task engine, the reachability matrix X of available task engine A, B1、X2It is as follows:
Reachability matrix X { the X of each task engine1,X2With each task point constitute Distance matrix D carry out dot product respectively, obtain
To matrix be each task engine flight path distance matrixFinally obtain the flight road of each task engine
Line length.For example, by the distance matrix that can obtain the walked distance of task engine A above are as follows:
So the air route length summation S of the task engine A after being distributed in the hope of algorithmA=1+6+2=9；Task engine B's
Air route length summation SB=4+7+3=14；
Step 4-3: after setting task engine formation has accessed all task points, the last one task point is rested on, without returning
Starting point is returned, task engine ideal air route l as shown in Figure 3 is obtained0；
Step 5: obtaining the position of energy recharge point P, the multiple no-manned plane task point under a kind of power consumption constraint as shown in Figure 4
With air route schematic diagram；
Step 5-1: the host computer combination constraint condition (power consumption constraint of task engine: SA=SB≤S0And the energy is mended
To point positioned at the midpoint constraint condition of the line midpoint of two frame task engine A, B:
SA`P=SB`P) task engine is simulated in ideal air route l0Upper flight obtains first energy recharge point P0Position, as shown in Figure 4
Energy recharge point P；
The point of step 5-2: two task engines A, B and server C are point P, and disregard and converge spent by the moment
Time, the flight route of server C are that each energy recharge point is successively gone to from starting point.Also, for iteration each time, system
Judge the air route length of individual task machine whether to be up to air route limiting value that (single air route maximum value is S0), when a certain task
When the air route of machine will reach the limit values, energy recharge point is gone in task engine.Wherein, the flying speed of task engine A, B is 1M (1M
As one times of velocity of sound), the flying speed of server C is 3M；Iteration cycle is 200ms；A, the flight route length difference of B, C
Are as follows: SA、SB、SC；A, the linear distance between B is SAB；A', B' are that task engine deviates ideal air route l0Deviation point；Task engine one
The secondary maximum flight route length by power consumption constraint is S0；SA`B`For the linear distance of deviation point A' and B'；SA`PFor deviation point A'
With the linear distance of point P；SB`PFor the linear distance of deviation point B' and point P；The power consumption constraint of task engine, with
And the midpoint constraint condition of the position of energy recharge point are as follows:
By the position of certain energy recharge point P during the available task engine formation execution task of constraints above condition
It sets；
Step 6: unmanned plane is formed into columns after first energy recharge point of arrival, and system makees task minute to remaining task point again
Match, seeks next energy recharge point in combination with constraint condition.That is circular flow step 2, step 3, step 4, step always
5, until the distribution of all task points is completed, the final air route for determining multiple no-manned plane and being distributed based on the task of limited energy；
Step 7: after all task points are assigned, console issues control command, and unmanned plane, which is formed into columns, receives order, from
O point sets out, and goes to task point to execute task along specified air route.
To sum up, the present invention devises the method for allocating tasks of a kind of optimization and balanced unmanned plane formation air route, obtains nobody
The planning air route of machine；Simultaneously, it is contemplated that the energy recharge problem in practical application, while task engine accesses each task point,
Energy recharge point solution procedure is set.This method has the advantages that practical field suitable for other aerial multi-task plannings
It closes；Multiple no-manned plane task distribution application field is expanded, provides good distribution side for air transport and the planning of aerial battlefield
Method.
Claims (5)
1. the multiple no-manned plane method for allocating tasks under a kind of power consumption constraint, it is characterised in that: in the formation of given unmanned plane, each task
It puts and under the consumption constraint of task function, air route is minimized as Target Assignment task, to make task simultaneously using balance optimizing air route
The energy that machine expends is few as far as possible, the air route point of task engine is determined on the basis of the ideal air route of task engine, and by server
Energy recharge is provided in point, the complete air route of unmanned plane is finally obtained, completes the distribution of all task points；Specifically include as
Lower step:
Step 1: the initialization of task distribution system；It mainly include that host computer initializes, ground control cabinet initializes, clock is initial
Change, unmanned plane is formed into columns and initialized；The unmanned plane formation has at least three frame unmanned planes, respectively task engine A, task engine B, to appoint
The server C of the business machine offer energy；There is multiple tasks point in setting front, and task point number is more than task engine number, unmanned plane
It forms into columns before receiving out and saying the word, awaits orders positioned at starting point O；
Step 2: the communication established between ground control cabinet, position location satellite and unmanned plane formation: when executing front task, passing through
Satellite positioning tech obtains the location information of each task point, unmanned plane in real time；
Step 3: the location information of each task point of unmanned plane and front is returned to ground control cabinet, ground control in real time by position location satellite
Platform real-time reception processed, analysis simultaneously handle these location informations；
Step 4: host computer introduces three exchange crossover operators for each task point and distributes to each task by Revised genetic algorithum
Machine obtains ideal air route l0, i.e., task engine is not by the air route of power consumption constraint；Establishing task point distribution model is class MTSP model,
As class MTSP model of the travelling salesman without return；
Step 5: obtain the position of energy recharge point P: host computer combination constraint condition simulates task engine in preliminary ideal air route l0
Upper flight obtains first energy recharge point P0Position；
The constraint condition includes the power consumption constraint and midpoint constraint condition of task engine；
The power consumption constraint of task engine are as follows: SA=SB≤S0And energy recharge point is located in the line of two frame task engine A, B
Midpoint constraint condition at point are as follows:SA`P=SB`P；
Wherein, the flight route length of A are as follows: SA；The flight route length of B are as follows: SB；A, the linear distance between B is SAB；Task
Machine is once S by the maximum flight route length of power consumption constraint0；SA`B`For the linear distance of deviation point A' and B'；SA`PTo deviate
The linear distance of point A' and point P；SB`PFor the linear distance of deviation point B' and point P；
Step 6: setting unmanned plane, which is formed into columns, reaches first energy recharge point P0Afterwards, task distribution system is to remaining task point weight
The distribution of new work task, while seeking next energy recharge point, i.e. circular flow step 2, step 3, step 4, step 5 always,
Until completing the distribution of all task points；
Step 7: after all task points are assigned, ground control cabinet issues control command, and unmanned plane, which is formed into columns, receives order, from
O point sets out, and goes to each task point to execute distributing for task along specified planning air route.
2. the multiple no-manned plane method for allocating tasks under power consumption constraint according to claim 1, it is characterised in that: task engine A
It is isomorphism unmanned plane with task engine B, server C and task engine are isomery unmanned plane, and the limitation of server C energy consumption has been enough to ensure that
At all tasks, task engine A and task engine B limited energy need to carry out energy supplement in flight course.
3. the multiple no-manned plane method for allocating tasks under power consumption constraint according to claim 1, it is characterised in that: the energy
Supply point P is the point of task engine depleted of energy and the point of server C and task engine A, task engine B；And disregard remittance
Time spent by the conjunction, the flight route of server C are that each energy recharge point is successively gone to from starting point；It changes at each
For judging whether each task engine needs to carry out energy recharge in the period, energy recharge point P is located at a certain moment task engine A and B
Line midpoint, when a team form into columns reach energy recharge point P when, task engine A, task engine B the energy exhausted, server C
Flight path be by starting point O to the straight line path of energy recharge point P.
4. the multiple no-manned plane method for allocating tasks under power consumption constraint according to claim 1, it is characterised in that: individual task
When the air route length of machine is up to air route limiting value, which goes to energy recharge point.
5. the multiple no-manned plane method for allocating tasks under power consumption constraint according to claim 1, it is characterised in that: in step 4
After setting task engine formation has accessed all task points, the last one task point is rested on without returning to starting point and obtains task
Machine it is preliminary not by the air route l of power consumption constraint0, namely ideal air route l0。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610341065.1A CN106020230B (en) | 2016-05-20 | 2016-05-20 | A kind of multiple no-manned plane method for allocating tasks under power consumption constraint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610341065.1A CN106020230B (en) | 2016-05-20 | 2016-05-20 | A kind of multiple no-manned plane method for allocating tasks under power consumption constraint |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106020230A CN106020230A (en) | 2016-10-12 |
| CN106020230B true CN106020230B (en) | 2019-01-15 |
Family
ID=57095529
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610341065.1A Active CN106020230B (en) | 2016-05-20 | 2016-05-20 | A kind of multiple no-manned plane method for allocating tasks under power consumption constraint |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106020230B (en) |
Families Citing this family (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106527446B (en) | 2016-12-02 | 2020-11-13 | 北京小米移动软件有限公司 | Control method and device of sweeping robot |
| CN106600147B (en) * | 2016-12-15 | 2017-10-13 | 合肥工业大学 | Towards the multiple no-manned plane method for allocating tasks and device of decomposable asymmetric choice net task |
| CN107037826B (en) * | 2017-04-14 | 2019-11-12 | 合肥工业大学 | Unmanned plane detection mission distribution method and device |
| CN107037829A (en) * | 2017-05-09 | 2017-08-11 | 广东容祺智能科技有限公司 | A kind of cluster unmanned plane route planning method |
| CN107203221B (en) * | 2017-06-01 | 2020-09-01 | 合肥工业大学 | Online information distribution method and device in unmanned aerial vehicle and manned machine mixed formation |
| CN107239078B (en) * | 2017-06-26 | 2020-03-27 | 中国人民解放军国防科学技术大学 | Unmanned aerial vehicle base station site selection and patrol path optimization method and device |
| CN107563653B (en) * | 2017-09-08 | 2020-04-28 | 中国人民解放军国防科技大学 | Multi-robot full-coverage task allocation method |
| CN108416488B (en) * | 2017-12-21 | 2022-05-03 | 中南大学 | Dynamic task-oriented multi-intelligent-robot task allocation method |
| CN108594813B (en) * | 2018-04-18 | 2021-04-09 | 苏州大学张家港工业技术研究院 | Task allocation method for multiple cleaning robots in large-scale indoor environment |
| CN108762294B (en) * | 2018-04-28 | 2023-12-22 | 广州亿航智能技术有限公司 | Unmanned aerial vehicle path planning method and management system for aerial photography |
| CN108871335B (en) * | 2018-06-14 | 2020-10-23 | 华南农业大学 | Unmanned aerial vehicle spraying operation air route and load planning method |
| CN108759824A (en) * | 2018-07-24 | 2018-11-06 | 顺丰科技有限公司 | High accuracy positioning navigation system and method |
| CN109270934B (en) * | 2018-11-01 | 2020-05-12 | 中国科学技术大学 | Multi-robot formation cruising method based on pilot switching |
| CN109922137B (en) * | 2019-01-28 | 2021-06-25 | 中国人民解放军国防科技大学 | Unmanned aerial vehicle assisted calculation migration method |
| CN109784585B (en) * | 2019-03-07 | 2020-10-16 | 广东华中科技大学工业技术研究院 | Hybrid deployment and scheduling method for unmanned aerial vehicle unmanned ship |
| CN110069074B (en) * | 2019-04-17 | 2020-08-18 | 南京航空航天大学 | Unmanned aerial vehicle collaborative flight path planning method based on multi-target three-point positioning |
| CN109976164B (en) * | 2019-04-25 | 2022-11-29 | 南开大学 | Energy optimization visual coverage trajectory planning method for multi-rotor unmanned aerial vehicle |
| CN110233658B (en) * | 2019-06-02 | 2021-05-18 | 西安电子科技大学 | Unmanned aerial vehicle energy supply scheduling method based on idle unmanned aerial vehicle replacement |
| CN110889625B (en) * | 2019-11-25 | 2022-05-24 | 航天时代飞鸿技术有限公司 | Task planning method for swarm unmanned aerial vehicle system |
| CN111240356B (en) * | 2020-01-14 | 2022-09-02 | 西北工业大学 | Unmanned aerial vehicle cluster convergence method based on deep reinforcement learning |
| CN113268088B (en) * | 2021-06-10 | 2021-12-28 | 中国电子科技集团公司第二十八研究所 | Unmanned aerial vehicle task allocation method based on minimum cost and maximum flow |
| CN113788012B (en) * | 2021-10-13 | 2023-01-06 | 北京三快在线科技有限公司 | Unmanned vehicle parking method, device, equipment and storage medium |
| CN114489135B (en) * | 2022-01-28 | 2023-10-13 | 北京航空航天大学东营研究院 | Multitasking route design method |
| CN114200963B (en) * | 2022-02-17 | 2022-05-10 | 佛山科学技术学院 | Unmanned aerial vehicle autonomous mission planning method and device under dynamic environment and storage medium |
| CN115454147B (en) * | 2022-10-29 | 2023-03-14 | 中国电子科技集团公司第十五研究所 | Unmanned aerial vehicle cluster cooperative task allocation method based on dynamic optimization |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6055523A (en) * | 1997-07-15 | 2000-04-25 | The United States Of America As Represented By The Secretary Of The Army | Method and apparatus for multi-sensor, multi-target tracking using a genetic algorithm |
| CN101286071A (en) * | 2008-04-24 | 2008-10-15 | 北京航空航天大学 | Multiple no-manned plane three-dimensional formation reconfiguration method based on particle swarm optimization and genetic algorithm |
| CN104216382A (en) * | 2014-09-19 | 2014-12-17 | 北京航天长征飞行器研究所 | Spatial small aircraft formation flying control system |
| CN104359473A (en) * | 2014-10-24 | 2015-02-18 | 南京航空航天大学 | Collaborative flight path intelligent planning method for formation flying of unmanned planes under dynamic environment |
| CN104407619A (en) * | 2014-11-05 | 2015-03-11 | 沈阳航空航天大学 | Method enabling multiple unmanned aerial vehicles to reach multiple targets simultaneously under uncertain environments |
| CN104700165A (en) * | 2015-03-27 | 2015-06-10 | 合肥工业大学 | Multi-UAV (unmanned aerial vehicle) helicopter and warship cooperating path planning method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8447419B1 (en) * | 2012-05-02 | 2013-05-21 | Ether Dynamics Corporation | Pseudo-genetic meta-knowledge artificial intelligence systems and methods |
-
2016
- 2016-05-20 CN CN201610341065.1A patent/CN106020230B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6055523A (en) * | 1997-07-15 | 2000-04-25 | The United States Of America As Represented By The Secretary Of The Army | Method and apparatus for multi-sensor, multi-target tracking using a genetic algorithm |
| CN101286071A (en) * | 2008-04-24 | 2008-10-15 | 北京航空航天大学 | Multiple no-manned plane three-dimensional formation reconfiguration method based on particle swarm optimization and genetic algorithm |
| CN104216382A (en) * | 2014-09-19 | 2014-12-17 | 北京航天长征飞行器研究所 | Spatial small aircraft formation flying control system |
| CN104359473A (en) * | 2014-10-24 | 2015-02-18 | 南京航空航天大学 | Collaborative flight path intelligent planning method for formation flying of unmanned planes under dynamic environment |
| CN104407619A (en) * | 2014-11-05 | 2015-03-11 | 沈阳航空航天大学 | Method enabling multiple unmanned aerial vehicles to reach multiple targets simultaneously under uncertain environments |
| CN104700165A (en) * | 2015-03-27 | 2015-06-10 | 合肥工业大学 | Multi-UAV (unmanned aerial vehicle) helicopter and warship cooperating path planning method |
Non-Patent Citations (3)
| Title |
|---|
| path planning for multiple unmanned aerial vehicles;Halit Ergezer，Kemal Leblebicioglu;《Signal Processing and Communications Applications Conference（SIU）》;20120530;第1-4页 |
| 一种新的求解旅行商问题的混合遗传算法;陈乔礼，吴怀宇，刘亮;《武汉科技大学学报（自然科学版）》;20070228;第30卷(第1期);第74-78页 |
| 无人机编队信息传输能耗平衡协同设计;方堃，茹乐，于强强，于云龙;《计算机仿真》;20160430;第33卷(第4期);第63-66页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106020230A (en) | 2016-10-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106020230B (en) | A kind of multiple no-manned plane method for allocating tasks under power consumption constraint | |
| CN107238388B (en) | Multiple no-manned plane task is distributed and trajectory planning combined optimization method and device | |
| Liu et al. | Energy-efficient UAV crowdsensing with multiple charging stations by deep learning | |
| CN105717926A (en) | Mobile robot traveling salesman optimization method based on improved ant colony algorithm | |
| Liang et al. | Applying genetic algorithm and ant colony optimization algorithm into marine investigation path planning model | |
| CN107103164B (en) | Distribution method and device for unmanned aerial vehicle to execute multiple tasks | |
| CN103279793A (en) | Task allocation method for formation of unmanned aerial vehicles in certain environment | |
| CN105225003A (en) | A kind of cuckoo searching algorithm solves the method for UAV multitask investigation decision problem | |
| CN105045095A (en) | Multi-unmanned-plane emergency rescue task distribution method | |
| CN112013829A (en) | Multi-UAV/UGV (unmanned aerial vehicle/user generated Unit) cooperative long-term operation path planning method based on multi-objective optimization | |
| CN111121784B (en) | Unmanned reconnaissance aircraft route planning method | |
| CN107330588A (en) | A kind of mission planning method of many base isomery unmanned plane coordinated investigations | |
| CN109885082A (en) | The method that a kind of lower unmanned aerial vehicle flight path of task based access control driving is planned | |
| CN106228265A (en) | Based on Modified particle swarm optimization always drag phase transport project dispatching algorithm | |
| Tang et al. | Spatial task assignment based on information gain in crowdsourcing | |
| CN110530373A (en) | A kind of robot path planning method, controller and system | |
| Huang et al. | TMA-DPSO: Towards efficient multi-task allocation with time constraints for next generation multiple access | |
| Li et al. | Fedvanet: Efficient federated learning with non-iid data for vehicular ad hoc networks | |
| CN111157002A (en) | Aircraft 3D path planning method based on multi-agent evolutionary algorithm | |
| Su et al. | An improved adaptive differential evolution algorithm for single unmanned aerial vehicle multitasking | |
| Yu et al. | Balanced multi-region coverage path planning for unmanned aerial vehicles | |
| Zhang et al. | Rlcharge: Imitative multi-agent spatiotemporal reinforcement learning for electric vehicle charging station recommendation | |
| Li et al. | CAAS: a novel collective action-based ant system algorithm for solving TSP problem | |
| CN113805609A (en) | Unmanned aerial vehicle group target searching method based on chaos lost pigeon group optimization mechanism | |
| CN109840551A (en) | A method of the optimization random forest parameter for machine learning model training |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |