A method for generating a jammer control strategy for networked radar false tracks

Abstract

The application discloses a method for generating a control strategy of an unmanned jammer for networked radars, and is applied to the technical field of electronic countermeasure, and aims at solving the problem that the existing technology cannot consider the implementation completeness of false tracks and the total driving distance of ECAVs simultaneously. First, a networked radar deception jamming implementation scene is constructed, then, discretization is performed on two dimensions of time and space, constraint conditions are derived, an ECAV control strategy optimization model is regarded as an optimal path search problem of a tree, finally, a single-player Monte Carlo tree search algorithm is combined, and the problem is iteratively optimized through four stages of selection, expansion, simulation and backtracking to obtain an optimal flight path. The method has better performance than existing algorithms in single false track scenes and multiple false track scenes.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of electronic countermeasures, and particularly relates to a control strategy planning technology for networked radar deception jamming of unmanned aerial vehicles. BACKGROUND

[0002] In the field of networked radar deception jamming, electronic combat air vehicles (ECAVs, also known as unmanned jammer clusters) are attracting much attention due to their controllability and flexibility. A large number of studies adopt a "many-to-many" jamming mode, in which ECAVs cooperatively control the delay forwarding of each station detection signal of the networked radar, while maintaining the correlation of each ECAV in time and space, so as to generate a false track for the networked radar and achieve track deception jamming of the networked radar. ECAVs are the control objects for implementing false track deception jamming, and by controlling the path and forwarding delay of the ECAVs, a non-existent false track is generated in the data center of the networked radar. When implementing deception jamming, the control factors of the ECAVs include two aspects: the motion trajectory and the forwarding delay. When the false track is determined, if the motion trajectory can be determined, the forwarding delay can be determined by subtracting the motion trajectory of the ECAVs from the false track. Therefore, the research on the control strategy of the ECAVs is usually directed to the motion trajectory of the ECAVs.

[0003] Under the background of false track deception jamming, there are different research ideas for the open research on ECAVs motion trajectory planning problem: (1) Starting from the perspective of constraint control, this type takes the control relationship between false track and ECAV as the starting point for research. By analyzing the motion relationship and spatial geometric relationship between the two, the feasible motion region or trajectory of ECAV is derived. The literature "Concepts for generating coherent radar phantom tracks using cooperating vehicles" explains the coupling between the false target and the ECAV in two-dimensional space, analyzes and derives the constraint conditions that need to be met from the perspective of motion relationship, and derives the feasible motion region of ECAV for uniform straight and uniform circular false tracks. The literature "Feasible flight paths for cooperative generation of a phantom radar track" analyzes the ECAV motion trajectory selection problem under the motion constraint, relaxes the ECAV dynamics constraint and spatial constraint, and relaxes the initial conditions of ECAV, providing more choices for ECAV trajectory planning (2) Starting from the optimization angle, this type of research starts from the optimization angle, takes the motion state and spatial geometric relationship as the constraint, and takes the specific strategic intention as the evaluation index, establishes a constraint optimization model, and solves the motion trajectory of ECAV. The literature "Online control strategies for highly coupled cooperative uavs" considers the speed and turning angle constraints of ECAV, takes minimizing the distance between ECAV and radar and the curvature of ECAV trajectory as the target, establishes an optimal control model, converts the problem into a constraint optimization problem with limited dimension of independent variable by parameterizing the control, and solves it by using the sequential quadratic programming method. The literature "Research on method of phantom track generation based on cooperative control of uavs" establishes an optimization model to reduce the total distance of ECAV operation, and after simplification, proposes an iterative algorithm.

[0004] In summary, for the ECAVs trajectory planning problem under the background of networked radar deception jamming, there are currently a large number of studies, but most of them consider single false target scenarios, and there is insufficient research in multi-false target scenarios. At the same time, current research lacks consideration of the multi-frame coupling relationship and time constraints of ECAVs and false tracks. Due to the cooperation of false tracks and protected targets in range gate decoy jamming, the implementation time deviation of false tracks will reduce the jamming performance. Moreover, most of the current research on ECAVs control optimization methods is focused on the modeling process, and there is less research on solution algorithms. Therefore, it is necessary to further deepen the research on ECAVs control strategy planning methods for networked radar track deception. SUMMARY

[0005] To solve the above technical problems, the present application provides an unmanned jammer control strategy generation method for networked radar false tracks, which can balance the implementation completeness of false tracks and the total travel distance of ECAVs.

[0006] The technical scheme adopted by the present application is as follows: an unmanned jammer control strategy generation method for networked radar false tracks, comprising:

[0007] S1, constructing a networked radar deception jamming implementation scenario and determining ECAVs motion constraints; the implementation scenario is as follows: for each radar site of the networked radar, at each time, a group of unmanned jammers needs to be located between the radar and the false target, and the ECAVs will transmit the echo with time delay to the radar, so that the radar detects the false track; the ECAVs motion constraints are that the ECAVs satisfy the condition that they are on the connecting line between the radar and the false track at each time, and satisfy the minimum safety distance constraint between the ECAVs and the radar, the minimum trailing distance constraint, the ECAVs speed constraint, the ECAVs acceleration constraint, and the ECAVs turning angle constraint;

[0008] S2, determining the constraints of the ECAVs control strategy optimization model by discretizing the time and space dimensions;

[0009] S3, completing the construction of the ECAVs control strategy optimization model; according to the state transition process of the ECAVs, a tree T{V, E} is established with the initial state of the ECAVs as the top vertex, wherein V is the vertex set of the tree, each vertex represents a state of the ECAVs cluster, and E is the edge set of the tree, each edge represents a state transition process of the ECAVs cluster; thereby converting the ECAVs control strategy optimization problem into an optimal path search problem in the tree T;

[0010]

[0011]

[0012] S4, search the tree T by using a single player-Monte Carlo tree search (SP-MCTS) algorithm to obtain an optimal control strategy.

[0013] The application has the beneficial effects that the application proposes a SP-MCTS-based deception jamming ECAVs control strategy planning method. Firstly, a networking radar deception jamming implementation scene is constructed, then a constraint relationship among ECAVs, false targets and radars is determined according to a networking radar deception jamming principle, and an ECAVs control strategy planning model is established by taking the ECAVs maneuvering capability as a constraint and considering the false track implementation completion degree and the ECAVs driving distance, the model is discretized in time and space dimensions, a tree structure is established according to the motion constraint of the ECAVs, the optimization model is transformed into an optimal path search problem of the tree structure, and finally the SP-MCTS is used for solving. The simulation results in a single false track scene show that the method can meet the constraint and completely implement the false track, in the case of strict constraint, the false track completion degree and the ECAVs driving distance are considered, and the completion degree is primarily considered, so that the effect of the deception jamming is ensured. The simulation results in a multiple false track scene show that the method can introduce more possibilities into path planning by considering the replacement of the matching relationship between the ECAVs and the false track at different times, so that a more optimal flight path is planned. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 It is a networking radar deception jamming implementation scene diagram.

[0015] Figure 2 It is a radar, ECAV, false target space position relationship diagram.

[0016] Figure 3 It is an unmanned aerial vehicle speed constraint diagram.

[0017] Figure 4 It is a single false target path planning scene diagram.

[0018] Figure 5 It is a single false target scene path planning tree structure diagram.

[0019] Figure 6 It is a multiple false target path planning scene diagram.

[0020] Among them, (a) is to maintain the matching relationship; (b) is to change the matching relationship.

[0021] Figure 7 It is a multiple false target scene path planning tree structure diagram.

[0022] Figure 8 Simulation results of UAV path planning for scenario a.

[0023] Figure 9 Simulation results of UAV path planning for scenario b.

[0024] Figure 10 Simulation results of UAV path planning for scenario c. DETAILED DESCRIPTION

[0025] The present application is mainly verified by numerical simulation experiment method, and all steps and conclusions are verified correct on Matlab R2022a. The following will be combined with the drawings to further illustrate the present application. Figures 1-10 Further description of the present application.

[0026] Step 1: Construct a networked radar deception jamming implementation scenario, and determine the ECAVs motion constraint

[0027] 1.1, a typical networked radar deception jamming implementation scenario is shown in FIG. 1, for each radar, at each time, a set of ECAVs needs to be located between the radar and the false target, the unmanned jammer transmits the echo with time delay to the radar, so that the radar detects the false point track. Figure 1

[0028] Since there may be multiple false tracks in the scene, it is assumed that the total number of false tracks is N F At different times, the matching relationship between ECAVs and false tracks is not fixed, only need to ensure that all point tracks are formed. Without loss of generality, the number of false tracks matched by the ECAV numbered i at time t is recorded as m i (t), in order to ensure that all false tracks can be formed, the following conditions need to be met

[0029]

[0030] That is, at any time, ensure that the matching relationship between false tracks and ECAVs meets the "one-to-one" matching relationship.

[0031] 1.2, the ECAVs motion constraint is analyzed and described below. In Figure 2 , the radar position is P R For the jth false track, the distance of the false target relative to the radar is R j (t), the azimuth angle is θ j (t), due to the RGPO (Range Gate PullOff, distance wave door drag interference) interference principle, the ECAV is on the radar and the false target line of sight (Line of sight, LOS), the ECAV is numbered as 1, 2, …, N F , at time t, for the ECAV numbered i, v​e,i (t) represents its velocity, r i (t) represents its distance to radar, φ e,i (t) represents the angle between ECAV heading and the line of sight, and represents the angle between ECAV heading and the horizontal direction, then the angle of each ECAV needs to satisfy

[0032]

[0033] Combining Figure 2 , the distance-to-radar movement of ECAVs needs to satisfy

[0034]

[0035] the azimuth movement needs to satisfy

[0036]

[0037] where, represents the rate of change of distance to radar, represents the rate of change of azimuth.

[0038] Further derivation can obtain the following constraint relationship

[0039]

[0040] The above constraints make ECAVs satisfy that each moment is on the connecting line of radar and false point, at the same time, the movement of ECAVs also needs to satisfy the following basic movement constraints, that is, the ECAVs maneuvering ability, the minimum safety distance constraint between ECAVs and radar

[0041] r i (t) ≥ r i,min (6)

[0042] the minimum trailing distance constraint

[0043]

[0044] the ECAVs speed constraint

[0045] v e,i,min ≤ v e,i (t) ≤ v e,i,max (8)

[0046] the ECAVs acceleration constraint

[0047]

[0048] the ECAVs turning angle constraint

[0049]

[0050] where,△R i,min is the minimum trailing distance, which is 1km in this embodiment; r i,min is the minimum safety distance between the ECAV and the radar, which is 5km in this embodiment; is the ECAV acceleration, is the ECAV angular velocity, v e,i,min , v e,i,max , a e,i,min , a e,i,max , w e,i,min , w e,i,max are the upper and lower limits of the ECAV speed, acceleration, and angular velocity, respectively.

[0051] Step 2: Determine the constraints of the ECAV control strategy optimization model by discretizing the time and space dimensions

[0052] 2.1, Discretize the scene in the time and space dimensions. First, the time dimension is discretized into frames with △T as the interval, and the positions of the ECAVs at times △T, 2△T, 3△T, etc. are focused on. Then, the space dimension is discretized, and at each time, the ECAVs need to be on the LOS between the radar and the false point track. In order to reduce the search difficulty, the LOS is discretized into a point sequence with △D as the interval. After discretization, the changes in the ECAV state are caused by the changes in the ECAV position at different times.

[0053] 2.2, Determine the constraints of the ECAV control strategy optimization model after time and space discretization. As shown in Figure 3 , assume that the current position of the ECAV is (x e , y e ), and the position of the false target formed by trailing in the next frame is (x t , y t ). Due to the distance gate deception jamming principle, the ECAV needs to be on the line connecting the radar and the target in the next frame, and its position (x, y) needs to satisfy the following relationship:

[0054]

[0055] Assume that the time difference between two frames is △T, and the maximum acceleration time t acc of the ECAV in this period is

[0056]

[0057] The maximum distance d e,max that can be traveled is

[0058]

[0059] The longest deceleration time tdec For

[0060]

[0061] The minimum distance d e,min For

[0062]

[0063] Therefore, after the time period of ΔT, the distance that the ECAV can travel is between d e,min and d e,max The next frame of the ECAV may exist in the position that needs to satisfy the following conditions

[0064]

[0065] Substitute equation (11) into equation (16) and move terms to simplify to obtain the following two quadratic inequalities

[0066]

[0067] Wherein, A, B, C1, C2 are quadratic inequality coefficients, which can be calculated according to the current known quantities, and the calculation method is

[0068]

[0069] For the first condition of equation (17), the solution set that satisfies the condition is

[0070]

[0071] Wherein, Φ represents an empty set, is the quadratic equation root determination formula.

[0072] Similarly, for the second condition, the solution set is

[0073]

[0074] Wherein, Υ represents a universal set, is the quadratic equation root determination formula.

[0075] Finally, the feasible region of the ECAV position is

[0076]

[0077] Step 3: Complete the construction of the ECAV control strategy optimization model

[0078] When there is only one false target, one ECAV needs to be implemented to interfere, and the change of the ECAV state is as shown in Figure 4 Assume that the initial position of the ECAV is P init, according to the initial speed, orientation, and motion constraints, etc. information, the optional range of the first frame ECAV position can be calculated, assuming that there are two positions available, P 11 and P 12 , when moving from the initial position to different positions, the required acceleration, rotation angle, etc. information are different, which will result in different states of the first frame ECAV. When moving to P 11 position, assuming that the optional positions of the second frame are P 21 , P 22 , P 23 , when moving to P 12 position, assuming that the optional positions of the second frame are P 23 , P 24 , P 25 , similarly, moving to different positions makes the ECAV have different states, based on the state transition relationship as above, a tree structure as shown in Figure 5 is established.

[0079] When there are multiple false targets, the state transition process also needs to consider the matching relationship between ECAVs and false tracks, as shown in Figure 6 , when two ECAVs are used to implement two false tracks, assuming that the initial positions of the two ECAVs are P 1,init and P 2,init , and in the initial state, ECAV1 matches false track 1, and ECAV2 matches false track 2, the state transition process that maintains the matching relationship is shown in Figure 6 (a), the optional positions of ECAV1 are P 1,11 and P 1,12 , and the optional positions of ECAV2 are P 2,11 and P 2,12 , the state transition process that changes the matching relationship is shown in Figure 6 (b), the optional positions of ECAV1 are P 1,13 and P 1,14 , and the optional positions of ECAV2 are P 2,13 and P 2,14 , after establishing the state transition relationship according to the optional positions of each case, a tree structure as shown in Figure 7 is obtained.

[0080] In summary, when the scene is discretized, a tree T {V, E} with the initial state of ECAVs as the top vertex can be established according to the state transition process of ECAVs, wherein V is the vertex set of the tree, each vertex represents a state of the ECAVs cluster, and E is the edge set of the tree, each edge represents a state transition process of the ECAVs cluster. The size of the tree T is related to the total number of false tracks, the discretization granularity, and the number of ECAVs. The depth of the tree increases with the increase of the number of false track frames, and the width of the tree increases with the increase of the discretization precision and the number of ECAVs.

[0081] When the tree structure is established, the ECAVs control strategy optimization problem can be converted into a shortest path search problem in the tree T, that is, searching for the shortest path from the root node to the leaf node in the tree. Assuming that the false track has K frames, the problem can be converted into the following problem

[0082]

[0083] wherein p is a path from the root node S root to the leaf node , that is, an ECAVs control strategy, S k,i represents the i-th vertex in the k-th layer of the tree T, and also represents a state of the ECAVs, and M is the number of vertices in the path except the root node. Since the ECAVs have motion restrictions, the entire K frames of false tracks may not be completely implemented, so M≤K; V(k) represents the vertex set of the k-th layer of the tree T, E(S k,i ) represents the edge set with the node S k,i as the starting point, e(S k,i →S k+1,j ) represents an edge with the node S k,i as the starting point and the node S k+1,j as the end point. C(·) is a path evaluation objective function, that is, an ECAVs control strategy evaluation index;

[0084] The ECAVs control strategy evaluation index needs to consider two factors: the false track implementation completion degree and the ECAVs flight distance. When comparing different control strategies, the following needs to be met: (1) If the false track implementation completion degrees of two control strategies are different, the control strategy with the higher completion degree is better; (2) If the false track implementation completion degrees of two control strategies are the same, the control strategy with the shorter ECAVs flight distance is better. Considering the implementation completeness of the false track and the ECAVs cluster travel distance, the following objective function can be used for optimization:

[0085]

[0086] wherein N F represents the number of false points generated by ECAVs at a certain moment, ve,i,max the maximum speed of the ECAV, denotes a node transitions to a node the total distance of the ECAV, and other symbols are defined above. The objective function includes two terms, the first term is the maximum distance of all ECAVs flying M frames, and the second term is the average length of the actual flight path of each frame, which considers both the implementation degree of the false track and the flight distance of the ECAV. The higher the degree of completion, the greater the calculated value of the path, and the shorter the actual flight distance, the greater the calculated value of the path.

[0087] Step 4: Taking the SP-MCTS algorithm as the framework, the four stages of selection (Selection), expansion (Expand), simulation (Simulation), and backpropagation (Backpropagation) in the ECAV motion trajectory planning scene are determined, and the ECAV control strategy optimization model is solved.

[0088] 4.1, the ECAV control strategy optimization model is established in the foregoing step, and the model is converted into a tree optimal path search problem through discretization. Since any change in the state of the ECAV will increase the nodes in the tree, the size of the tree is large, and it is difficult to solve globally by traditional algorithms, therefore, the SP-MCTS algorithm is introduced to solve the numerical solution. The SP-MCTS algorithm is a heuristic tree search algorithm, which has the characteristics that path search and tree establishment are carried out at the same time, and UCT (Upper Confidence bounds applied to Trees) is used to guide the search direction, which can reduce the calculation amount of path search and is suitable for the problem. The overall framework of the ECAV path cooperative planning algorithm based on SP-MCTS is shown in Table 1:

[0089] Table 1 ECAV control strategy planning algorithm based on SP-MCTS

[0090]

[0091] The algorithm mainly includes four stages: selection (Selection), expansion (Expand), simulation (Simulation), and backpropagation (Backpropagation), each stage corresponds to a key processing function, which will be described in detail in the following steps. In Table 1, Eval(S leaf ) represents the strategy evaluation function, which uses formula (23) described in step 3 to evaluate the control strategy, S leaf represents a leaf node, which can uniquely determine the control strategy in the tree, and r represents the evaluation of S leaf) the calculated policy evaluation result, i.e. the reward value, S cur represents the selected node in the current iteration.

[0092] 4.2, the SP-MCTS selection phase processing function Selection(), whose flow is shown in Table 2

[0093] Table 2 SP-MCTS selection phase processing flow

[0094]

[0095] In this process, the calculation of the upper confidence bound UCT requires recording the total number of visits N of the tree T total , and the number of visits N(S i ) and the cumulative reward value Q(S i ) of each node S i , N(S i ) and Q(S leaf ) are initialized to 0, and the specific iteration method is given by the backtracking phase processing function Backpropagation() in step 4.5. Using these data, UCT can be calculated, and the calculation formula is

[0096]

[0097] Where c is a control coefficient.

[0098] The selection phase processing function Selection(), whose purpose is to select a leaf node S leaf in the tree T according to the guidance of UCT. This function starts from the root node S leaf and gradually selects to the leaf node. For non-leaf nodes, if all its child nodes have been visited, the node with the maximum UCT value in the child nodes is selected for the next iteration, and if there are child nodes that have not been visited, a randomly selected unvisited child node is selected. Through the guidance of UCT, the selection process is biased towards the optimal path, avoiding global search of the tree T.

[0099] 4.3, the SP-MCTS expansion phase processing function Expand(), whose flow is shown in Table 3

[0100] Table 3 SP-MCTS expansion phase processing flow

[0101]

[0102] The expansion phase processing function Expand(), whose purpose is to calculate the leaf node S i according to the motion constraints of ECAVs.The set V contains all subsequent nodes. This function needs to handle the matching relationships between ECAVs and fake tracks. Here, a matching vector m represents the matching relationship between ECAVs and fake tracks. For an ECAV with number i, the number of the fake track it matches is denoted as m(i). During Expand() processing, the set V is first initialized to an empty set; if S... leaf If it corresponds to the last frame, there are no subsequent nodes, and the result of Expand() is an empty set; if it is not the last frame, then it is necessary to traverse each matching relationship m between ECAVs and pseudo tracks, and calculate the set of possible coordinates PS for the next frame of all ECAVs under each matching relationship. i For ECAV numbered i, according to the RGPO false track... With the matching vector m, we can obtain the false dots (x) that need to be formed in the next frame. t ,y t Then, with △D as the interval, the radar -(x) t ,y t Discretize the connecting line segments to obtain a set of candidate coordinates. Then, filter the candidate coordinates using the motion constraints derived in step 2.2 to obtain the feasible set PS. i After obtaining the selectable coordinates of ECAVs, if the current frame is for N... F A feasible set of spurious dots If none of them are empty, then from these N... F In each set, select a coordinate as the next frame position for the corresponding ECAV. The Comb() function in Table 3 selects a coordinate from the feasible set. Choose one coordinate from each of the options, and iterate through all feasible choices. Each choice corresponds to S. leaf A subsequent child node; using Comb() to continuously expand set V, eventually obtaining S. leaf The set of subsequent nodes.

[0103] 4.4 The Simulation() function, used in the SP-MCTS simulation stage, has a flowchart shown in Table 4.

[0104] Table 4. Processing Flow of SP-MCTS Simulation Phase

[0105]

[0106] The simulation phase processing function `Simulation()` aims to evaluate the input using the Monte Carlo method and obtain its reward value `r`. This function takes a leaf node as input, randomly selects its subsequent nodes, generates a random policy, and calculates the reward value `r` for that random policy. Although this process uses `Expand()` to calculate the optional child nodes of a node, it does not update the tree `T`, thus preventing the tree `T` from expanding in size.

[0107] 4.5, SP-MCTS backtracking phase processing function Backpropagation(), its flow is shown in Table 5

[0108] Table 5 SP-MCTS backtracking phase processing flow

[0109]

[0110] The backtracking phase processing function Backpropagation() is to update the tree T using the strategy reward value r. The function starts from the leaf node, updates the node access frequency N, reward cumulative value Q, and confidence upper limit UCT, and goes up level by level until the root node.

[0111] 4.6, using the above algorithm, the establishment of the tree and the optimal strategy search are carried out synchronously. For single false track scene, after t max iterations, the Selection() is executed to obtain the optimal control strategy of ECAV; for multi-false track scene involving track switching, after t max iterations, the correspondence between ECAV and false track is determined, and the point sequence corresponding to the jth ECAV is recorded as The point sequence set of the corresponding false track of all ECVAs is defined as After that, replace with At this time, the multi-false track scene can be regarded as N F single false track scenes without track switching, and the optimal control strategy of ECAV can be obtained by iterating optimization again. t max In the simulation of the embodiment, the value is 100000.

[0112] The effect of the application is further illustrated by the following three scene simulation comparison experiments:

[0113] Simulation scene a: single false track, and ECAVs are closer to the networked radar, and are biased to detect the total travel distance index of ECAVs.

[0114] Simulation scene b: single false track, and ECAVs are closer to the false track, and are biased to detect the integrity index of the false track.

[0115] Simulation scene c: multi-false track, and is biased to detect whether the algorithm can introduce more possibilities for path planning by considering the replacement of the matching relationship between ECAVs and false tracks at different times, so as to plan a more optimal travel path (in order to avoid repetition, one radar is taken as an example for illustration, and the effects of other radar sites are equivalent).

[0116] In each scenario, 15 frames of false tracks are considered, the inter-frame time interval ΔT is 1s, the UCT control coefficient c is set to 0.7, and the spatial discrete interval ΔD is set to 2m. The simulation parameters corresponding to the simulation scenario a, the simulation scenario b and the simulation scenario c are shown in Table 6, Table 7 and Table 8 respectively. The simulation scenario a solves all radar sites in turn, which can present the global effect of the RGPO interference of the networked radar, but since the present application independently considers different radar sites, in order to avoid repetition, only one radar site is considered in the simulation scenario b and c.

[0117] Table 6 Simulation parameter table of scenario a

[0118]

[0119] Table 7 Simulation parameter table of scenario b

[0120]

[0121] Table 8 Simulation parameter table of scenario c

[0122]

[0123] The simulation results in the three scenarios are shown in Figure 8 , Figure 9 , Figure 10 From the planning results, it can be seen that the algorithm described in the present application solves the feasible ECAV path, in scenario a, the movement trend of the ECAV is to first adjust the movement direction and speed, and then gradually adjust to the direction approximately perpendicular to the LOS line, and then basically keep this direction unchanged to pursue a shorter total travel distance. In scenario b, in order to meet the minimum tow distance constraint and maximize the completion rate of the false track, the ECAV selects a longer path in the 5th-10th frame, abandoning the selection of the single-step shortest path. In scenario c, the false targets responsible by each ECAV at different times are different, in 1-7 frames, ECAV1 is responsible for forming false track 1, and ECAV2 is responsible for forming false track 2, in the subsequent 8-15 frames, the false targets responsible by the two are exchanged, ECAV1 is responsible for forming false track 2, and ECAV2 is responsible for forming false track 1, through track switching, the movement constraint of ECAV1 is relaxed, so that it can complete all 15 frames of interference.

[0124] In order to prove the effectiveness of the method described in the present application, a comparative experiment with other algorithms is carried out, and Table 9 and Table 10, Table 11 and Table 12 respectively show the results of the method described in the present application compared with other algorithms in the three scenarios.

[0125] Table 9 Summary table of left ECAV path planning results in scenario a

[0126]

[0127] Table 10 Summary table of path planning results of ECAVs on the right side of scenario a

[0128]

[0129] Table 11 Summary table of simulation results of scenario b

[0130]

[0131] Table 12 Comparison table of completion degrees of false tracks of scenario c

[0132]

[0133]

[0134] In summary of the above three simulation experiments, the path planned by the method described in the present application can be close to or better than other methods in terms of driving distance under relatively wide constraints, the path planned can maximize the completion degree of false track implementation under critical conditions, and in the multi-false track scenario, more possibilities are introduced into path planning by considering the replacement of ECAVs and false track matching relationship at different times, and a better driving path can be planned.

[0135] Those skilled in the art will realize that the embodiments described herein are for the purpose of helping the reader understand the principles of the present application and should be understood as not limiting the scope of protection of the present application to such specific statements and embodiments. The present application can have various modifications and changes for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the claims of the present application.

Claims

1. A method for generating a control strategy of a jammer aiming at false tracks of a networked radar, characterized in that, Comprise: S1, build a networked radar deception jamming implementation scenario, determine the ECAVs motion constraints; Implementation scenario is specific: for each radar site of networked radar, at each time, a group of unmanned jammer needs to be located between the radar and the false target, ECAV will send the echo with time delay to the radar, so that the radar detects the false point track; The ECAVs motion constraints are that the ECAVs satisfy being on the connecting line between the radar and the false point track at each time, and satisfy the ECAVs and radar minimum safety distance constraint, minimum tow distance constraint, ECAVs speed constraint, ECAVs acceleration constraint, ECAVs corner constraint; S2, through the discretization of time and space dimensions, determine the ECAVs control strategy optimization model constraints; S3, complete the construction of the ECAVs control strategy optimization model; According to the state transition process of ECAVs, a tree T{V,E} with ECAVs initial state as the top point is established, wherein V is the vertex set of the tree, each vertex represents a state of the ECAVs cluster, E is the edge set of the tree, each edge represents a state transition process of the ECAVs cluster; So as to convert the ECAVs control strategy optimization problem into the optimal path search problem in the tree T; where p is a path from root node S root to leaf node , i.e. a control strategy of ECAVs; S k,i represents the i-th vertex in the k-th layer of tree T, i.e. a state of ECAVs; M is the number of vertices in the path except the root node, so M≤K; V(k) represents the vertex set of the k-th layer of tree T, E(S k,i ) represents the edge set with node S k,i as the starting point, e(S k,i →S k+1,j ) represents the edge with node S k,i as the starting point and node S k+1,j as the ending point; C(.) is the path evaluation objective function; S4, search the tree T by using SP-MCTS algorithm, and obtain the optimal control strategy.

2. The method of claim 1, wherein the method further comprises: determining a number of radar false tracks; and determining a number of radar false track groups based on the number of radar false tracks. The expression of C(p) is: where N F represents the number of false points generated by ECAVs at a certain moment, v e,i,max is the maximum speed of ECAVs, represents the total distance traveled by ECAVs when transitioning from node to node , and △T represents the interval time of adjacent frames.

3. The method of claim 2, wherein the method further comprises: determining a number of radar groups; and determining a number of radar groups to be jammed. Step S4 specifically includes the following steps: S41, select a stage processing function from the root node S root Start to select step by step to the leaf node, for non-leaf node, if all its child nodes are visited, select the node with the maximum UCT value in the child nodes for the next iteration, if there are child nodes not visited, randomly select one unvisited child node; S42, the extension stage processing function calculates all the subsequent node set V of the leaf node S according to the motion constraint of the ECAV leaf ; S43, the simulation stage processing function evaluates the control strategy represented by the input leaf node by randomly selecting the subsequent strategy, and obtains the reward value r S44, the backtracking stage processing function updates the tree T by using the strategy reward value r; S45, after t max After the sub-iteration, the selection phase processing function is executed to obtain the optimal control strategy of the ECAV.

4. The method of claim 3, wherein the method further comprises: The UCT value calculation formula is: where c is a control coefficient, N total represents the total number of visits to tree T, N(S i ) represents the number of visits to node S i , Q(S i ) represents the cumulative reward value of node S i , and N(S i ) and Q(S i ) are initialized to 0.

Images