CN113190038A - Method for distributing reconnaissance tasks in unmanned aerial vehicle cluster air area - Google Patents
Method for distributing reconnaissance tasks in unmanned aerial vehicle cluster air area Download PDFInfo
- Publication number
- CN113190038A CN113190038A CN202110420280.1A CN202110420280A CN113190038A CN 113190038 A CN113190038 A CN 113190038A CN 202110420280 A CN202110420280 A CN 202110420280A CN 113190038 A CN113190038 A CN 113190038A
- Authority
- CN
- China
- Prior art keywords
- unmanned aerial
- reconnaissance
- task
- aerial vehicle
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
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
Abstract
The invention discloses a method for distributing reconnaissance tasks in an unmanned aerial vehicle cluster aerial region, which comprises the following steps: establishing a profit model, initializing a bundle set, performing task allocation and resolving conflict allocation. The method for distributing the reconnaissance tasks in the unmanned aerial vehicle cluster air area saves the whole task time and the recovery consumption.
Description
Technical Field
The invention relates to a task allocation method, in particular to an allocation method of a reconnaissance task in an unmanned aerial vehicle cluster aerial region, and belongs to the field of unmanned aerial vehicle control.
Background
The distribution problem of the reconnaissance tasks in the aerial region of the unmanned aerial vehicle cluster is essentially that regions with different reconnaissance values are reasonably distributed to multiple types of unmanned aerial vehicles carrying sensors with different performances, so that the efficiency and the benefit of completing the reconnaissance tasks are the largest, and the cost is the smallest.
In the prior art, a CBAA algorithm (consensus based allocation algorithm) with a single task allocation problem is improved to a CBBA algorithm (consensus based allocated bundle algorithm) suitable for a multi-task allocation problem, and the methods can better react to task situations of each node of a cluster based on a distributed auction algorithm and a consistency idea. However, these methods are only suitable for the task allocation process, but are not suitable for the need of recovering the unmanned aerial vehicle in the actual reconnaissance task.
Moreover, the task allocation problem of the multi-unmanned aerial vehicle applying the CBBA algorithm focuses on solving the mathematical scheduling problem of task allocation, when the target of 'low and slow' is faced in the actual reconnaissance task, the target has probability, the benefits of the reconnaissance tasks allocated to different areas are not accurate values, and the recovery of the unmanned aerial vehicle cluster is also important except the completion of the task allocation in the actual reconnaissance task. Particularly, for widely dispersed 'low-slow small' targets, a large number of unmanned aerial vehicles need to consume a large amount of energy to complete the return flight requirement after the reconnaissance mission is finished, and the start of the next reconnaissance mission is indirectly influenced, so that the reconnaissance efficiency and the benefit are greatly reduced.
On the other hand, in consideration of constraints of two influences of the total flight distance of the unmanned aerial vehicle cluster and the task distance reached by each unmanned aerial vehicle, an optimal task decision method is adopted in the prior art, but in the optimal task decision method, the effect that each unmanned aerial vehicle tends to execute a task closer to the takeoff position of the unmanned aerial vehicle is not obvious. Particularly, when the target is low, slow and small, the detection area is scattered and uncertain due to the wide target dispersion range, and the traditional optimal task decision method is not suitable.
Therefore, it is necessary to provide a method for allocating the unmanned aerial vehicle cluster aerial area reconnaissance missions to solve the above problems.
Disclosure of Invention
In order to overcome the problems, the inventor of the present invention has made an intensive study and provides a method for allocating a reconnaissance mission in an unmanned aerial vehicle cluster air region, wherein the method includes the following steps:
s1, establishing a profit model;
s2, initializing a beam set;
s3, performing primary task allocation according to the income model;
and S4, resolving conflict allocation in the preliminary task allocation to obtain final task allocation.
Further, in step S1, the benefit model is used to describe the scout benefits of the drone cluster, including establishing a scout matrix and determining a benefit index,
the reconnaissance matrix is used for describing the performability of different unmanned aerial vehicles to different reconnaissance areas and is represented as NuLine NtA two-dimensional matrix of columns, wherein NuRepresenting the total number of unmanned aerial vehicles, NtRepresenting the total number of scout areas.
The revenue indicator may be expressed as
Wherein x isijIndicates whether the unmanned aerial vehicle i performs reconnaissance on the reconnaissance area j, xijDenotes that drone i performs reconnaissance on reconnaissance area j, x ij0 means that drone i does not perform reconnaissance on reconnaissance area j;
Piindicating the path sequence during the reconnaissance of drone i, cijAs a function of the benefit.
The beam set includes:
the task bundle set is used for describing a task sequence of the unmanned aerial vehicle;
the task time sequence set is used for describing the task time sequence of the unmanned aerial vehicle;
the execution time set is used for describing the time for the unmanned aerial vehicle to scout different tasks according to the task time sequence set;
a winner set used for describing whether the unmanned aerial vehicle successfully bids on different reconnaissance areas;
a winner offer set used for representing the maximum output value of each unmanned aerial vehicle when the unmanned aerial vehicle auctions on the reconnaissance area at the current moment;
a set of timestamps to represent a last time information interaction time between drone i and its neighboring drone.
In step S3, the post-task allocation benefits are made more accurate by adding a probabilistic constraint on the existence of the target.
Further, in step S3, the following substeps are included:
s31, obtaining marginal benefits of the unmanned aerial vehicle i to the reconnaissance area j;
s32, judging whether the unmanned aerial vehicle i can obtain the task of the reconnaissance area j;
s33, determining the path sequence P of the scout area jiThe insertion position in (1);
and S34, updating the beam set of the unmanned aerial vehicle i.
Specifically, in step S31, the marginal profit cij(Pi) Comprises the following steps:
wherein, tijIndicating the order of unmanned plane i along path PiTime spent to scout area j;
λ is the elapsed time tijThe affected income factor parameters can be set according to actual needs;
Valjvalue coefficient inherent to the scout area j, 0 < Valj＜1；
ZijAnd the completeness of the reconnaissance area j by the unmanned plane i in the reconnaissance matrix is represented.
In step S33, the scout area j is inserted into different positions in the route order, the profit is calculated, and the position at the time of the maximum profit is taken as the route order P of the scout area jiThe insertion position in (1) is represented as:
wherein the content of the first and second substances,
adding the scout region j to the path sequence P when the representation yield is maximumiN denotes the path order PiThe element position in (1).
In step S34, steps S31 to S33 are repeated, the scout area allocation and the route order determination are performed for each drone, and the task bundle set, the task time sequence set, the winner set, and the winner bid set in the bundle set are updated to complete the task allocation.
In step S4, the following substeps are included:
s41, all unmanned aerial vehicles communicate with each other, and the unmanned aerial vehicle corresponding to the maximum bid value in the winner bid set is selected to execute the task of the reconnaissance area j;
and S42, repeating the step S41 and completing the distribution of all conflict tasks.
The invention has the advantages that:
(1) according to the method for distributing the reconnaissance tasks in the aerial region of the unmanned aerial vehicle cluster, the tasks distributed by each unmanned aerial vehicle comprise the return flight requirement, so that the reconnaissance tasks are completed, the unmanned aerial vehicle cluster is recovered, the maximum benefit is still ensured, and the time and the recovery consumption are saved for other tasks;
(2) according to the method for distributing the reconnaissance tasks in the air region of the unmanned aerial vehicle cluster, provided by the invention, the probability mujThe profit value is corrected well, so that a pure mathematical algorithm is avoided from being separated from the reality, and a better evaluation value is provided for the profit of completing each task;
(3) according to the distribution method of the unmanned aerial vehicle cluster aerial region reconnaissance tasks, information is communicated and shared, and multiple unmanned aerial vehicles in the same task are prevented from being repeatedly executed;
(4) according to the distribution method of the unmanned aerial vehicle cluster aerial area reconnaissance tasks, the communication condition is recorded in a time stamp mode, communication burden caused by repeated communication is avoided, and efficiency is prevented from being reduced.
Drawings
Fig. 1 is a schematic flow chart illustrating a method for allocating a mission of an unmanned aerial vehicle cluster in an air region according to a preferred embodiment of the present invention;
FIG. 2 is a graph showing the results in example 1;
FIG. 3 is a graph showing the results in comparative example 1;
FIG. 4 is a graph showing the results in example 2;
fig. 5 shows a graph of the results in comparative example 2.
Detailed Description
The invention is explained in more detail below with reference to the figures and examples. The features and advantages of the present invention will become more apparent from the description.
The word "exemplary" is used exclusively herein to mean "serving as an example, embodiment, or illustration. Any embodiment described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments. While the various aspects of the embodiments are presented in drawings, the drawings are not necessarily drawn to scale unless specifically indicated.
The invention provides a method for allocating reconnaissance tasks in an unmanned aerial vehicle cluster aerial region, and particularly aims at low-speed and small-size targets, and by adding probability constraint to the existence of the targets, the benefits after the tasks are allocated are more accurate.
The low-slow small target refers to an aircraft with the flying height of below 1000 meters, the flying speed per hour of less than 200 kilometers and the radar reflection area of less than 2 square meters, has the problems of difficult discovery, difficult capture, difficult disposal, difficult coping and the like, and greatly threatens the air safety.
Further, in the present invention, it is preferable that the return recovery of the unmanned aerial vehicle cluster is taken as an extra cost to be included in the task allocation profit, so that the task sequence end point allocated by each unmanned aerial vehicle is entirely close to the departure point, so that the unmanned aerial vehicle cluster can rapidly enter the next reconnaissance task, and the energy consumption of the unmanned aerial vehicle cluster is reduced.
Specifically, the method comprises the following steps:
s1, establishing a profit model;
s2, initializing a beam set;
s3, performing primary task allocation according to the income model;
and S4, resolving conflict allocation in the preliminary task allocation to obtain final task allocation.
In step S1, the benefit model is used to describe the scout benefits of the drone cluster, including establishing a scout matrix and determining a benefit index.
The reconnaissance matrix is used for describing the completeness of different unmanned aerial vehicles to different reconnaissance areas.
In the present invention, the scout matrix is represented by one NuLine NtA two-dimensional matrix of columns, wherein NuIndicate the total number of unmanned aerial vehicles, different unmanned aerial vehicles are denoted as i, then i equals 1,2,3, …, Nu；NtDenotes the total number of scout regions, and j denotes a different scout region, j is 1,2,3, …, Nt。
In one embodiment, any one element Z in the scout matrixijCan be expressed as:
in a preferred embodiment, whether drone i can complete the reconnaissance mission of reconnaissance area j is determined by the reconnaissance capability matrix of drone i
With the object type matrix in the scout area j
After multiplication, it is determined that:
reconnaissance capability matrix of unmanned aerial vehicle i
Is shown as
Wherein the subscripts 1,2, …, m denote the kind of object to be spy,
respectively representing the identification capability of different sensitive elements carried in the unmanned aerial vehicle i on different targets to be detected, for example 4 targets to be detected in a detection area,
the method comprises the steps that sensitive elements carried in an unmanned aerial vehicle i cannot identify 1 st type and 4 th types of targets to be detected, the identification capacity of the 2 nd type of targets to be detected is 8 at most, and the identification capacity of the 3 rd type of targets to be detected is 4 at most;
object type matrix in scout region j
Is shown as
Wherein the content of the first and second substances,
respectively representing the number of different objects to be scout in the scout area j, e.g.
The number of 1 st objects to be detected in the detection area j is 0, and the number of 2 nd, 3 rd and 4 th objects to be detected is 1.
The profit index is a commonly used research method in the task allocation process, the optimal strategy can be selected from numerous combinations through the profit index, and in the invention, the profit index can be expressed as
Wherein x isijIndicates whether drone i performs a reconnaissance on reconnaissance area j, preferably xijDenotes that drone i performs reconnaissance on reconnaissance area j, xij0 means that drone i does not perform reconnaissance on reconnaissance area j,
Pirepresenting the path sequence of the unmanned plane during the reconnaissance, wherein the path sequence can be represented by a one-dimensional array, namely Pi＝{Pi1,Pi2,…,Pik,…},PikScout zones representing the kth secondary scout of drone i, e.g. drone i scout 4 areas to be scout in total, Pi＝{Pi1,Pi2,Pi3,Pi4And {1,5,7,4 denotes that the unmanned aerial vehicle i sequentially scouts the 1 st, 5 th, 7 th and 4 th scout areas;
cijas a function of gain, from xiAnd PiJoint decisions, i.e., joint decisions by path length and elapsed time to perform all tasks.
In the invention, the optimal task allocation result is obtained by maximizing the profit index, namely, the reconnaissance area to be executed by the unmanned aerial vehicle is determined, and the path sequence when the unmanned aerial vehicle reconnaissance is determined.
Because the different unmanned aerial vehicles carry different sensitive elements, the types of targets which can be detected by different unmanned aerial vehicles are different, the number of regions which can be detected by different unmanned aerial vehicles is different at most, and a large number of intermediate results can be generated in the task distribution process, in the invention, the intermediate results and the final results are recorded through the cluster.
In step S2, a beam set is provided in each drone, the beam set including:
set of task bundles, denoted as
Sequence of tasks to describe drone i, where LiRepresenting the number of scout missions (i.e. scout areas) that the drone i is capable of performing at most, the elements in the set of task bundles being boolean values, e.g. bi11 denotes that drone i performs the mission of the 1 st reconnaissance area, bi3No. 0 indicates that drone i does not perform the 3 rd scout mission;
a time-ordered set of tasks, denoted as
To describe the task timing of drone i, i.e. to perform a set of task bundles BiThe order of the tasks in (1);
set of execution times, denoted as
Time-ordered set according to task P to describe drone iiGo to the time for scouting different tasks;
set of winners, denoted as
Elements in the set are used for describing whether the unmanned aerial vehicle i successfully bids in different reconnaissance areas, if the unmanned aerial vehicle i bids the highest value in the reconnaissance area j at the current moment and becomes a winner, the unmanned aerial vehicle i bids the highest value in the reconnaissance area j, the unmanned aerial vehicle i bids the highest value, the unmanned aerial vehicle i successfully bids the highest value, and zijElse, bid fails, zij＝0；
Set of winner bids, represented as
The elements in the set are used for representing the maximum output value of each unmanned aerial vehicle when the unmanned aerial vehicle auctions on the scouting area j at the current moment, and if no unmanned aerial vehicle auctions on the scouting area j, yj＝0；
Set of time stamps, denoted as
The elements in the set are used to represent the last time of information interaction between drone i and its neighboring drone, which is the exclusion of this nullAll unmanned aerial vehicles except man-machine.
Before task allocation, a bundle set is initialized, and all sets are set as empty sets.
In step S3, the following substeps are included:
s31, obtaining marginal benefits of the unmanned aerial vehicle i to the reconnaissance area j;
s32, judging whether the unmanned aerial vehicle i can obtain the task of the reconnaissance area j;
s33, determining the path sequence P of the scout area jiThe insertion position in (1);
and S34, updating the beam set of the unmanned aerial vehicle i.
In step S31, the marginal benefit is that the drone i follows the path sequence piThe benefit of reconnaissance area j can be expressed as:
wherein the content of the first and second substances,
indicating that drone i executes in path order PiTotal revenue obtained from all tasks allocated;
|Pii denotes the modulus of the task path, i.e. in path order PiThe path length over which the scout is performed;
operation of
It is shown that the insertion action is performed,
to add scout region j to path order PiIn the n-th position of the optical fiber,
indicates that the scout region j is inserted into the path sequence PiThe last position of (a);
By taking the starting point of unmanned aerial vehicle i as a task and inserting the starting point of unmanned aerial vehicle i in the path sequence PiThe last position of (a) is constrained by the last need to return to the starting point when calculating the benefit of drone i to perform the task.
Further, an unmanned aerial vehicle i starting point is used as a reconnaissance area and inserted into the path sequence PiAfter the last position, drone i executes the order of path PiThe total revenue obtained for all tasks allocated can be expressed as:
wherein, mujThe probability of the target to be detected in the detection area j is represented, namely the existence of the flight target with low speed and small speed in the actual detection task is considered to be uncertain, and the probability mu is needed for the calculated fixed income valuejCorrecting;
τinrepresenting that when the unmanned aerial vehicle i carries out the path consumption of the current task and the next task, namely the path consumption is used as a constraint for influencing the income;
further, the air conditioner is provided with a fan,
τcurrepresenting the moment at which drone i executes reconnaissance area j,
indicating the sequence P from the departure point to the path of the unmanned plane iiTime consuming in the j-th scout area.
According to the formulas (two) - (four), the unmanned aerial vehicle i follows the path sequence PiGain of scouting jth scout areaIs determined by the marginal benefit and path consumption constraint for completing the task and the existence probability correction, the marginal benefit cij(Pi) Can be simplified as follows:
wherein, tijIndicating the order of unmanned plane i along path PiTime spent to scout area j;
λ is the elapsed time tijThe affected income factor parameters can be set according to actual needs;
Valjvalue coefficients inherent to the scout area j, typically 0 < ValjAnd < 1, further, different scout areas j have different value coefficients, the larger the target hazard in the scout area j is, the higher the value coefficient of the scout area j is, and the specific coefficient value can be set by a person skilled in the art according to actual requirements.
In the invention, the takeoff position of the unmanned aerial vehicle is fixed at the end point of the task path, so the recovery consumption of the unmanned aerial vehicle cluster is also considered, the return voyage of the unmanned aerial vehicle cluster also belongs to the task, the unmanned aerial vehicle cluster is more suitable for actual reconnaissance task distribution, and the task path of each unmanned aerial vehicle approaches to form a closed loop, which is in line with the aim of facilitating recovery.
In step S32, the drone i determines whether the reconnaissance area j exceeds its flight range, which is determined by the remaining power or fuel of the drone, the flight speed of the drone, and the like, which are not described in detail herein.
Further, if the flight range is exceeded, the unmanned aerial vehicle i cannot obtain the mission of the reconnaissance area j, and if the flight range is within the flight range, the unmanned aerial vehicle i can obtain the mission of the reconnaissance area j.
Repeating the process, judging all the missions of the reconnaissance area of the unmanned aerial vehicle i, and updating the missions to a mission beam set.
In step S33, the scout area j is inserted into different positions in the route order, the profit is calculated, and the position at the time of the maximum profit is taken as the scout area jThe order of the paths P of the region of interest jiThe insertion position in (1) can be expressed as:
wherein the content of the first and second substances,
indicating the location of n where the benefit is greatest.
And further, updating the obtained path sequence to a task time sequence set, and further updating the obtained marginal profit to a winner bid set.
In step S34, steps S31 to S33 are repeated, reconnaissance area allocation and route order determination are performed for each drone, and the task bundle set, the task time sequence set, the winner set, and the winner bid set in the bundle set are updated to complete task allocation.
In step S3, since the task allocation is realized based on the maximum benefit of each drone to the reconnaissance area, which may cause the situation of task duplicate allocation, in step S4, the resolution of the duplicate allocation task is performed, which includes the following sub-steps:
s41, all unmanned aerial vehicles communicate with each other, and the unmanned aerial vehicle corresponding to the maximum bid value in the winner bid set is selected to execute the task of the reconnaissance area j;
and S42, repeating the step S41 and completing the distribution of all conflict tasks.
In step S41, the drones communicate with each other, share their respective winner and winner bid sets, and record the time of communication.
Further, if a plurality of unmanned aerial vehicles successfully bid on the same reconnaissance area j in the winner set, selecting the unmanned aerial vehicle corresponding to the maximum bid value in the winner set to execute the task of the reconnaissance area j;
specifically, by comparing the marginal benefit of the unmanned aerial vehicle i with the marginal benefits of other unmanned aerial vehicles, whether the unmanned aerial vehicle i obtains the mission of the reconnaissance area j is determined, which may be expressed as:
wherein, yijRepresents the highest of the marginal gains of other drones:
if the profit of the unmanned aerial vehicle i in the reconnaissance area j task is higher than the maximum value of the marginal profits of other unmanned aerial vehicles, the auction succeeds, and the reconnaissance area j task is obtained;
if the income of the j task of the unmanned aerial vehicle i in the reconnaissance area is lower than the maximum value of the marginal income of other unmanned aerial vehicles, the auction fails, and the j task of the reconnaissance area is not executed
Further, for other drones not obtaining the task of the reconnaissance area j, the task in the reconnaissance area j and the tasks after the reconnaissance area j are emptied, and step S3 is repeated to re-distribute the task after the reconnaissance area j, so as to update the drone bundle.
Further, after unmanned aerial vehicle i communicates with other unmanned aerial vehicles each other, the timestamp that unmanned aerial vehicle i restrainted in the set is updated, and the expression is:
wherein, gikRepresenting the communication connectivity situation of drone i and drone k, gikIf the communication between drone i and drone k is possible, the interaction time τ at this time is recorded as 1r，
If communication is not possible, the beam set of drone k is obtained from other drones, for example, drone m, and the latest interaction time when drone k information is obtained from drone m is recorded.
The mode of communication between the machine and update time stamp avoids many unmanned aerial vehicles to carry out close task respectively, avoids the repeated execution of task.
In step S42, after each drone updates the beam set, step S41 is repeated multiple times until all reconnaissance areas j complete a single allocation, completing conflict resolution.
Further preferably, after the conflict resolution is completed, the unmanned aerial vehicles in some reconnaissance areas may not execute, and at this time, the steps S3 to S4 are repeated, so that the reconnaissance areas can be allocated to the executable unmanned aerial vehicles, and thus conflict-free task allocation is completed.
Examples
Example 1
The 3 unmanned aerial vehicles carry the same sensitive elements with numbers of U1, U2 and U3, the flying speeds are all 20m/s, and the initial coordinates are (930,1010), (1690,1610) and (860,320) respectively.
Reconnaissance area coordinate and probability mujAs shown in table one.
Watch 1
| Scout area label | Reconnaissance area coordinates/km | Probability muj |
| T 1 | (120,1740) | 0.95 |
| T 2 | (1970,1460) | 0.95 |
| T 3 | (940,1590) | 0.95 |
| T 4 | (1980,1970) | 0.95 |
| T 5 | (1100,1320) | 0.90 |
| T 6 | (1410,1770) | 0.90 |
| T 7 | (170,1180) | 0.90 |
| T 8 | (950,1210) | 0.90 |
| T 9 | (1430,860) | 0.90 |
The distribution method comprises the following steps:
s1, establishing a profit model;
s2, initializing a beam set;
s3, performing primary task allocation according to the income model;
and S4, resolving conflict allocation in the preliminary task allocation to obtain final task allocation.
In step S1, creating the revenue model includes creating a scout matrix and determining a revenue indicator, the scout matrix being
The profit index is
In step S2, beam sets are set for U1, U2, and U3, respectively, and the beam set for the drone U1 includes a task beam set B1Task time ordered set P1Winner set z1Set of time stamps s1Winner offer set y1、y2、…、y9The beam set of the unmanned plane U2 comprises a task beam set B2Task time ordered set P2Winner set z2Set of time stamps s2Winner offer set y1、y2、…、y9The beam set of the unmanned plane U3 comprises a task beam set B3Task time ordered set P3Winner set z3Set of time stamps s3Winner offer set y1、y2、…、y9And initializing all the beam sets to be empty sets.
In step S3, the following substeps are included:
s31, obtaining marginal benefits of the unmanned aerial vehicle i to the reconnaissance area j;
s32, judging whether the unmanned aerial vehicle i can obtain the task of the reconnaissance area j;
s33, determining the path sequence P of the scout area jiThe insertion position in (1);
and S34, updating the beam set of the unmanned aerial vehicle i.
In step S31, U1, U2 and U3 obtain the marginal benefits of different reconnaissance areas respectively by:
wherein lambda is 0.9, ValjThe marginal benefit of, for example, drone U1 for reconnaissance area T1 is 0.975:
t11the distance between the U1 position (930,1010) and the T1 position (120,1740) of the drone is divided by the drone speed, i.e.
In step S32, the drones U1, U2, and U3 respectively determine whether the reconnaissance areas T1 to T9 exceed their flight ranges, and if not, it indicates that the reconnaissance area mission can be obtained, and if it exceeds the flight range, the drones cannot obtain the reconnaissance area mission.
In step S33, the drone U1 inserts the reconnaissance area j into different positions in the route order, calculates earnings according to the earnings index in step S1, and takes the position at which the maximum earnings are obtained as the route order P of the reconnaissance area jiThe insertion position in (1) can be expressed as:
and updating the obtained path sequence to a task time sequence set, and updating the obtained marginal profit to a winner bid set.
Also, in step S34, the drones U2 and U3 perform updating of the mission timing set and the winner bid set, respectively.
In step S41, U1, U2 and U3 communicate with each other, share the respective winner and winner bid sets and record the time at which the communication took place.
For the reconnaissance area T1, the drone corresponding to the maximum bid value in the winner bid set is selected to perform the reconnaissance area mission by comparing the marginal profit of the drone with the marginal profit of other drones, which may be expressed as
U1 is finally selected as the drone of the reconnaissance area T1.
Further, after the unmanned aerial vehicles communicate with each other, the timestamp of the unmanned aerial vehicle bundle set is updated.
For other unmanned aerial vehicles (U2 and U3) which do not obtain the tasks of the scout area T1, the tasks in the scout area T1 and the subsequent tasks in the U2 and U3 task bundles are emptied, and the steps S3 to S41 are repeated, so that the distribution of all the tasks can be completed.
The results of the experiment are shown in FIG. 2.
Example 2
4 unmanned aerial vehicles carry out target reconnaissance task distribution on 20 areas with high possibility of targets in the range of 500 x 500km,
the unmanned aerial vehicle carries a sensitive element, and the speed and the position of the sensitive element are shown in a table II; scout region type and probability mujAs shown in table three, the coordinates of the scout area are shown in table four.
Watch two
Watch III
Watch four
| Reference numerals | Task area type | Task point coordinates/km | Reference numerals | Task area type | Task point coordinates/km |
| T 1 | Type I | (219,376) | T 11 | Type II | (138,420) |
| T 2 | Type I | (191,128) | T 12 | Type II | (340,127) |
| T 3 | Type I | (383,253) | T 13 | Type II | (328,407) |
| T 4 | Type I | (398,350) | T 14 | Type II | (81,122) |
| T 5 | Type I | (93,445) | T15 | Type II | (99,465) |
| T 6 | Type I | (245.480) | T 16 | Type II | (249,175) |
| T 7 | Type I | (283,284) | |
Type II | (480,98) |
| T 8 | Type I | (323,69) | |
Type II | (170,126) |
| T 9 | Type I | (355,75) | T 19 | Type II | (293,308) |
| T 10 | Type I | (377,129) | T 20 | Type II | (112,237) |
The distribution method comprises the following steps:
s1, establishing a profit model;
s2, initializing a beam set;
s3, performing primary task allocation according to the income model;
and S4, resolving conflict allocation in the preliminary task allocation to obtain final task allocation.
In step S1, creating a revenue model includes creating a scout matrix and determining a revenue indicator, the element Z in the scout matrixijFor scouting capability matrix
And a target type matrix in the scout area
Multiplying or obtaining to obtain a scout matrix:
the profit index is
In step S2, the bundle sets are set for U1 to U4, respectively, and the bundle sets are initialized to the empty sets.
In step S3, the following substeps are included:
s31, obtaining marginal benefits of the unmanned aerial vehicle i to the reconnaissance area j;
s32, judging whether the unmanned aerial vehicle i can obtain the task of the reconnaissance area j;
s33, determining the path sequence P of the scout area jiThe insertion position in (1);
and S34, updating the beam set of the unmanned aerial vehicle i.
In step S31, the U1 to U4 obtain the marginal benefits of different reconnaissance areas respectively by the following formula:
wherein lambda is 0.6, Valj＝0.975。
In the marginal profit calculation process, the reconnaissance capability of the type I unmanned aerial vehicle can better reconnaissance the type I area, and the profit of the type II area is reduced by half; similarly, the reconnaissance capability of the type II unmanned aerial vehicle can better reconnaissance the type II area, and the income of the type I area is halved.
In step S32, the drones U1 to U4 respectively determine whether the reconnaissance areas T1 to T20 exceed their flight ranges, and if not, it indicates that the reconnaissance area mission can be obtained, and if it exceeds the flight range, the drone cannot obtain the reconnaissance area mission.
In step S33, the drone U1 inserts the reconnaissance area j into different positions in the route order, calculates earnings according to the earnings index in step S1, and takes the position at which the maximum earnings are obtained as the route order P of the reconnaissance area jiThe insertion position in (1) can be expressed as:
and updating the obtained path sequence to a task time sequence set, and updating the obtained marginal profit to a winner bid set.
Similarly, in step S34, the drones U2 to U4 perform updating of the mission time series set and the winner bid set, respectively.
In step S41, U1 to U4 communicate with each other, share their own winner and winner bid sets, and record the time of communication.
For the reconnaissance area T1, the drone corresponding to the maximum bid value in the winner bid set is selected to perform the reconnaissance area mission by comparing the marginal profit of the drone with the marginal profit of other drones, which may be expressed as
U2 is finally selected as the drone of the reconnaissance area T1.
Further, after the unmanned aerial vehicles communicate with each other, the timestamp of the unmanned aerial vehicle bundle set is updated.
For other unmanned aerial vehicles (U1, U2 and U3) which do not obtain the tasks of the reconnaissance area T1, the tasks in the reconnaissance area T1 and the subsequent tasks are emptied in the U2 and U3 task bundles, and the steps S3 to S41 are repeated, so that all the tasks can be distributed.
The results of the experiment are shown in FIG. 4.
Comparative example 1
The same experiment as in example 1 was performed, except that the CBBA algorithm was used for task assignment.
The results of the experiment are shown in FIG. 3.
Comparative example 2
The same experiment as in example 2 was performed, except that the CBBA algorithm was used for task assignment.
The results of the experiment are shown in FIG. 5.
Experimental example 1
Comparing the result of example 1 (fig. 2) with the result of comparative example 1 (fig. 3), it can be clearly seen that the task allocation scheme formed in example 1 enables the task path of each drone to be more close to forming a closed loop, thereby facilitating recovery of the drone and facilitating the drone to perform other subsequent tasks.
Comparing the result of embodiment 2 (fig. 4) and the result of comparative example 2 (fig. 5), it can obviously be found that embodiment 2 forms the task allocation scheme, and every unmanned aerial vehicle's task route more tends to form the closed loop, compares in the task allocation of comparative example 2, and the low actual reconnaissance task allocation that is more suitable for of the low recovery consumption of unmanned aerial vehicle cluster, and simultaneously, the mode of communication between the planes and update time stamp avoids many unmanned aerial vehicles to carry out close task respectively, avoids the repeated execution of task.
In the description of the present invention, it should be noted that the terms "upper", "lower", "inner", "outer", "front", "rear", and the like indicate orientations or positional relationships based on operational states of the present invention, and are only used for convenience of description and simplification of description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," "third," and "fourth" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the description of the present invention, it should be noted that, unless otherwise specifically stated or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning either a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; the connection may be direct or indirect via an intermediate medium, and may be a communication between the two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
The present invention has been described above in connection with preferred embodiments, but these embodiments are merely exemplary and merely illustrative. On the basis of the above, the invention can be subjected to various substitutions and modifications, and the substitutions and the modifications are all within the protection scope of the invention.
Claims (10)
1. A method for distributing a reconnaissance task in an unmanned aerial vehicle cluster air area is characterized by comprising the following steps:
s1, establishing a profit model;
s2, initializing a beam set;
s3, performing primary task allocation according to the income model;
and S4, resolving conflict allocation in the preliminary task allocation to obtain final task allocation.
2. The method of claim 1,
in step S1, the benefit model is used to describe the scout benefits of the drone cluster, including establishing a scout matrix and determining a benefit index,
the reconnaissance matrix is used for describing the performability of different unmanned aerial vehicles to different reconnaissance areas and is represented as NuLine NtA two-dimensional matrix of columns, wherein NuRepresenting the total number of unmanned aerial vehicles, NtRepresenting the total number of scout areas.
3. The method of claim 2,
the revenue indicator may be expressed as
Wherein x isijIndicates whether the unmanned aerial vehicle i performs reconnaissance on the reconnaissance area j, xijDenotes that drone i performs reconnaissance on reconnaissance area j, xij0 means that drone i does not perform reconnaissance on reconnaissance area j;
Piindicating the path sequence during the reconnaissance of drone i, cijAs a function of the benefit.
4. The method of claim 1,
the beam set includes:
the task bundle set is used for describing a task sequence of the unmanned aerial vehicle;
the task time sequence set is used for describing the task time sequence of the unmanned aerial vehicle;
the execution time set is used for describing the time for the unmanned aerial vehicle to scout different tasks according to the task time sequence set;
a winner set used for describing whether the unmanned aerial vehicle successfully bids on different reconnaissance areas;
a winner offer set used for representing the maximum output value of each unmanned aerial vehicle when the unmanned aerial vehicle auctions on the reconnaissance area at the current moment;
a set of timestamps to represent a last time information interaction time between drone i and its neighboring drone.
5. The method of claim 1,
in step S3, the post-task allocation benefits are made more accurate by adding a probabilistic constraint on the existence of the target.
6. The method of claim 1,
in step S3, the following substeps are included:
s31, obtaining marginal benefits of the unmanned aerial vehicle i to the reconnaissance area j;
s32, judging whether the unmanned aerial vehicle i can obtain the task of the reconnaissance area j;
s33, determining the path sequence P of the scout area jiThe insertion position in (1);
and S34, updating the beam set of the unmanned aerial vehicle i.
7. The method of claim 6,
in step S31, the marginal profit cij(Pi) Comprises the following steps:
wherein, tijIndicating the order of unmanned plane i along path PiTime spent to scout area j;
λ is the elapsed time tijThe affected income factor parameters can be set according to actual needs;
Valjvalue coefficient inherent to the scout area j, 0 < Valj＜1；
ZijAnd the completeness of the reconnaissance area j by the unmanned plane i in the reconnaissance matrix is represented.
8. The method of claim 6,
in step S33, the scout area j is inserted into different positions in the route order, the profit is calculated, and the position at the time of the maximum profit is taken as the route order P of the scout area jiThe insertion position in (1) is represented as:
wherein n isi,jiAdding the scout region j to the path sequence P when the representation yield is maximumiPosition of (2), indicating the route order PiThe element position in (1).
9. The method of claim 6,
in step S34, steps S31 to S33 are repeated, the scout area allocation and the route order determination are performed for each drone, and the task bundle set, the task time sequence set, the winner set, and the winner bid set in the bundle set are updated to complete the task allocation.
10. The method of claim 1,
in step S4, the following substeps are included:
s41, all unmanned aerial vehicles communicate with each other, and the unmanned aerial vehicle corresponding to the maximum bid value in the winner bid set is selected to execute the task of the reconnaissance area j;
and S42, repeating the step S41 and completing the distribution of all conflict tasks.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110420280.1A CN113190038A (en) | 2021-04-19 | 2021-04-19 | Method for distributing reconnaissance tasks in unmanned aerial vehicle cluster air area |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110420280.1A CN113190038A (en) | 2021-04-19 | 2021-04-19 | Method for distributing reconnaissance tasks in unmanned aerial vehicle cluster air area |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113190038A true CN113190038A (en) | 2021-07-30 |
Family
ID=76977426
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110420280.1A Pending CN113190038A (en) | 2021-04-19 | 2021-04-19 | Method for distributing reconnaissance tasks in unmanned aerial vehicle cluster air area |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113190038A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113867418A (en) * | 2021-09-17 | 2021-12-31 | 南京信息工程大学 | Unmanned aerial vehicle cluster autonomous cooperative scout task scheduling method |
| CN114200964A (en) * | 2022-02-17 | 2022-03-18 | 南京信息工程大学 | Unmanned aerial vehicle cluster cooperative reconnaissance coverage distributed autonomous optimization method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107330588A (en) * | 2017-06-19 | 2017-11-07 | 西北工业大学 | A kind of mission planning method of many base isomery unmanned plane coordinated investigations |
| KR20180085562A (en) * | 2017-01-19 | 2018-07-27 | 금오공과대학교 산학협력단 | Search and reconnaissance method by multiple drones using particle swarm algorithm |
| US10140875B1 (en) * | 2017-05-27 | 2018-11-27 | Hefei University Of Technology | Method and apparatus for joint optimization of multi-UAV task assignment and path planning |
| CN109901616A (en) * | 2019-03-29 | 2019-06-18 | 北京航空航天大学 | A kind of isomery unmanned aerial vehicle group distributed task scheduling planing method |
| CN110134146A (en) * | 2019-06-14 | 2019-08-16 | 西北工业大学 | A kind of distributed multiple no-manned plane method for allocating tasks under uncertain environment |
| CN111695776A (en) * | 2020-05-11 | 2020-09-22 | 清华大学 | Unmanned aerial vehicle cluster distributed online cooperative area reconnaissance method and device |
-
2021
- 2021-04-19 CN CN202110420280.1A patent/CN113190038A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20180085562A (en) * | 2017-01-19 | 2018-07-27 | 금오공과대학교 산학협력단 | Search and reconnaissance method by multiple drones using particle swarm algorithm |
| US10140875B1 (en) * | 2017-05-27 | 2018-11-27 | Hefei University Of Technology | Method and apparatus for joint optimization of multi-UAV task assignment and path planning |
| CN107330588A (en) * | 2017-06-19 | 2017-11-07 | 西北工业大学 | A kind of mission planning method of many base isomery unmanned plane coordinated investigations |
| CN109901616A (en) * | 2019-03-29 | 2019-06-18 | 北京航空航天大学 | A kind of isomery unmanned aerial vehicle group distributed task scheduling planing method |
| CN110134146A (en) * | 2019-06-14 | 2019-08-16 | 西北工业大学 | A kind of distributed multiple no-manned plane method for allocating tasks under uncertain environment |
| CN111695776A (en) * | 2020-05-11 | 2020-09-22 | 清华大学 | Unmanned aerial vehicle cluster distributed online cooperative area reconnaissance method and device |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113867418A (en) * | 2021-09-17 | 2021-12-31 | 南京信息工程大学 | Unmanned aerial vehicle cluster autonomous cooperative scout task scheduling method |
| CN114200964A (en) * | 2022-02-17 | 2022-03-18 | 南京信息工程大学 | Unmanned aerial vehicle cluster cooperative reconnaissance coverage distributed autonomous optimization method |
| CN114200964B (en) * | 2022-02-17 | 2022-04-26 | 南京信息工程大学 | Unmanned aerial vehicle cluster cooperative reconnaissance coverage distributed autonomous optimization method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107886201B (en) | Multi-objective optimization method and device for multi-unmanned aerial vehicle task allocation | |
| CN113190038A (en) | Method for distributing reconnaissance tasks in unmanned aerial vehicle cluster air area | |
| CN107330588B (en) | Task planning method for cooperative reconnaissance of multi-base heterogeneous unmanned aerial vehicle | |
| CN109409773B (en) | Dynamic planning method for earth observation resources based on contract network mechanism | |
| CN109919484B (en) | On-satellite autonomous task planning method | |
| CN115525068B (en) | Unmanned aerial vehicle cluster cooperative task allocation method based on iterative optimization | |
| CN116166048B (en) | Unmanned aerial vehicle group fault-tolerant task planning method | |
| CN104156943B (en) | Multi objective fuzzy cluster image change detection method based on non-dominant neighborhood immune algorithm | |
| CN112580801A (en) | Reinforced learning training method and decision-making method based on reinforced learning | |
| CN112633654A (en) | Multi-unmanned aerial vehicle task allocation method based on improved cluster expansion consistency bundle algorithm | |
| CN107180309B (en) | Collaborative planning method for space-sky-ground observation resources | |
| CN112966773B (en) | Unmanned aerial vehicle flight condition mode identification method and system | |
| CN114172942A (en) | Cooperative task allocation and track optimization method for multi-unmanned aerial vehicle assisted Internet of things | |
| CN114841055A (en) | Unmanned aerial vehicle cluster task pre-distribution method based on generation of countermeasure network | |
| CN114399161A (en) | Multi-unmanned aerial vehicle cooperative task allocation method based on discrete mapping differential evolution algorithm | |
| CN115688568A (en) | Scheduling method of hypersensitive agile satellite multi-satellite regional imaging task | |
| CN116088586B (en) | Method for planning on-line tasks in unmanned aerial vehicle combat process | |
| CN104833359B (en) | A kind of star chart mode identification method based on discrete Markov characteristic sequence model | |
| CN113065094A (en) | Situation assessment method and system based on accumulated foreground value and three-branch decision | |
| CN112179335A (en) | Star map identification method based on simplest general subgraph | |
| CN116739739A (en) | Loan amount evaluation method and device, electronic equipment and storage medium | |
| CN114565203A (en) | Multi-satellite multi-task planning method based on large neighborhood search algorithm and K-means | |
| CN114995503A (en) | Unmanned aerial vehicle routing inspection path optimization method | |
| CN114967748A (en) | Unmanned aerial vehicle path planning method based on space deformation | |
| CN115080388A (en) | Simulation test scene generation method for automatic driving system |
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 |