A dynamic target assignment method suitable for anti-UAV swarm control

Abstract

The application discloses a dynamic target allocation method suitable for anti-UAV group control, input target information and conversion to obtain the interception mode set of each target, genetic algorithm is used to determine the target priority and the preferred interception candidate mode of each target; the target priority is in descending order, and each target is tried to be added to the scheduling diagram according to the preferred interception candidate mode; for the target that is not successfully added, each target is considered in turn according to the priority from large to small, for each target, various interception modes of the target are tried in turn according to the order of entering the interception range from early to late, until the target is successfully added to the scheduling diagram or all modes are tried; and the allocation scheme is generated according to the scheduling diagram. The genetic algorithm of the application only needs to search the target priority and the preferred interception mode, and then uses the scheduling diagram to greedily construct the scheduling scheme, so that the discretization need of dividing time into segments is got rid of, and the precision and efficiency are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of firepower and command control, in particular to a dynamic target assignment method suitable for anti-UAV group control. BACKGROUND

[0002] In the field of firepower and command control, there is a problem of weapon scheduling, which is called weapon-target assignment. The traditional static weapon-target assignment (SWTA) scheme cannot meet the needs of rapidly changing battlefields. The static weapon-target assignment is based on the assumption that all weapons are assigned to be launched at the same time, without considering time, and cannot plan future assignments. The ideal dynamic weapon-target assignment (DWTA) unfolds the firepower resources in the time dimension, allowing the use of future firepower resources and considering future threat target assignment strategies. Dynamic weapon-target assignment is more suitable for actual combat scenarios, does not blindly consider current assignments, focuses on the global, and can achieve better overall combat capability. The traditional weapon-target assignment method has the defect of being difficult to adapt to large-scale, especially when facing dynamic weapon-target assignment problems that consider time. Due to the increase in optimization variables, traditional methods are difficult to effectively handle. SUMMARY

[0003] The purpose of the present application is to provide a dynamic target assignment method suitable for anti-UAV group control, which solves the problem that the existing weapon-target assignment method has the defect of being difficult to adapt to large-scale, especially when facing dynamic weapon-target assignment problems that consider time. Due to the increase in optimization variables, traditional methods are difficult to effectively handle.

[0004] The present application solves the above problems through the following technical solutions:

[0005] A dynamic target assignment method suitable for anti-UAV group control, comprising:

[0006] Step S10, inputting the position, speed, estimated processing time of being simultaneously intercepted by different numbers of weapons, position of the intercepting weapon, and intercepting radius of the target;

[0007] Step S20, converting the input information to obtain a set of interception methods for each target, the target interception method including the used interceptor, the time of entering the interception range, the time of leaving the interception range, and the estimated duration required for interception;

[0008] Step S30, determining the target priority and the preferred candidate interception method for each target using a genetic algorithm;

[0009] Step S40, according to the target priority obtained by the genetic algorithm, the target is tried to be added to the scheduling graph in the preferred intercepted candidate mode in the order from large to small; for the target that fails to be added, each target is considered in turn according to the priority from large to small, and for each target, various interception modes of the target are tried in turn according to the order from early to late in the entering interception range, until the target is successfully added to the scheduling graph or all modes are tried;

[0010] Step S50, the allocation scheme is generated according to the scheduling graph.

[0011] The present application takes time constraint as the primary consideration, and the actually feasible interception mode is pre-calculated in the calculation process of the optimal allocation mode, which makes the search of the genetic algorithm more effective. Most of the existing methods for solving dynamic weapon allocation need to divide time into segments and discretize time. The finer the time division, the higher the precision of the scheduling scheme. However, relatively, in order to achieve the same effect, the solving time will increase due to the expansion of the search space. The genetic algorithm part of the present application only needs to search the target priority and the preferred interception mode, and then uses the scheduling graph to greedily construct the scheduling scheme. This process is free from the discretization requirement of dividing time into segments, so there is no need to worry about the accuracy of the scheme, and the generated scheduling time can be a real value. Based on the same reason, since only the target priority and the preferred interception mode of each target need to be iteratively searched, compared with directly optimizing the search of the interception start and end time, the search space is greatly reduced, and the search efficiency is improved.

[0012] Further, the method for converting the input information to obtain the interception mode set of each target in step S20 is:

[0013] Step S21, the time when the target enters and leaves the interception range of each interceptor is calculated according to the geometric relationship;

[0014] Step S22, for the case that the target is intercepted by k interceptors at the same time, the set of interceptors corresponding to the interception range of the target path is denoted as W, and various combinations of selecting k interceptors from W are enumerated to form a set S;

[0015] Step S23, the intersection of the time range of the interception range of each interceptor in the set S is calculated, and if the length of the intersection is not less than the expected interception duration of the k interceptors intercepting the target at the same time, the interception mode is added to the interception mode set of the target.

[0016] Further, the step S30 specifically includes:

[0017] Step S31, randomly generate multiple groups of solutions X = [target_priority, first_option] to form a population, wherein target_priority = [p_i | p_i ∈ R, 1≤i≤|T|], first_option = [o_i | 1≤o_i≤|O_i|, 1≤i≤|T|], p_i represents the priority of the i th target, o_i represents the serial number of the first preferred intercepted candidate mode of the i th target, |O_i| represents the size of the intercepted candidate mode set O_i of the i th target, and |T| represents the number of targets;

[0018] Step S32, construct a scheduling graph for all solutions in the population and calculate the fitness =∑1[i∈S']*(threat_i-ε*(endTime_i-minTime) / (maxTime-minTime)), wherein S' is a serial number set of threat targets successfully added to the scheduling graph, 1[i∈S'] is 1 when the i th target is successfully added to the scheduling graph, and 0 otherwise; threat_i is the threat degree of the i th target, endTime_i is the time when the i th target is successfully intercepted, minTime represents the minimum start interception time in the candidate mode of all targets, maxTime represents the maximum interception completion time in the candidate mode of all targets, and ε is the weight of the completion time penalty term;

[0019] Step S33, resample and select the solutions in the population according to the fitness to obtain a new population with the same size as the original population;

[0020] Step S34, randomly combine the solutions in the new population in pairs, and exchange the contents of the two solutions in the combination at random positions;

[0021] Step S35, randomly replace the values of the elements of the solutions in the new population with new values according to a preset probability;

[0022] Step S36, repeat steps S32 to S35 until a set number of iterations is reached or the fitness no longer improves within a preset time period.

[0023] Further, the method for constructing the scheduling graph is:

[0024] a. Construct an initial scheduling graph structure, including a simple directed graph and a same number of single-directional linked lists as the weapons, the simple directed graph describes the occurrence sequence of the interception events in the global, each single-directional linked list maintains the sequence of the corresponding weapon intercepting each target, the nodes in the simple directed graph and the nodes in the linked list set are in a one-to-many relationship, initially, the simple directed graph has a same number of source points as the weapons and a sink point, the source points point to the sink point, and each linked list has a head node and a tail node;

[0025] b. add the candidate interception mode of the target into the simple directed graph and the linked list in order of priority from large to small, and skip if failed;

[0026] c. add the target failed in b into the simple directed graph and the linked list in order of priority from large to small, and for each target, add the candidate interception mode of the target into the simple directed graph and the linked list in order of time from early to late, and consider the candidate interception mode of the next target if the addition is successful;

[0027] d. during the topological sorting of the simple directed graph, update the earliest start time of the interception mode of the target corresponding to each node on the simple directed graph as:

[0028] max{max_i{prevNode_i_earliest+prevNode_i_duration},node_enter}

[0029] wherein prevNode_i_earliest is the earliest start time of the interception mode of the target corresponding to the i-th predecessor node, prevNode_i_duration is the interception duration required by the interception mode of the target corresponding to the i-th predecessor node, and node_enter is the time when the target corresponding to the node enters the interception range;

[0030] e. during the inverse topological sorting of the simple directed graph, update the latest end time of the interception mode of the target corresponding to each node on the simple directed graph as:

[0031] min{min_i{nextNode_i_latest}-node_duration,node_leave},

[0032] wherein nextNode_i_latest is the latest end time of the interception mode of the target corresponding to the i-th successor node, nextNode_i_duration is the interception duration required by the interception mode of the target corresponding to the i-th successor node, and node_leave is the time when the target corresponding to the node leaves the interception range.

[0033] Further, the specific method for adding the candidate interception mode of the target into the simple directed graph and the linked list in b and c is:

[0034] 1) find the insertion position set on each linked list corresponding to the weapon required by the interception mode, and the maximum time range of the insertion position on each linked list is:

[0035] [prevNode_earliest+prevNode_duration,nextNode_latest–nextNode_duration], where prevNode_earliest is the earliest start time of the interception mode of the target corresponding to the predecessor node, nextNode_latest is the latest end time of the interception mode of the target corresponding to the successor node, prev / nextNode_duration is the interception duration required by the interception mode of the target corresponding to the predecessor / successor node, the intersection of the maximum time range in the insertion position set is denoted as [earliest, latest], and the found insertion position set should satisfy:

[0036] min{latest,node_leave}-max{earliest,node_enter}>node_duration;

[0037] 2) inserting the node in the linked list according to the selected insertion position set, and inserting the node into the graph at the corresponding position to establish the correspondence between the graph node and the linked list node.

[0038] Compared with the prior art, the present application has the following advantages and beneficial effects:

[0039] (1) In the present application, time constraints are taken as the primary consideration, and the actually feasible interception mode is pre-calculated in the calculation process of the optimal allocation mode, which makes the search of the genetic algorithm more effective. The genetic algorithm of the present application only needs to search the target priority and the first interception mode, and then uses the scheduling graph to greedily construct the scheduling scheme. This process is free from the discretization requirement of dividing time into segments, and therefore does not need to worry about the accuracy of the scheme, and the generated scheduling time can be a real value. Compared with directly optimizing the search of the interception start and end time, the search space is greatly reduced, and the efficiency of dynamic target allocation is improved.

[0040] (2) The algorithm of the present application uses the scheduling graph to maintain the scheduling scheme, and greedily constructs the scheduling scheme on the scheduling graph through the target priority and the first interception mode, so as to complete the fitness evaluation of the solution in the genetic algorithm. BRIEF DESCRIPTION OF DRAWINGS

[0041] Figure 1 It is a schematic diagram of the interception candidate mode to be allocated in the present application;

[0042] Figure 2 It is the order of the target interception mode when the scheduling graph is constructed in the present application;

[0043] Figure 3 It is a schematic diagram of the correspondence between the data structure and an actual scheduling scheme in the present application;

[0044] Figure 4 Figure 1 is a diagram showing the change of the algorithm runtime optimization target in the present application with the number of iterations. DETAILED DESCRIPTION

[0045] The present application will be further described in detail below with reference to the examples, but the embodiments of the present application are not limited thereto.

[0046] Embodiment:

[0047] A dynamic target allocation method suitable for anti-UAV swarm control, comprising:

[0048] Step S10, inputting the position, speed, estimated processing time of simultaneous interception by different numbers of weapons, position of the interception weapon, interception radius, working state of the target, and position, radius of the protection zone;

[0049] Step S20, converting the input information to obtain a set of interception methods for each target, the target interception method including the used interceptor, the time of entering the interception range, the time of leaving the interception range, and the estimated duration required for interception;

[0050] Step S30, determining the target priority and the preferred candidate interception method for each target by using a genetic algorithm;

[0051] Step S40, according to the target priority obtained by the genetic algorithm in descending order, attempting to add the target to the scheduling diagram according to the preferred candidate interception method one by one, as shown in Figure 2 The order of the target interception method is considered when constructing the scheduling diagram; for the targets that are not successfully added, the priority is considered again in descending order, and each target is considered in turn, and for each target, the various interception methods of the target are continuously attempted in the order of entering the interception range from early to late, until the target is successfully added to the scheduling diagram or all methods are tried;

[0052] Step S50, generating an allocation scheme according to the scheduling diagram.

[0053] As shown in Figure 1 , there is an interception radius for the interceptor, and the target can only be intercepted when it enters this range. During the interception process, the interceptor needs to track the target until the target is destroyed or leaves the interception range. Figure 1The right half of the figure shows the many-to-one case, and the scope of the interception processing mode is the intersection of the two interception ranges, and the two interceptors simultaneously track and intercept a target. Using multiple interceptors to intercept a target can spend less time to complete the interception. The task of the present application is to solve the problem of how to reasonably allocate multiple targets to multiple interceptors, use which interception mode, and when to intercept. The present application proposes to use a genetic algorithm to determine the order of considering targets and the optimal processing mode of the target priority, and based on this order, each target and the corresponding processing mode are considered in turn, and the entire scheduling allocation scheme is constructed from zero. The scheduling scheme is maintained using a data structure during the construction process. As shown in Figure 3 the scheduling diagram includes a simple directed graph and multiple linked lists, and a scheduling scheme above is described by two data structures below. The final scheduling scheme is finally constructed by continuously inserting nodes in the data structure.

[0054] Further, the method for converting the input information to obtain the interception mode set of each target in step S20 is:

[0055] Step S21, calculate the time when the target enters and leaves the interception range of each interceptor according to the geometric relationship, that is, the interception time window;

[0056] Step S22, for the case that the target is simultaneously intercepted by k interceptors, record the set of interceptors corresponding to the interception range of the target path as W, enumerate various combinations of selecting k interceptors from W to form a set S;

[0057] Step S23, calculate the intersection of the time range of the interception range of each interceptor in set S, if the length of the intersection is not less than the expected interception duration of k interceptors simultaneously intercepting the target, then add this interception mode to the interception mode set of the target.

[0058] Further, the step S30 specifically includes:

[0059] Step S31, randomly generate multiple groups of solutions X=[target_priority, first_option] to form a population, where target_priority=[p_i|p_i∈R,1≤i≤|T|], first_option=[o_i|1≤o_i≤|O_i|,1≤i≤|T|], p_i represents the priority of the i-th target, o_i represents the serial number of the first preferred candidate mode of the i-th target, |O_i| represents the size of the candidate mode set O_i of the i-th target, and |T| represents the number of targets;

[0060] Step S32, all solutions in the population are used to construct a schedule graph, and the fitness is calculated as follows: fitness =∑1[i∈S']*(threat_i-ε*(endTime_i-minTime) / (maxTime-minTime)), where S' is the set of the serial numbers of the threat targets successfully added to the schedule graph, 1[i∈S'] is 1 when the ith target is successfully added to the schedule graph, otherwise 0; threat_i is the threat degree of the ith target, endTime_i is the time when the ith target is successfully intercepted, minTime is the minimum start-interception time of all candidate modes of the targets, maxTime is the maximum interception-completion time of all candidate modes of the targets, and ε is the weight of the completion-time penalty term;

[0061] Step S33, the solutions in the population are resampled and selected according to the fitness, and a new population with the same size as the original population is obtained;

[0062] Step S34, the solutions in the new population are randomly combined two by two, and the contents of the two solutions in the combination are exchanged at random positions;

[0063] Step S35, the elements of the solutions in the new population are randomly replaced with new values according to a preset probability;

[0064] Step S36, steps S32 to S35 are repeated until a preset number of iterations is reached or the fitness no longer improves within a preset time period.

[0065] Further, the method for constructing the schedule graph is as follows:

[0066] a. An initial schedule graph structure is constructed, including a simple directed graph and a same number of single-directional linked lists as the weapons, the simple directed graph describes the sequence of the interception events in the global, each single-directional linked list maintains the sequence of the corresponding weapon intercepting each target, the nodes in the simple directed graph and the nodes in the linked list set are in a one-to-many relationship, initially, the simple directed graph has a same number of source points as the weapons and a sink point, the source points point to the sink point, and each linked list has a head node and a tail node;

[0067] b. The candidate interception modes of the targets are added to the simple directed graph and the linked lists in the order of the priority from large to small, and if the addition fails, the target is skipped;

[0068] c. The targets whose addition fails in b are considered to be added to the simple directed graph and the linked lists in the order of the priority from large to small, for each target, the candidate interception modes of the target are added to the simple directed graph and the linked lists in the order of the entering-interception-range time from early to late, if the addition is successful, the candidate interception modes of the next target are considered;

[0069] d. In the process of topological sorting of the simple directed graph, the earliest start time of the interception mode of the target corresponding to each node on the simple directed graph is updated as:

[0070] max{max_i{prevNode_i_earliest+prevNode_i_duration},node_enter}

[0071] Where prevNode_i_earliest is the earliest start time of the interception mode of the target corresponding to the ith predecessor node, prevNode_i_duration is the interception duration required by the interception mode of the target corresponding to the ith predecessor node, and node_enter is the time when the target corresponding to the node enters the interception range.

[0072] e. In the process of reverse topological sorting of the simple directed graph, the latest end time of the interception mode of the target corresponding to each node on the simple directed graph is updated as:

[0073] min{min_i{nextNode_i_latest}-node_duration,node_leave},

[0074] Where nextNode_i_latest is the latest end time of the interception mode of the target corresponding to the ith successor node, nextNode_i_duration is the interception duration required by the interception mode of the target corresponding to the ith successor node, and node_leave is the time when the target corresponding to the node leaves the interception range.

[0075] Further, the specific method of adding candidate interception modes of the target to the simple directed graph and the linked list in b and c is:

[0076] 1) Find the insertion position set on each linked list corresponding to the weapon required by the interception mode, and the maximum time range of the insertion position on each linked list is:

[0077] [prevNode_earliest+prevNode_duration,nextNode_latest-nextNode_duration], where prevNode_earliest is the earliest start time of the interception mode of the target corresponding to the predecessor node, nextNode_latest is the latest end time of the interception mode of the target corresponding to the successor node, prev / nextNode_duration is the interception duration required by the interception mode of the target corresponding to the predecessor / successor node, and the intersection of the maximum time range in the insertion position set is denoted as [earliest,latest], and the found insertion position set should satisfy:

[0078] min{latest,node_leave} - max{earliest,node_enter} > node_duration;

[0079] 2), insert the node in the linked list according to the selected insertion position set, and insert the node into the graph at the corresponding position, to establish the correspondence between the graph node and the linked list node.

[0080] As shown in Figure 4 , the process of gradually improving the optimization target value of the algorithm with population iteration in one scene is shown. The scene setting is as follows:

[0081] 1, 50 targets are randomly generated in the range of [0, 200] 2 , fly to the fire area, the speed is randomly generated in [10, 50], the threat degree is randomly generated in [1, 10], and it is divided into four types according to the different expected interception time of the target:

[0082] a) the expected time required for simultaneous interception by 1 / 2 / 3 interceptors is 10 / 7 / 5;

[0083] b) the expected time required for simultaneous interception by 1 / 2 / 3 interceptors is 7 / 4 / 3;

[0084] c) the expected time required for simultaneous interception by 1 / 2 / 3 interceptors is 6 / 3 / 2;

[0085] d) the expected time required for simultaneous interception by 1 / 2 / 3 interceptors is 22 / 15 / 9;

[0086] 2, 4 interceptors, interception radius is 1000, position is respectively in [1800, 1800], [1800, 2200], [2200, 1800], [2200, 2200];

[0087] The population size of genetic algorithm is set to 50, the crossover probability is set to 0.5, and the mutation probability is set to 0.01. Figure 4 It can be seen from the figure that the algorithm converges at about 160 iterations. The algorithm used in the comparison also uses the scheduling graph described in the invention to construct the scheduling scheme, but replaces the genetic algorithm with a greedy strategy, that is, considering each target in order of threat degree from large to small, and considering the interception processing mode of each target according to the first-come-first-served principle. As can be seen from the figure, after optimization by genetic algorithm, better results can be obtained than based on the simple rules in this example.

[0088] In conclusion, the hybrid algorithm scheme combines the global search ability of the genetic algorithm and the ability of quickly generating a scheduling scheme on a scheduling graph by using greediness, and is suitable for dynamic target allocation and scheduling of a laser system using thermal energy accumulation as a killing means.

[0089] Although the present application has been described with reference to the explanatory embodiments thereof, the above embodiments are merely the preferred embodiments of the present application, and the embodiments of the present application are not limited to the above embodiments, and it should be understood by those skilled in the art that many other modifications and embodiments can be designed, and these modifications and embodiments will fall within the scope and spirit of the principles disclosed in the present application.

Claims

1. A dynamic target assignment method suitable for counter-UAV swarm control, characterized in that, The method comprises the following steps: Step S10, inputting the position, speed, estimated processing time of simultaneous interception by different numbers of weapons, position of the interception weapon and interception radius of each target; Step S20, converting the input information to obtain a set of interception modes of each target, the interception mode of the target comprising a used interceptor, time of entering the interception range, time of leaving the interception range and duration of the estimated interception; Step S30, determining the target priority and the first preferred candidate interception mode of each target by using a genetic algorithm; Step S40, according to the target priority obtained by the genetic algorithm, attempting to add the target to the scheduling diagram according to the first preferred candidate interception mode in descending order of the target priority; for the target that fails to be added, considering each target in descending order of the target priority again, and for each target, attempting the interception mode of the target in ascending order of the time of entering the interception range until the target is successfully added to the scheduling diagram or all the modes are tried; Step S50, generating an allocation scheme according to the scheduling diagram; The method for converting the input information to obtain the set of interception modes of each target in the step S20 comprises the following steps: Step S21, calculating the time of entering and leaving the interception range of each interceptor according to geometric relationship; Step S22, for the case that the target is simultaneously intercepted by k interceptors, recording the set of interceptors corresponding to the interception range of the target as W, enumerating various combinations of selecting k interceptors from W to form a set S; Step S23, calculating the intersection of the time range of the interception range of each interceptor in the set S, and if the duration of the intersection is not less than the estimated interception duration of the k interceptors simultaneously intercepting the target, the interception mode is added to the set of interception modes of the target; The step S30 specifically comprises the following steps: Step S31, randomly generating multiple groups of solutions X=[target_priority, first_option] to form a population, wherein target_priority=[p_i|p_i∈R, 1≤i≤|T|], first_option=[o_i|1≤o_i≤|O_i|, 1≤i≤|T|], p_i represents the priority of the i-th target, o_i represents the serial number of the first preferred candidate interception mode of the i-th target, |O_i| represents the size of the set of candidate interception modes O_i of the i-th target, and |T| represents the number of targets; Step S32, all solutions in the population are constructed into a schedule graph, and fitness=∑1[i∈S']*(threat_i-ε*(endTime_i-minTime) / (maxTime-minTime)) is calculated, where S' is a set of serial numbers of threat targets successfully added into the schedule graph, 1[i∈S'] is 1 when the ith target is successfully added into the schedule graph, otherwise 0; threat_i is the threat degree of the ith target, endTime_i is the time when the ith target is successfully intercepted, minTime is the minimum start intercepting time in the candidate mode of all targets, maxTime is the maximum intercepting completion time in the candidate mode of all targets, and ε is the weight of the completion time penalty term; Step S33, the solutions in the population are resampled and selected according to fitness, and a new population with the same size as the original population is obtained; Step S34, the solutions in the new population are randomly combined in pairs, and the contents of the two solutions in the combination are exchanged at random positions; Step S35, the elements of the solutions in the new population are randomly replaced with new values according to a preset probability; Step S36, steps S32 to S35 are repeated until a set number of iterations is reached or the fitness does not improve within a preset time period.

2. The dynamic target assignment method for counter-UAV swarm control according to claim 1, wherein, The method for constructing the schedule graph is: a. Construct an initial schedule graph structure, including a simple directed graph and a same number of single-directional linked lists as the weapons, the simple directed graph describes the sequence of the occurrence of each intercepting event in the global, each single-directional linked list maintains the sequence of the corresponding weapon intercepting each target, the nodes in the simple directed graph and the nodes in the linked list set are in a one-to-many relationship, initially, the simple directed graph has a same number of source points as the weapons and a sink point, the source points point to the sink point, and each linked list has a head node and a tail node; b. Add the candidate intercepting modes of the targets into the simple directed graph and the linked lists in the order of the priority from large to small, and skip if failed; c. Add the targets in b that fail to be added into the simple directed graph and the linked lists in the order of the priority from large to small, for each target, add the candidate intercepting modes of the target into the simple directed graph and the linked lists in the order of the entering intercepting range time from early to late, and consider the candidate intercepting modes of the next target if the addition is successful; d. In the process of topological sorting of the simple directed graph, the earliest start time of the intercepting mode of each node corresponding to the target on the simple directed graph is updated as: max{max_i{prevNode_i_earliest+prevNode_i_duration},node_enter} where prevNode_i_earliest is the earliest start time of the intercepting mode of the ith predecessor node corresponding to the target, prevNode_i_duration is the intercepting duration required by the intercepting mode of the ith predecessor node corresponding to the target, and node_enter is the time when the node corresponding to the target enters the intercepting range. ​ ​ e、In the process of reverse topological sorting of the simple directed graph, the latest end time of the interception mode of the target corresponding to each node on the simple directed graph is updated as: min{min_i{nextNode_i_latest}-node_duration,node_leave}, where nextNode_i_latest is the latest end time of the interception mode of the target corresponding to the ith successor node, nextNode_i_duration is the interception duration required by the interception mode of the target corresponding to the ith successor node, and node_leave is the time when the target leaves the interception range of the node.

3. The dynamic target assignment method for counter-UAV swarm control according to claim 2, wherein, The specific method of adding the candidate interception mode of the target into the simple directed graph and the linked list in b and c is as follows: 1) find the insertion position set on each linked list corresponding to the weapon required by the interception mode, and the maximum time range of the insertion position on each linked list is: [prevNode_earliest+prevNode_duration,nextNode_latest-nextNode_duration], where prevNode_earliest is the earliest start time of the interception mode of the target corresponding to the predecessor node, nextNode_latest is the latest end time of the interception mode of the target corresponding to the successor node, and prev / nextNode_duration is the interception duration required by the interception mode of the target corresponding to the predecessor / successor node, and the intersection of the maximum time range in the insertion position set is denoted as [earliest,latest], and the found insertion position set should satisfy: min{latest,node_leave}-max{earliest,node_enter}>node_duration; 2) insert the node in the linked list according to the selected insertion position set, and simultaneously insert the node into the corresponding position in the graph, so as to establish the corresponding relationship between the graph node and the linked list node.

Images