CN112149959B - Distributed sensor-weapon-target joint allocation method - Google Patents
Distributed sensor-weapon-target joint allocation method Download PDFInfo
- Publication number
- CN112149959B CN112149959B CN202010870029.0A CN202010870029A CN112149959B CN 112149959 B CN112149959 B CN 112149959B CN 202010870029 A CN202010870029 A CN 202010870029A CN 112149959 B CN112149959 B CN 112149959B
- Authority
- CN
- China
- Prior art keywords
- weapon
- target
- unit
- sensor
- sensor unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06311—Scheduling, planning or task assignment for a person or group
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06316—Sequencing of tasks or work
Abstract
The invention discloses a sensor-weapon-target joint allocation method which can effectively and quickly perform reconnaissance task allocation and attack task allocation on multi-target tasks, reduce communication bandwidth and improve allocation efficiency. Each sensor unit determines a target with the highest self-capturing success rate, determines the targets with the highest capturing success rates of all the sensor units through communication interaction, and broadcasts the optimal sensor-target distribution relation to the weapon units; the weapon units acquire the hitting success rate of the optimal target, and the weapon units with the highest hitting success rate in all the weapon units are determined through communication interaction to obtain the optimal weapon-target distribution relation. This results in the sensor-weapon-target assignment for the current round. And circularly executing the distribution operation to complete the distribution of all the sensor-weapon-targets. The sensor unit and the weapon unit execute tasks according to the allocation result of the sensor-weapon-target.
Description
Technical Field
The invention relates to the technical field of multi-agent, in particular to a sensor-weapon-target joint distribution method.
Background
The problem of sensor-weapon-target joint distribution is that a plurality of attacking targets attack the local site or the base, the local (defending party) detects and captures by using a sensor unit, and the weapon unit is damaged and hit.
In a complex real dynamic environment, due to factors such as environmental change, time constraint and uneven resource distribution, the requirements of the parties on the allocation efficiency and the instantaneity of resources such as sensors and weapons are extremely high. The sensor unit and the weapon unit need to solve coordination and cooperation problems of resource allocation, task scheduling, behavior coordination, conflict resolution and the like under the condition of limited time and limited resources. If a centralized decision-making method is adopted, the combat command system is used for uniformly distributing combat resources, the requirements of task distribution efficiency and calculation real-time performance can not be met, and the requirement on communication bandwidth is high. Meanwhile, the centralized decision command system becomes a main attack target of an enemy, and once the command system is attacked, the decision capability and the fighting capability of the unit of the party are greatly weakened.
Disclosure of Invention
In view of this, the invention provides a sensor-weapon-target joint allocation method, which can effectively and quickly perform scout task allocation and attack task allocation on multi-target tasks, reduce communication bandwidth and improve allocation efficiency.
In order to solve the technical problem, the invention is realized as follows:
a sensor-weapon-target joint distribution method is characterized in that a defense party has S sensor units and W weapon units, and the sensor units and the weapon units can communicate with each other; the joint distribution method comprises the following steps:
step 1: initializing sensor-weapon-target allocation scheme matrix X, X = [ X ] ijk ]Wherein i =1,2, \8230;, S; j =1,2, \8230;, W; k =1,2, \ 8230;, T, x ijk =0, t is the number of enemy attack targets; the size of the X matrix is S × W × T; initializing the maximum number of detectable targets y of the sensor unit i And the maximum number of targets of the weapon unit z j ；
And 2, step: for sensor unit i, if y i Jumping to the step 5 when the value is less than or equal to 0; if y is i >0, determining a target with the highest self-capturing success rate by the sensor unit i to form a first distribution relation between the sensor unit and the best captured target; the first distribution relation records the ID of the sensor unit, the ID of the target and parameters representing the capturing success rate; each sensor unit interacts with the first distribution relation through communication, and the optimal first distribution relation with the highest capturing success rate in all the sensor units is obtained
A consensus is reached; so far, all sensors agree to identify i * Number sensor unit execution k * Detecting a task by a number target, and taking the task as a sensor-target distribution result; the sensor unit also assigns an optimal first distribution relationship
Broadcast to weapon units;
and step 3: for weapon unit j, if z j Jumping to the step 5 when the value is less than or equal to 0; if z is j >0, then the weapon unit j acquires its own striking target k * Forming weapon unit and target k * The second allocation relationship of (a); the second distribution relation records weapon unit ID, target ID and parameters representing striking success rate; each weapon unit interacts the second distribution relation through communication, and the optimal second distribution relation with the highest hitting success rate is obtained
Reaching a consensus; so far, all weapon units agree on j * Weapon unit execution k * Number target hit task, and as weapon-target assignment result;
and 4, step 4: the weapon unit broadcasts the weapon-target assignment result to each sensor unit, and the weapon unit and the sensor unit broadcast the elements in the matrix X of the respective assignment schemes
Updating to 1; will be provided with
From decreasing by 1, will
Self-subtracting 1 to update the constraint condition, and returning to the step 2;
and 5: each sensor unit and weapon unit executes the task according to the information of the task assignment matrix X.
Preferably, the step 2 is:
step 201: for sensor unit i, if y i Jumping to the step 5 when the speed is less than or equal to 0; if y is i >0, acquiring the capturing success rate of capturing each target by the sensor unit i, and selecting the target with the highest capturing success rate to form a first distribution relation;
step 202: based on the auction algorithm, the sensor unit i broadcasts a first distribution relation to other sensor units; within a time limit delta t, a sensor unit i receives first distribution relations sent by other sensor units; if the parameter representing the capturing success rate in the first distribution relation received by the sensor unit i is larger, replacing the first distribution relation of the sensor unit i, and broadcasting a new first distribution relation of the sensor unit i again;
through the operations of steps 201 and 202, each sensor unit obtains the optimal first distribution relation with the highest capturing success rate;
step 203: the sensor unit broadcasts the optimal first allocation relationship to the weapon units.
Preferably, the step 3 is:
step 301: for weapon unit j, if z j Jumping to the step 5 when the value is less than or equal to 0; if z is j >0, the weapon unit j acquires the self-hit target k * Forming a second distribution relationship;
step 302: based on the auction algorithm, weapon unit j broadcasts a second allocation relationship to the other weapon units; within time limit Δ t, weapon unit j receives a second allocation relationship sent by another weapon unit; if the parameter for representing the striking success rate in the second distribution relation received by the weapon unit j is larger, replacing the second distribution relation of the weapon unit j, and broadcasting a new second distribution relation of the weapon unit j again;
through the operations of steps 301 and 302, each weapon unit obtains the optimal second distribution relationship with the highest striking success rate.
Preferably, the parameter representing the capturing success rate is marginal profit of the sensor;
m represents a loop execution stepThe m-th wheel from 2 to 4; marginal benefit delta of sensor unit i on target k at round m +1 pairing m+1 (i, k) the calculation formula is:
s m (k)＝α k ×[1-P mis (k)]
δ m+1 (i,k)＝s m+1 (i,k)-s m (k)
wherein s is m (k) The total capturing success rate of the target k when the sensor unit i is not added in the mth round is set; s m (k) The total capturing success rate of the target k when the sensor unit i is added into the (m + 1) th round; p is a radical of ik The probability of capture of target k for a known sensor unit i; if sensor unit i is assigned to target k, σ ik =1, otherwise, σ ik ＝0；α k Is the strike bias of target k; p mis (k) Represents the probability that each sensor unit assigned to target k cannot complete the capture of target k:
Similarly, the parameter representing the striking success rate is the marginal income of the weapon;
m represents the m-th round when steps 2-4 are executed circularly; marginal benefit delta of weapon unit j to target k in the m +1 th pairing m+1 (j, k) the formula is:
w m (k)＝α k ×[1-P ad (k)]
δ m+1 (j,k)＝s m+1 (j,k)-s m (k)
wherein, w m (k) The total hit success rate of the target k when no weapon unit j is added in the mth round; s is m (k) The total hit success rate to the target k when the weapon unit j is added to the (m + 1) th round; q. q of jk The hit probability of a known weapon unit j on target k; if weapon unit j is assigned to target k, σ jk =1, otherwise, σ jk ＝0；α k Is the strike bias of target k; p is ad (k) Representing the probability that each weapon unit assigned to target k is unable to complete a strike on target k:
θ k representing a set of weapon units that have engaged the target k-strike task.
Preferably, the striking bias α of the target k is k Using the threat value tr of target k k 。
Preferably, the sensor unit will also optimize the first distribution relation
The broadcast to the weapon units is: wherein one sensor unit will optimize the first distribution relation
Broadcast to all weapon units.
Has the advantages that:
(1) The invention provides a sensor-weapon-target joint allocation method, which adopts a distributed allocation scheme, wherein each unit autonomously decides to detect or attack a target by analyzing the current operation form and considering the threat value of an attack target, communicates with other operation units and executes tasks according to the allocation scheme achieving consensus. The distributed task allocation has the advantages that: each combat unit can calculate required parameters in parallel, and a target unit is independently decided, so that the calculation speed is high, and the real-time performance is strong; the communication pressure is lower because the central command unit is not available for the party. Compared with a centralized decision, even if the command unit is hit, the fighting loss of the party is small; each combat unit has strong expandability and robustness, and the number of the sensors and the weapon units can be dynamically increased and decreased.
(2) The invention is based on the execution sequence of weapon striking detected by the sensor firstly, and adopts the scheme of firstly determining the sensor-target and then determining the target-weapon, compared with the scheme of parallel allocation of the sensor and the weapon, the serial allocation scheme can prevent the conflict of task allocation. And after the sensor-target is determined, only target information needs to be sent to the weapon unit, so that the information sending amount is small, and the communication burden is reduced.
(3) The auction algorithm is adopted to achieve consensus among the distributed sensors according to the sensor target distribution scheme, consensus among the distributed weapons according to the weapon target distribution scheme is achieved, the real-time performance is strong, each sensor and each weapon unit can be added or deleted dynamically, information communication is mainly concentrated in the coordination process of the task scheme, and the requirement of communication bandwidth can be effectively reduced.
(4) The invention adopts the marginal profit idea in economics to perform distributed target allocation on target tasks. The method can dynamically explore extra income brought by the newly added scheme according to the known distribution scheme of 'sensor-weapon-target', and update the distribution scheme according to the size of the newly added income value, so that a more reasonable distribution mode is found, and defensive side resources are maximized.
Drawings
FIG. 1 is a flow chart of a distributed sensor-weapon-target joint assignment method;
FIG. 2 is a schematic diagram of the flow of information to each unit;
fig. 3 is a diagram illustrating different types of messages.
Detailed Description
The invention is described in detail below by way of example with reference to the accompanying drawings.
The task scene of the invention is as follows: the defense party of our part needs to carry out two tasks of detecting and capturing and target hitting on a plurality of attacking targets of the enemy. The sensor unit of my party is responsible for detecting and capturing tasks, and when the enemy target is successfully detected and captured, the weapon unit strikes the tracked and captured enemy target.
Based on the scene, the invention provides a distributed sensor-weapon-target joint distribution method, and a rapid and reasonable target distribution scheme is generated by distributed negotiation among multiple platform units. The invention takes the improved auction algorithm as the solving algorithm of the distributed task allocation, and adopts the idea of the economic border income when evaluating the value income. The scheme has low requirement on communication bandwidth, high task allocation speed and can effectively prevent task allocation conflict.
The specific situation of the problem is as follows: assume that there are T enemy attack targets, numbered 1,2, \8230, respectively, and a T target. I have S sensor units and W weapon units, denoted 1,2, \ 8230;, S and 1,2, \ 8230;, W, respectively, for the defending party. Let vk denote the threat value of target k, k =1,2, \ 8230;, T. When a sensor unit i is assigned to an object k, the probability that the object is successfully captured by the sensor unit is p ik . When a weapon unit j is assigned to a target k, the probability of occurrence of the event that the target k successfully destroys is recorded as q jk . In the present invention p ik And q is jk In known amounts.
Probability of successful destruction of a target k by the combined action of a sensor and a weapon
Comprises the following steps:
wherein the content of the first and second substances,
represents the probability of occurrence of the event that target k is successfully captured by the sensors assigned to it:
wherein σ is given if sensor cell i is assigned to target k ik =1; otherwise, σ ik ＝0。
wherein if a weapon j is assigned to a target k, σ jk =1, otherwise, σ jk ＝0。
The goal of the problem is to maximize the expected devastating value of defender resources to the attacking target.
Constraints of the problem include:
the maximum number of detection targets of the sensor unit i is not more than y i A, y i Is a positive integer.
Maximum number of targets hit by weapon Unit j is not greater than z j Z is j Is a positive integer.
Communication topology constraint:
the sensor units and the weapon units are provided with communication modules, and all the sensor units and the weapon units can send and receive messages to each other.
As shown in fig. 1 and fig. 2, the distributed sensor-weapon-target joint distribution method provided by the present invention includes the following steps:
step 1: for each sensor unit i, the capture probability p of each target by the sensor unit i is given ik K =1,2, \ 8230;, T. For each weapon unit j, the probability q of striking the targets by the weapon unit j is given jk K =1,2, \8230;, T. Initializing sensor-weapon-target allocation scheme matrix X, X = [ X ] ijk ]Wherein i =1,2, \8230;, S; j =1,2, \ 8230;, W; k =1,2, \8230;, T, x ijk =0,X matrix sizeS × W × T. Threat value tr for a given target k . An initialization count variable m =1. Initializing the maximum number of detectable targets y of the sensor unit i (ii) a Initializing maximum number of targets of weapon Unit j 。
Step 2: for sensor unit i, if y i And (5) jumping to the step 5. If y is i >And 0, determining the target with the highest self-capturing success rate by the sensor unit i, and forming a first distribution relation between the sensor unit and the best captured target. The first distribution relationship records a sensor unit ID, a target ID and a parameter representing the capture success rate. Each sensor unit interacts with the first distribution relation through communication, and the optimal first distribution relation with the highest capture success rate in all the sensor units is obtained
A consensus was reached. So far, all sensors agree to i * Number sensor unit execution k * And detecting the task by the number target and serving as a sensor-target distribution result. The sensor unit will also assign the relation optimally to the first
Broadcast to the weapon units.
In the preferred embodiment of the invention, the auction algorithm is adopted to realize each sensor unit pair
And in the process of consensus achievement, the first distribution relation is expressed by using a value vector, and meanwhile, the parameter representing the capturing success rate is expressed by using the marginal income of the sensor. Then, the step 2 specifically includes:
step 201: for sensor unit i, if y i Jumping to the step 5 when the speed is less than or equal to 0; if y is i >0, calculating all target marginal benefits by the sensor unit i to obtain a marginal benefit vector { delta m (i,1),δ m (i,2),…,δ m The (i, K) } vector, i is the number of the sensor unit, i =1,2, \ 8230, S, marginal profit vector contains K elements. Assuming the most of the marginal benefit vectorsThe large marginal profit value is
k is its corresponding object number. The sensor unit i obtains the value vector of the sensor unit i
sensor _ id is the sensor unit number; target _ id is the enemy target number, and reward _ value is the maximum marginal profit
Step 202: value vectors broadcast by sensor unit i to other sensor units based on auction algorithm
See message type 1 in fig. 3. Within the time limit Δ t, sensor unit i receives broadcasts of other sensor units (i.e., message type 1). If the value vector received by sensor unit i
If the marginal profit value is larger, the value vector of the user is replaced, and a new value vector of the user is broadcasted again
Through the operations of the above steps 201 and 202, the value vector of each sensor unit stores the sensor and the target with the largest marginal profit:
all sensors have been able to achieve "consensus" in the sense that: in the auction round, all sensors reach "consensus", i * Number sensor execution k * And (5) detecting a task.
Step 203: the sensor unit is to
Broadcast to weapon units, see message type 2 in fig. 3. Here, one of the sensor units may be used to measure
Broadcast to the various weapon units.
Marginal benefit is a concept in economics, meaning the increase in utility by adding a unit of element. The idea of marginal profit adopted by the invention follows the following principle:
1. the greater the marginal benefit that a certain "sensor-weapon-target" triple can bring, it should be given a higher priority to process; because the invention is divided into two sections of sensor-target and weapon-target distribution, each section should consider the marginal benefit brought by the binary combination to be larger.
2. If any valid "sensor-weapon" combination is assigned to a target, the threat value of that target will be reduced.
The specific calculation method of the marginal profit of the sensor is described below.
Marginal yield delta of sensor unit i to target k at bid in round m +1 m+1 (i, k) the calculation formula is:
wherein s is m (k) The total capturing success rate of the target k when the sensor unit i is not added in the mth round is set; s m (k) The total capturing success rate of the target k when the sensor unit i is added into the (m + 1) th round; p is a radical of ik The probability of capture of target k for a known sensor unit i; if sensor unit i is assigned to target k, σ ik =1, otherwise, σ ik ＝0。P mis (k) Indicating the probability that each sensor unit assigned to target k is unable to complete the capture of target k.
Tr in the above formula (1) k Which is the threat value of target k, appears as a weight of hit success rate. If the target threat value is large, the marginal benefit needs to be increased by a multiple. In practice tr k Other strike tendencies α characterizing the target k may also be employed k Instead of the parameters, for example, the striking bias α may be set for different target types k ，α k The greater the striking is required.
And step 3: knowledge of k of the weapon Unit that the Current sensor Unit has achieved "consensus * After target number. For weapon unit j, if z j And (5) jumping to the step 5. If z is j >0, then the weapon unit j acquires its own striking target k * Forming weapon unit and optimum striking target k * The second allocation relationship of (1); the second distribution relation records weapon unit ID, target ID and parameters representing striking success rate; each weapon unit interacts the second distribution relation through communication, and the optimal second distribution relation with the highest striking success rate is aimed at
A consensus was reached. So far, all weapon units agree on j * Weapon unit execution k * The horn hits the task and is the result of weapon-target assignment.
In the preferred embodiment of the invention, each weapon unit pair is realized by using auction algorithm
And in the process of consensus achievement, the second distribution relation is expressed by using a value vector, and the parameter representing the striking success rate is expressed by using the marginal profit of the weapon. Then, the step 3 is specifically:
step 301: for weapon unit j, if z j Jumping to the step 5 when the value is less than or equal to 0; if z is j >0, weapon units j to k * Calculating marginal profit of the target to obtain delta m (j,k * ) Where the subscript j is the weapon unit number, j =1,2, \8230, W. The weapon units are sorted to obtain a value vector
Step 302: based on auction algorithm, weapon unit j broadcasts value vector to other weapon units
See message type 3 in fig. 3. Within time limit Δ t, weapon unit j receives the value vectors (still message type 3) sent by the other weapon units. If the marginal profit value in the received value vectors is larger, replacing the value vectors per se, and broadcasting new value vectors again
To this end, the weapon units achieve a "consensus":
the meaning is that: in this round of auctions, all weapon units agree to j * Horn weapon unit execution k * And (5) striking the target.
The following describes a specific calculation method for the weapon marginal profit.
Marginal receipts of target k by weapon Unit j at bid in round m +1Benefit delta m+1 The formula (j, k) is calculated as:
w m (k) The total hit success rate of the target k when no weapon unit j is added in the mth round; s m (k) The total hit success rate of the target k when the weapon unit j is added into the (m + 1) th round; q. q of jk The hit probability of a known weapon unit j on target k; if weapon unit j is assigned to target k, σ jk =1, otherwise, σ jk ＝0；α k Is the strike bias of target k; p is ad (k) Represents the probability that each weapon unit assigned to target k fails to complete a hit on target k:
θ k representing a set of weapon units that have engaged in a target k-strike task.
Tr in the above formula (3) k Is the threat value of target k, which appears as a weight of hit success rate. If the target threat value is large, the marginal benefit needs to be increased by a multiple. In practice tr k Other strike bias alpha characterizing the target k may also be used k Instead of the parameters, for example, the striking bias α may be set for different target types k ，α k The greater the striking is required.
And 4, step 4: the weapon unit sends information to the sensor unit, and the information of the weapon unit distribution in step 3
Broadcast to the sensor units, see message type 4 in fig. 3. Weapon and sensor units will be assigned to the scheme matrix
Let m add 1 and update the constraint:
and returning to the step 2.
And 5: each sensor unit and weapon unit executes the task according to the information of the task assignment matrix X.
The above embodiments only describe the design principle of the present invention, and the shapes and names of the components in the description may be different without limitation. Therefore, a person skilled in the art of the present invention can modify or substitute the technical solutions described in the foregoing embodiments; such modifications and substitutions do not depart from the spirit and scope of the present invention.
Claims (5)
1. A sensor-weapon-target joint distribution method is characterized in that a defense party has S sensor units and W weapon units, and the sensor units and the weapon units can communicate with each other; the joint distribution method comprises the following steps:
step 1: initializing sensor-weapon-target allocation scheme matrix X, X = [ X ] ijk ]Wherein i =1,2, \8230;, S; j =1,2, \ 8230;, W; k =1,2, \ 8230;, T, x ijk =0, t is the number of enemy attack targets; the size of the X matrix is S × W × T; initializing the maximum number of detectable objects y of the sensor unit i And the maximum number of targets of the weapon unit z j ；
Step 2: for sensor unit i, if y i Jumping to the step 5 when the value is less than or equal to 0; if y is i >0, determining a target with the highest self-capturing success rate by the sensor unit i to form a first distribution relation between the sensor unit and the best captured target; the first distribution relation records the ID of the sensor unit, the ID of the target and parameters representing the capturing success rate; each sensor unit interacts with the first distribution relation through communication, and the optimal first distribution relation with the highest capturing success rate in all the sensor units is obtained
A consensus is reached; so far, all sensors agree to i * Number sensor unit execution k * Detecting a task by a number target, and taking the task as a sensor-target distribution result; the sensor unit also assigns an optimal first distribution relationship
Broadcast to weapon units;
and 3, step 3: for weapon unit j, if z j Jumping to the step 5 when the speed is less than or equal to 0; if z is j >0, then the weapon unit j acquires its own striking target k * Success rate of striking, forming weapon unit and target k * The second allocation relationship of (1); the second distribution relation records weapon unit ID, target ID and parameters representing striking success rate; each weapon unit interacts the second distribution relation through communication, and the optimal second distribution relation with the highest hitting success rate is obtained
A consensus is reached; so far, all weapon units agree on j * Horn weapon unit execution k * Number target hit task, and as weapon-target assignment result;
and 4, step 4: the weapon unit broadcasts the weapon-target assignment result to each sensor unit, and the weapon unit and the sensor unit broadcast the elements in the matrix X of the respective assignment schemes
Updating to 1; will be provided with
From decreasing by 1, will
Self-subtracting 1 to update the constraint condition, and returning to the step 2;
and 5: each sensor unit and each weapon unit execute tasks according to the information of the task allocation matrix X;
the step 2 is as follows:
step 201: for sensor unit i, if y i Jumping to the step 5 when the speed is less than or equal to 0; if y is i >0, acquiring the capturing success rate of capturing each target by the sensor unit i, and selecting the target with the highest capturing success rate to form a first distribution relation;
step 202: based on the auction algorithm, the sensor unit i broadcasts a first distribution relation to other sensor units; within the time limit Δ t, the sensor unit i receives the first distribution relation sent by other sensor units; if the parameter representing the capturing success rate in the first distribution relationship received by the sensor unit i is larger, replacing the first distribution relationship of the sensor unit i, and broadcasting a new first distribution relationship of the sensor unit i again;
through the operations of steps 201 and 202, each sensor unit obtains the optimal first distribution relationship with the highest capture success rate;
step 203: the sensor unit broadcasts the optimal first allocation relationship to the weapon units;
the step 3 is as follows:
step 301: for weapon unit j, if z j Jumping to the step 5 when the value is less than or equal to 0; if z is j >0, weapon unit j acquires its own hit target k * Forming a second distribution relation;
step 302: based on the auction algorithm, weapon unit j broadcasts a second allocation relationship to the other weapon units; within time limit Δ t, weapon unit j receives a second allocation relationship sent by other weapon units; if the parameter for representing the striking success rate in the second distribution relation received by the weapon unit j is larger, replacing the second distribution relation of the weapon unit j, and broadcasting a new second distribution relation of the weapon unit j again;
through the operations of steps 301 and 302, each weapon unit obtains the optimal second distribution relation with the highest striking success rate.
2. The sensor-weapon-target joint allocation method of claim 1, wherein the parameter characterizing capture success rate is marginal gain of the sensor;
m represents the m-th round when the steps 2-4 are executed circularly; marginal benefit delta of sensor unit i on target k at round m +1 pairing m+1 (i, k) the formula is:
s m (k)＝α k ×[1-P mis (k)]
δ m+1 (i,k)＝s m+1 (i,k)-s m (k)
wherein s is m (k) The total capturing success rate of the target k when the sensor unit i is not added in the mth round is set; s m+1 (i, k) is the total capturing success rate of the target k when the sensor unit i is added into the (m + 1) th round; p is a radical of ik The probability of capturing the target k for a known sensor unit i; if sensor unit i is assigned to target k, σ ik =1, otherwise, σ ik ＝0；α k Is the strike bias of target k; p mis (k) Representing the probability that each sensor unit assigned to target k cannot complete the capture of target k:
3. The sensor-weapon-target joint allocation method of claim 1, wherein the parameter characterizing the success rate of the blows is the marginal gain of the weapon;
m represents the m-th round when the steps 2-4 are executed circularly; marginal benefit delta of weapon unit j to target k in the m +1 th pairing m+1 (j, k) meterThe calculation formula is as follows:
w m (k)＝α k ×[1-P ad (k)]
δ m+1 (j,k)＝w m+1 (j,k)-w m (k)
wherein, w m (k) The total hit success rate of the target k when no weapon unit j is added in the mth round; w is a m+1 (j, k) is the total hit success rate of the target k when the weapon unit j is added into the (m + 1) th round; q. q.s jk The probability of strike on target k for a known weapon unit j; if weapon unit j is assigned to target k, σ jk =1, otherwise, σ jk ＝0；α k Is the strike bias of target k; p is ad (k) Representing the probability that each weapon unit assigned to target k is unable to complete a strike on target k:
θ k representing a set of weapon units that have engaged in a target k-strike task.
4. The combined sensor-weapon-target distribution method according to claim 2 or 3, characterized in that said target k has a strike bias α k Using the threat value tr of target k k 。
5. The sensor-weapon-target joint allocation method of claim 1, wherein the sensor unit further assigns an optimal first allocation relationship
The broadcast to the weapon units is: wherein one sensor unit will optimize the first distribution relation
Broadcast to all weapon units.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010870029.0A CN112149959B (en) | 2020-08-26 | 2020-08-26 | Distributed sensor-weapon-target joint allocation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010870029.0A CN112149959B (en) | 2020-08-26 | 2020-08-26 | Distributed sensor-weapon-target joint allocation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112149959A CN112149959A (en) | 2020-12-29 |
| CN112149959B true CN112149959B (en) | 2022-10-21 |
Family
ID=73888150
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010870029.0A Active CN112149959B (en) | 2020-08-26 | 2020-08-26 | Distributed sensor-weapon-target joint allocation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112149959B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116542400B (en) * | 2023-07-05 | 2023-09-19 | 北京航空航天大学 | Weapon target distribution method, system, equipment and medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103336885A (en) * | 2013-06-03 | 2013-10-02 | 北京航空航天大学 | Method for solving weapon-target assignment problem based on differential evolution algorithm |
| CN110096822A (en) * | 2019-05-08 | 2019-08-06 | 北京理工大学 | Multi-platform cooperative dynamic task allocation method under a kind of condition of uncertainty |
| CN111582553A (en) * | 2020-04-17 | 2020-08-25 | 上海机电工程研究所 | Information and firepower integrated task planning method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9779185B2 (en) * | 2012-04-11 | 2017-10-03 | Aselsan Elektronik Sanayi Ve Ticaret Anonim Sirketi | Test and analysis system and a method for threat evaluation and sensor/weapon assignment algorithms |
-
2020
- 2020-08-26 CN CN202010870029.0A patent/CN112149959B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103336885A (en) * | 2013-06-03 | 2013-10-02 | 北京航空航天大学 | Method for solving weapon-target assignment problem based on differential evolution algorithm |
| CN110096822A (en) * | 2019-05-08 | 2019-08-06 | 北京理工大学 | Multi-platform cooperative dynamic task allocation method under a kind of condition of uncertainty |
| CN111582553A (en) * | 2020-04-17 | 2020-08-25 | 上海机电工程研究所 | Information and firepower integrated task planning method |
Non-Patent Citations (3)
| Title |
|---|
| An Efficient Marginal-Return-Based Constructive Heuristic to Solve the Sensor-Weapon-Target Assignment Problem;Bin Xin 等;《IEEE TRANSACTIONS ON SYSTEMS, MAN, AND CYBERNETICS: SYSTEMS》;20191231;第49卷(第12期);第2536-2547页 * |
| 多阶段传感器–武器–目标分配问题的建模与优化求解;王艺鹏 等;《控制理论与应用》;20191130;第36卷(第11期);第1886-1895页 * |
| 武器目标分配问题的离散差分进化算法;张春美 等;《北京理工大学学报》;20140331;第34卷(第3期);第289-293、321页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112149959A (en) | 2020-12-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109218440B (en) | Dynamic scheduling method for heterogeneous executive bodies of scene simulation web server | |
| CN108833049B (en) | Deception anti-interference method and device based on cognition in unmanned aerial vehicle network | |
| CN112149959B (en) | Distributed sensor-weapon-target joint allocation method | |
| US11443221B2 (en) | Distributed incorruptible accordant management of nonlocal data fusion, unified scheduling and engage-ability | |
| CN113485456B (en) | Unmanned aerial vehicle group distributed online self-adaptive task planning method | |
| CN104950673A (en) | Method for distributing targets cooperatively attacked by unmanned aerial vehicle group | |
| CN112649002A (en) | Wild protection unmanned aerial vehicle monitoring system based on edge calculation | |
| CN113109770B (en) | Interference resource allocation method and system | |
| CN108876086A (en) | A kind of multiple no-manned plane method for allocating tasks based on particle swarm algorithm | |
| CN115333825A (en) | Defense method aiming at gradient attack of federal learning neurons | |
| CN111061995B (en) | Combat resource allocation method, first equipment and second equipment | |
| CN110163519B (en) | UUV red and blue threat assessment method for base attack and defense tasks | |
| CN112642161A (en) | Cheating detection and model training method and device for shooting game and storage medium | |
| CN110096822A (en) | Multi-platform cooperative dynamic task allocation method under a kind of condition of uncertainty | |
| CN101893441A (en) | Unmanned aerial vehicle flight path optimization method based on deviation maximization and grey correlation analysis | |
| CN111142439A (en) | Fire indicating integrated control method and system based on cloud computing | |
| CN113532193A (en) | Intelligent combat confrontation training system and method for team tactics | |
| CN111797966B (en) | Multi-machine collaborative global target distribution method based on improved flock algorithm | |
| CN111083106B (en) | Detection method and detection system for attack robot in multi-robot network | |
| CN116389093A (en) | Method and system for defending Bayesian attack in federal learning scene | |
| CN113792985A (en) | Sensor-weapon dynamic joint task multi-target allocation method | |
| Pang et al. | Risk-based sensor scheduling for target detection | |
| CN110163502A (en) | A kind of more bullet collaboration multistage target assignment methods | |
| Qiu et al. | MT-MTD: Muti-training based moving target defense trojaning attack in edged-AI network | |
| CN108174055B (en) | Intelligent monitoring method, system, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |