CN-122018557-A - Unmanned aerial vehicle cluster weapon target cooperative allocation method and system under space-time constraint

CN122018557ACN 122018557 ACN122018557 ACN 122018557ACN-122018557-A

Abstract

The unmanned aerial vehicle cluster weapon target collaborative distribution method and system under space-time constraint are provided, a collaborative distribution object is defined, a collaborative combat scene objective function is constructed, a collaborative combat scene constraint condition is defined, finally, the objective function is solved by utilizing a polygenic population parallel ant colony algorithm, wherein weapon sets, target sets and unmanned aerial vehicle sets are respectively encoded into weapon gene sequences, target gene sequences and unmanned aerial vehicle gene sequences, an initial population is generated, the cross mutation operation of an information guiding mechanism and a genetic algorithm of a fusion ant colony algorithm is used for carrying out phased joint optimization, an optimal collaborative distribution scheme is obtained by searching under the constraint condition, unmanned aerial vehicle task distribution results can be better obtained and optimized in a collaborative distribution mode, and unmanned aerial vehicle collaborative distribution efficiency is improved.

Inventors

DENG HENG
SHEN BOYANG
ZHANG LIGUO

Assignees

北京工业大学

Dates

Publication Date: 20260512
Application Date: 20260212

Claims (10)

1. A method for collaborative allocation of unmanned aerial vehicle cluster weapon targets under space-time constraints, the method comprising: s10, defining a cooperative allocation object, wherein the cooperative allocation object comprises an unmanned plane set, a target set and a weapon set; S20, constructing a collaborative combat scene objective function, wherein the collaborative combat scene objective function is used for simultaneously optimizing four targets, namely a target survival probability index, an unmanned aerial vehicle total flight range index, a task planning time index for completing a task and an unmanned aerial vehicle survival probability index; S30, defining a collaborative combat scene constraint condition, wherein the constraint condition at least comprises a time constraint, a space constraint and an unmanned aerial vehicle capability constraint; S40, solving the objective function by utilizing a polygenic population parallel ant colony algorithm, wherein a weapon set, a target set and an unmanned aerial vehicle set are respectively encoded into a weapon gene sequence, a target gene sequence and an unmanned aerial vehicle gene sequence to generate an initial population, carrying out phased joint optimization by means of an informative guide mechanism fusing the ant colony algorithm and a crossover variation operation of a genetic algorithm, and searching under the condition of meeting the constraint condition to obtain an optimal collaborative distribution scheme among the unmanned aerial vehicle, the weapon and the target; and S50, in the task execution process, if the unmanned aerial vehicle is detected to be invalid, triggering a dynamic reassignment strategy, and only locally re-optimizing the affected target and the rest unmanned aerial vehicles to generate a new collaborative assignment scheme.
2. The method of claim 1, wherein the collaborative combat scene objective function is represented by the following formula: ; Wherein, the An indicator of the probability of survival of the target, Represents the total course index of the unmanned aerial vehicle, The time index of the task planning is represented, The survival probability index of the unmanned aerial vehicle is represented, Weight values respectively representing the target survival probability, the unmanned aerial vehicle total range, the task planning time and the unmanned aerial vehicle survival probability, and ; , The probability of survival of the target is indicated, , The maximum range of the unmanned aerial vehicle is represented, The total number of the unmanned aerial vehicle is represented, Representing the total number of targets, Representing the total number of weapons, Representing weapon type For the target Is used for the treatment of the fracture of the steel plate, Representing allocation to targets A kind of electronic device The number of weapons to be weapons, A weight representing the target j; , The total course of the unmanned aerial vehicle is represented, ; Represent the first Target slave unmanned plane To the target Is used for the distance of euclidean distance, Representing a binary decision variable; , the time of the task planning is indicated, , Represent the first Target slave unmanned plane Flying to target Is used for the time of flight of (a), Representing objects The time window that is in possession of, Representing the cruising speed of the unmanned plane; , The survival probability of the unmanned aerial vehicle is represented, , The maximum range of the unmanned aerial vehicle is represented, Representing unmanned aerial vehicle quilt targets Probability of damage.
3. The method of claim 1, wherein the collaborative combat scene constraint is expressed by the following formula: time constraint conditions: ; ; ; ; Wherein, the Represent the first Unmanned aerial vehicle reaches target Is used for the time period of (a), Representing a specified task start time; Represent the first Target slave unmanned plane Flying to target Is used for the time of flight of (a), Represent the first The flight time of the individual drone from the start point to the target i, Representing objects The time window that is in possession of, Representing a binary decision variable, Indicating that the ith unmanned aerial vehicle is no later than The task is completed and the task is completed, Representing the p-th unmanned aerial vehicle slave target Flying to target Is a binary decision variable on the flight path of (a), Representing the flight of a p-th drone from a start point to a target Binary decision variables on the flight path of (a); Space constraint conditions: ; ; Wherein, the And All represent that the p-th unmanned aerial vehicle takes off from the starting point and needs to fly back to the starting point again after the execution task is finished; unmanned aerial vehicle ability constraint: ; ; Wherein, the Represent the first Allocation of unmanned aerial vehicle to attack target Is a function of the number of weapons, Indicating the maximum ammunition quantity of the unmanned aerial vehicle, The maximum range of the unmanned aerial vehicle is represented, Represent the first Target slave unmanned plane To the target Is used for the distance of euclidean distance, Representing a binary decision variable.
4. The method according to claim 1, wherein the initial population in step S4 is obtained by: Encoding the weapon set, the target set and the unmanned aerial vehicle set into a weapon gene sequence, a target gene sequence and an unmanned aerial vehicle gene sequence respectively; and executing random deletion and replacement operation on the target gene sequence, executing random sequencing on the unmanned aerial vehicle gene sequence, and generating an initial population.
5. The method according to claim 1, wherein step S4 comprises: optimizing the unmanned aerial vehicle gene sequence based on the constraint condition to determine the task access sequence of each unmanned aerial vehicle; fixing the assignment relation between the unmanned aerial vehicle and the target, and combining the unmanned aerial vehicle gene sequence and the target gene sequence into a multidimensional gene sequence; optimizing weapon gene sequences for the multidimensional gene sequences in combination with weapon parameters of each weapon; fusing the optimized parallel gene sequences, performing Pareto non-dominant sorting on candidate solutions by using the objective function, screening elite solutions, and updating pheromones based on the elite solutions; And directly transmitting each generation of elite solution to the next generation of population, updating the population through crossover and mutation operation, and performing iterative optimization until convergence conditions are met.
6. The method of claim 1, wherein the pheromone is updated by: ; ; Wherein, the A pheromone representing the execution path from the target i to the target j; The evaporation rate is indicated by the expression, Indicating the initial pheromone volatilization rate, Represents the final pheromone volatilization rate and satisfies 0< < <1;T denotes the number of iteration algebra, And >0 is a decay rate parameter for controlling the transition rate of the volatility from the exploration phase to the development phase.
7. The method of claim 6, wherein the polygenic population parallel ant colony algorithm Pareto non-dominant ordering candidate solutions in each iteration generation, retaining non-dominant solutions as elite solution sets, and performing a pheromone enhancement operation based on the elite solution sets, wherein a pheromone increment delta # Inversely proportional to the overall target performance of the solution.
8. The method of claim 1, wherein the dynamic reassignment policy comprises: judging that the unmanned aerial vehicle fails based on the timeout of the heartbeat signal; determining a target set, which has the overlapping degree with the original allocation target of the invalid unmanned aerial vehicle on a time window larger than a preset threshold value or smaller than a preset radius on a space distance, as a space-time adjacent target subset; And forming a local model based on the space-time adjacent target subset and the rest available unmanned aerial vehicle, and limiting the number of solving iterations to ensure that a new allocation scheme is output within the response time allowed by combat.
9. The method of claim 1, wherein when an add-on drone joins a drone cluster, the method further comprises: the unmanned aerial vehicle with the current task load higher than a preset threshold is identified as a task migration candidate, and the newly added unmanned aerial vehicle and the candidate unmanned aerial vehicle are formed into a local optimized subset; And on the premise of meeting the constraint condition, migrating partial targets corresponding to the candidate unmanned aerial vehicles to the newly-added unmanned aerial vehicle, and generating an increment allocation instruction only comprising migration tasks, wherein the rest unmanned aerial vehicles maintain the original flight tasks unchanged.
10. A unmanned aerial vehicle cluster weapon target cooperative allocation system under space-time constraints, the system comprising: the collaborative distribution object definition module is used for defining a collaborative distribution object, wherein the collaborative distribution object comprises an unmanned plane set, a target set and a weapon set; The objective function construction module is used for constructing a collaborative combat scene objective function, wherein the collaborative combat scene objective function is used for simultaneously optimizing four targets, namely a target survival probability index, an unmanned aerial vehicle total flight range index, a task planning time index for completing a task and an unmanned aerial vehicle survival probability index; The constraint condition definition module is used for defining a collaborative combat scene constraint condition, and the constraint condition at least comprises a time constraint, a space constraint and an unmanned aerial vehicle capability constraint; the coordination distribution module is used for solving the objective function by utilizing a polygenic population parallel ant colony algorithm, wherein a weapon set, a target set and an unmanned aerial vehicle set are respectively encoded into a weapon gene sequence, a target gene sequence and an unmanned aerial vehicle gene sequence to generate an initial population, and the initial population is subjected to phased joint optimization through an pheromone guidance mechanism fusing the ant colony algorithm and a crossover variation operation of a genetic algorithm, so that an optimal collaborative distribution scheme among the unmanned aerial vehicle, the weapon and the target is obtained by searching under the condition of meeting the constraint condition; and the dynamic reassignment module is used for triggering a dynamic reassignment strategy if the unmanned aerial vehicle is detected to be invalid in the task execution process, and only carrying out local re-optimization on the affected target and the rest unmanned aerial vehicles to generate a new collaborative assignment scheme.

Description

Unmanned aerial vehicle cluster weapon target cooperative allocation method and system under space-time constraint Technical Field The invention belongs to the technical field of unmanned aerial vehicle planning, and particularly relates to a unmanned aerial vehicle cluster weapon target collaborative distribution method and system under space-time constraint. Background Unmanned aerial vehicle technology has rapidly developed over the past few years and has been widely used in a number of fields such as military, civilian, scientific research and commerce. With advances in unmanned aerial vehicle technology and cost reduction, unmanned aerial vehicle cluster technology is receiving wide attention. The unmanned aerial vehicle cluster means that a plurality of unmanned aerial vehicles realize the completion of one or more specific tasks through collaborative work. In conventional unmanned aerial vehicle operation, a single unmanned aerial vehicle may be limited by a number of factors, such as battery duration, flight speed, payload, etc., when it is to accomplish a particular task. The unmanned aerial vehicle clustering technology can fully exert the advantages of the clusters through the cooperative work of a plurality of unmanned aerial vehicles, and improves the efficiency and the success rate of tasks. Unmanned aerial vehicle cluster combat mission planning is an important topic of great concern in the military field. With the complexity and diversification of modern war environments, conventional combat modes have failed to meet the requirements for rapid acquisition and real-time monitoring of enemy conditions, topography and battlefield situations. The unmanned aerial vehicle is used as a flexible, hidden and efficient combat platform, and provides brand new possibility for the execution of military tasks. Firstly, the unmanned aerial vehicle has excellent reconnaissance monitoring capability, can execute various reconnaissance tasks on a battlefield, acquire important information such as enemy, topography, meteorological and the like, and provides reliable information support for combat command decision. Secondly, the unmanned aerial vehicle can execute the accurate striking task, and meanwhile, the casualties of friends and citizens can be reduced to the greatest extent under the condition of keeping the combat effort. The traditional unmanned aerial vehicle task allocation method is mainly realized through two parts, namely target allocation and weapon allocation. The goal allocation usually uses a Multiple TRAVELING SALESMAN Problem (MTSP) model, and the constraints such as unmanned plane range and time need to be considered in the allocation process, and the task sequence is reasonably planned. In the aspect of weapon distribution, a weapon-target distribution (Weapon TARGET ASSIGNMENT, abbreviated as WTA) problem model is mainly used, the types of weapons are modeled as a mathematical problem for different damage probabilities of targets, and the mathematical problem is solved through an optimization algorithm. However, the allocation process in this way does not consider the constraint relation and the synergy between the two, and has certain defects. Disclosure of Invention In view of the above problems in the prior art, an object of the present invention is to provide a method and a system for collaborative allocation of targets of unmanned aerial vehicle clusters under space-time constraint, which can improve efficiency of collaborative allocation of unmanned aerial vehicles. In order to solve the technical problems, the specific technical scheme is as follows: In one aspect, provided herein is a method of collaborative allocation of unmanned aerial vehicle cluster weapon targets under space-time constraints, the method comprising: s10, defining a cooperative allocation object, wherein the cooperative allocation object comprises an unmanned plane set, a target set and a weapon set; S20, constructing a collaborative combat scene objective function, wherein the collaborative combat scene objective function is used for simultaneously optimizing four targets, namely a target survival probability index, an unmanned aerial vehicle total flight range index, a task planning time index for completing a task and an unmanned aerial vehicle survival probability index; S30, defining a collaborative combat scene constraint condition, wherein the constraint condition at least comprises a time constraint, a space constraint and an unmanned aerial vehicle capability constraint; S40, solving the objective function by utilizing a polygenic population parallel ant colony algorithm, wherein a weapon set, a target set and an unmanned aerial vehicle set are respectively encoded into a weapon gene sequence, a target gene sequence and an unmanned aerial vehicle gene sequence to generate an initial population, carrying out phased joint optimization by means of an informative guide mechanism fusing the ant colony algo