Simulation demonstration and evaluation system for bee colony game algorithm of unmanned aerial vehicle

Abstract

The invention discloses a simulation demonstration and evaluation system for a bee colony game algorithm of an unmanned aerial vehicle, which comprises a site unit, an algorithm verification platform, a simulation confrontation target and a demonstration evaluation unit, and is characterized in that the site unit selects an indoor or open area, and guardrails are arranged around the site unit; arranging a red party and a blue party relative to the demonstration subjects according to fixed-point strike, airway interception and collaborative hunting set in the competition; selecting an unmanned aerial vehicle as an algorithm verification platform of a bee colony game algorithm, and selecting a ground remote control unmanned vehicle as a simulation confrontation target; the demonstration evaluation unit comprises a positioning module, a communication module and scoring software, the positioning module adopts a UWB positioning system and comprises a positioning base station and a UWB label, and the scoring software is used for recording all data of the algorithm verification platform in each demonstration subject and scoring according to a preset scoring standard; and meanwhile, real-time demonstration and playback of corresponding subjects can be carried out according to recorded data.

Description

Technical Field

[0001] The present invention relates to the field of drone attack and defense, and in particular to a simulation demonstration and evaluation system for a drone swarm game algorithm. Background Art

[0002] In recent years, unmanned aerial vehicles (UAVs), represented by drones, have been widely used in military and civilian applications. However, as the tasks to be performed become more and more complex and the environments they face become more and more complex, especially in actual confrontation scenarios represented by denial environments, a single UAV is often unable to complete certain large-scale tasks. This has also led to the development trend of unmanned aerial vehicles from a single UAV performing tasks to multiple UAVs or even clusters coordinating tasks. Unmanned aerial swarms form multiple formations to complete tasks collaboratively. On the one hand, they can complete tasks that a single UAV or a single formation cannot complete, such as collaborative search and tracking; on the other hand, they can improve the success rate of task execution. Therefore, drone swarm combat systems composed of high and low performance combinations have gradually emerged. With the development of swarm technology and the application of swarm weapons, game confrontation between drone swarms will become a new form of combat. However, there is no mature cluster attack and defense decision-making method in China, and it is impossible to effectively evaluate the effectiveness of drone swarm attack and defense decisions, resulting in a large number of gaps in the research on drone swarm game confrontation. Therefore, it is urgent to establish a drone swarm target trajectory and game algorithm evaluation system. Summary of the invention

[0003] The purpose of the present invention is to provide a simulation demonstration and evaluation system for a drone swarm game algorithm, which is applied to the simulation demonstration and evaluation of a drone swarm game algorithm.

[0004] In order to achieve the above tasks, the present invention adopts the following technical solutions:

[0005] A simulation demonstration and evaluation system for a drone swarm game algorithm includes a site unit, an algorithm verification platform, a simulated confrontation target, and a demonstration evaluation unit, wherein:

[0006] The venue unit is selected indoors or in an open area, and guardrails are arranged around the venue unit; the red team and the blue team are arranged according to the demonstration subjects set by the competition, such as fixed-point strike, route interception, coordinated round-up, and tit-for-tat;

[0007] The UAV is selected as the algorithm verification platform for the swarm game algorithm, and the ground remote-controlled unmanned vehicle is selected as the simulated confrontation target;

[0008] The demonstration and evaluation unit includes a positioning module, a communication module and scoring software. The positioning module adopts the UWB positioning system, including a positioning base station and a UWB tag. The communication module adopts a Wi-Fi device. The scoring software is used to record all data of the algorithm verification platform in each demonstration subject and score according to the preset scoring criteria. At the same time, real-time demonstration and playback of the corresponding subjects can be carried out based on the recorded data.

[0009] Furthermore, the unit area of ​​the site shall not be less than 50 meters in length and 50 meters in width, and shall be constructed with square tube trusses; the take-off and landing area, assembly and waiting area, no-fly / avoidance area, and confrontation / mission area of ​​the UAV shall be set up according to the requirements of the mission scenario.

[0010] Furthermore, the drone should be a pure electric four-rotor drone. The drone is required to be fully autonomously controlled throughout the competition, with a flight speed not exceeding 5m / s. The drone is equipped with a positioning module, and the location information recorded by the module will be transmitted to the competition venue's LAN. The drone obtains the real-time location of its own position, other drones and unmanned vehicles published in the competition venue's LAN by subscribing to ROS messages in the competition venue's LAN. The maximum movement speed of the ground-based remote-controlled unmanned vehicle is not less than 3m / s, and it can carry no less than two UWB tags at the same time; it has omnidirectional driving function, programmable driving, and adjustable speed function; at the same time, it can be equipped with a positioning module to report location information to the competition venue's LAN in real time.

[0011] Furthermore, the UWB positioning system adopts 2D positioning, and the positioning base station of the UWB positioning system is deployed outside the site unit, on road lamp poles, building roofs, container tops or constructed poles; the UWB positioning system can meet indoor and outdoor use, and can support no less than 5 teams to use it at the same time, and the positioning coverage area is not less than 100 meters × 100 meters.

[0012] Furthermore, the time data recorded by the scoring software should be accurate to 1 second, the position calculation should be accurate to 1mm, and the angle calculation should be accurate to 1°, so that it can calculate the coordinated strike angle and direction.

[0013] Furthermore, the fixed-point strike is a direct collision and coordinated strike by multiple drones of the red side against the blue side target at a fixed position in the field unit; the blue side target position is randomly set within the preset range of the field unit;

[0014] Scoring criteria:

[0015] There are two parts of scoring: effective strike and coordinated strike, among which coordinated strike is further divided into time coordination and angle coordination. The effective strike is determined by the distance between the red team's drone position and the blue team's target position. When multiple red team drones have caused effective strikes on the blue team's target, it is determined whether time coordination has been constituted by calculating whether the time between the previous and subsequent strikes meets the preset coordination time. When multiple red team drones constitute time coordination, the angle formed by the velocity vectors of two of the drones that constitute time coordination when they strike the blue team's target is recorded as the coordinated strike angle, and it is determined whether the coordinated strike angle meets the preset coordination angle requirements, thereby determining whether the two drones constitute angle coordination.

[0016] Furthermore, in the fixed-point strike demonstration subject, a UAV stationary avoidance zone is designed, and the UAV stationary avoidance zone is randomly set within the preset range of the site unit.

[0017] Furthermore, the route interception is a direct collision and coordinated attack by multiple drones of the red side on the blue side targets moving along a preset trajectory in the field unit;

[0018] Scoring criteria:

[0019] It is divided into three parts: effective strike, head-on strike and coordinated strike. The effective strike is determined by the distance between the red team's drone position and the blue team's target position. When a drone effectively strikes the blue team's target, the angle between the drone's speed vector and the blue team's target speed vector is calculated. When the angle is obtuse, the strike is considered a head-on strike. When multiple drones effectively strike the blue team's target, and the time difference between the strikes of the two drones is less than the preset time, it is considered a coordinated strike.

[0020] Furthermore, the coordinated capture is that multiple drones of the red side conduct time-coordinated strikes on the blue side targets running the game escape strategy in the field unit, and the blue side targets can autonomously avoid the strikes of the red side drones;

[0021] Scoring criteria:

[0022] It is set up in two parts: effective strike and coordinated strike. The effective strike is determined by the distance between the red side's drone position and the blue side's target position. When multiple drones of the red side have caused effective strikes on the blue side's target, and the time difference between the strikes of the two drones is less than the preset time, it is determined to have constituted a coordinated strike.

[0023] Furthermore, tit-for-tat is to divide multiple drones into two confrontation clusters, requiring them to act as the red side and the blue side to conduct offensive and defensive confrontations within a specified time; set up a confrontation area in the field unit, with the red side as the defender and its base located in the center of the confrontation area, and the blue side as the attacker; initially, the red side is in the warning area, and the blue side's drones are randomly distributed outside the warning area; the blue side's confrontation goal is to break through the red side's interception, destroy the red side's base, and successfully escape from the confrontation area, while the red side's confrontation goal is to destroy the enemy's drones and protect its own base;

[0024] Scoring criteria:

[0025] Set up three parts: successful interception, successful penetration, and survival after penetration; when the defender shoots down the attacker's drone, it is considered a successful interception; when the horizontal distance between the attacker's drone and the defender's base is less than the preset distance, the attacker's drone is considered to have successfully penetrated, and each attacker's drone can only be recorded as a successful penetration once; judgment of survival after penetration: when a drone of the attacker successfully penetrates and is not hit before escaping from the confrontation area, it is considered to have survived the penetration.

[0026] Furthermore, in the tit-for-tat demonstration subject, the confrontation area is a circular area with a radius of 12 meters. The red and blue sides should have at least 2 drones, and the blue drones are randomly distributed in the area 6 meters outside the red base.

[0027] Compared with the prior art, the present invention has the following technical features:

[0028] The present invention builds a test and evaluation system in a fixed site or a limited space, uses a given mounted flight platform to load the software algorithm, and simulates three-dimensional aerial attack and defense confrontation through two types of targets, air and ground, to provide strong support for the simulation demonstration, verification and evaluation of the UAV swarm game algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a schematic diagram of a site unit in the system of the present invention;

[0030] Figure 2 Schematic diagram of the deployment of UWB positioning base stations in the stadium;

[0031] Figure 3 This is a schematic diagram of the targeted strike subject;

[0032] Figure 4 This is a schematic diagram of the route interception subject;

[0033] Figure 5 It is a schematic diagram of the subjects for collaborative roundup;

[0034] Figure 6 Schematic diagram of collaborative roundup subjects. DETAILED DESCRIPTION

[0035] The present invention provides a simulation demonstration and evaluation system for a UAV swarm game algorithm, which adopts UWB to test the motion trajectory of aerial UAVs and the trajectory of ground unmanned vehicles simulating blue forces in real time, and evaluates the confrontation ability, complexity, practicality, etc. of the UAV swarm game algorithm through given scoring rules.

[0036] The system of the present invention includes a site unit, an algorithm verification platform, a simulated confrontation target, and a demonstration evaluation unit, wherein:

[0037] 1. Site unit

[0038] like Figure 1 As shown in the figure, as one of the main scenarios of swarm warfare, cities are limited by urban space. In order to avoid collateral damage caused by the two sides of the game, the site unit is selected in indoor or open areas with an area of ​​no less than 50 meters long × 50 meters wide. It is built with square tube trusses and 3 meters high. Guardrails are arranged around the site units. According to the requirements of the mission scenario, the take-off and landing areas, assembly and waiting areas, no-fly / avoidance areas, confrontation / mission areas of the drones are set up, and the mission objectives are arranged according to the different demonstration subjects set by the competition, such as fixed-point strikes, route interception, coordinated encirclement, tit-for-tat, etc.

[0039] 2. Algorithm verification platform and simulated adversarial targets

[0040] The UAV is selected as the algorithm verification platform for the swarm game algorithm. Different swarm game algorithms are loaded into the UAV according to the actual verification requirements. In order to ensure the safety of the confrontation process, a ground remote-controlled unmanned vehicle is selected as the simulated confrontation target.

[0041] The drone should be a pure electric four-rotor drone with a diagonal wheelbase not exceeding 350mm. The drone is required to be fully autonomously controlled throughout the competition and the flight speed shall not exceed 5m / s. At the same time, the drone is equipped with a positioning module, and the position information recorded by the module will be transmitted to the competition venue LAN. The drone obtains the real-time position of the drone itself, other drones and unmanned vehicles published in the competition venue LAN by subscribing to ROS messages (type is geometry_msgs / PoseStamped) in the competition venue LAN. In one embodiment of the present invention, the drone uses a DJI Tello drone.

[0042] The ground remote-controlled unmanned vehicle uses shelf products, which are used for simulating the position changes of ground moving targets and avoidance zones. The maximum movement speed is not less than 3m / s, and it can carry no less than two UWB tags at the same time; it has omnidirectional driving function, and the overall dimensions of length, width and height are not more than 350mm, and it has program-controlled driving and adjustable speed functions; at the same time, it can be equipped with a positioning module to report position information to the local area network of the stadium in real time; for example: in one embodiment of the present invention, the ground remote-controlled unmanned vehicle uses DJI Mech S1.

[0043] 3. Demonstration Evaluation Unit

[0044] The demonstration evaluation unit includes a positioning module, a communication module and scoring software, including:

[0045] The positioning module adopts the UWB positioning system. Since the site unit of the present invention is a 2-dimensional small area scene, the UWB positioning system adopts 2-dimensional positioning. The positioning base station of the UWB positioning system is deployed outside the site unit, and can be deployed on road lamp poles, rooftops, container tops, or erected poles. All positioning base stations together form a polygonal area, and the most ideal shape is a regular polygon; and the angle between the two sides of the polygon is preferably close to a right angle to avoid forming a large obtuse angle or acute angle. For example, a regular quadrilateral can be used. An embodiment of the present invention is as follows Figure 2 As shown, in order to support the autonomous navigation needs of the participating drones and the calculation of the key assessment indicators by the competition scoring system, 4 positioning base stations are set up with a height of 2m and arranged in a regular quadrilateral. At the same time, the UWB tags of the UWB positioning system include 16 LinkTrack PB tags, 8 LinkTrack P-BS tags (including self-assembly installation shells), 8 LinkTrack P-BS2 tags (including self-assembly installation shells) and installation brackets, power batteries, chargers, etc., which can meet the needs of indoor and outdoor use and support no less than 5 teams to use at the same time. The positioning coverage area is not less than 100 meters × 100 meters, and the battery independent power supply is guaranteed to be not less than 8 hours.

[0046] The communication module uses Wi-Fi equipment for communication between drones, ground-based remote-controlled drones, etc. and the venue's local area network; the Wi-Fi equipment should meet the Wi-Fi signal coverage requirement of no less than 100 meters x 100 meters, and be able to confirm the identity of the access IP; in one embodiment of the present invention, the Wi-Fi equipment includes: 1 ProE.AC integrated VPN router, 2 dual-band Wi-Fi6 and installation equipment, network cables and power supply guarantees; to ensure reliable operation of the demonstration site, one wireless controller AC and one wireless access point AP need to be added inside and outside the venue unit.

[0047] Scoring software: The scoring software is used to record all data of the algorithm verification platform in each demonstration subject (position and speed data of drones and unmanned vehicles, UWB tag data, etc.) and score according to the preset scoring standards. It also has the function of saving demonstrations and playback, and can perform real-time demonstrations and playback of corresponding subjects based on the recorded data; the time data recorded by the scoring software should be accurate to 1 second, the position calculation should be accurate to 1mm, the angle calculation should be accurate to 1°, and it can calculate the coordinated attack angle and direction. The remaining functions of the scoring software can be developed according to actual needs, and this is an existing technology and will not be repeated.

[0048] 4. Demonstration subjects and scoring criteria

[0049] The invention comprises four demonstration subjects: fixed-point strike, route interception, coordinated roundup and tit-for-tat.

[0050] (1) Targeted strikes

[0051] Set multiple drones as the red team and ground remote-controlled unmanned vehicles as the blue team targets; fixed-point strikes are multiple drones of the red team that conduct direct collision and coordinated strikes on the blue team targets at fixed positions in the field unit. The coordinated strike time between drones should be as close as possible, and the coordinated strike angle should meet the requirements; Figure 3 The red box in the middle is the take-off position of the red UAV, and the blue circle represents the ground remote-controlled unmanned vehicle; the blue target position is randomly set within the preset range of the site unit, and a UAV static avoidance zone is designed. The UAV static avoidance zone is randomly set within the preset range of the site unit.

[0052] Scoring criteria:

[0053] To simplify the calculation, the horizontal projection method is used, and two parts are set: effective strike and coordinated strike. The coordinated strike is divided into time coordination and angle coordination. The effective strike is determined by the distance between the red team's drone position and the blue team's target position. When multiple red team drones have caused effective strikes on the blue team's target, whether time coordination is constituted is determined by calculating whether the time before and after the strike meets the preset coordination time. When multiple red team drones constitute time coordination, the angle formed by the velocity vectors of two of the drones that constitute time coordination when they strike the blue team's target is recorded as the coordinated strike angle, and whether the coordinated strike angle meets the preset coordination angle requirements is judged, thereby determining whether the two drones constitute angle coordination.

[0054] (2) En-route interception

[0055] Set multiple drones as the red team and ground remote-controlled unmanned vehicles as the blue team targets; route interception is for multiple drones of the red team to conduct direct collision and coordinated strikes on the blue team targets moving along a preset trajectory (such as a sine function curve) in the field unit; the coordinated strike time between drones should be as close as possible, and head-on strikes should be achieved as much as possible; Figure 4 In the figure, the red box is the take-off position of the red drone, and the blue circle is the initial position of the blue target.

[0056] Scoring criteria:

[0057] To simplify the calculation, horizontal projection is used, and three parts are set: effective strike, head-on strike, and coordinated strike. An effective strike is determined by the distance between the red team's drone position and the blue team's target position. A head-on strike is: when a certain drone causes an effective strike on the blue team's target, the angle between the drone's velocity vector and the blue team's target velocity vector is calculated. When the angle is an obtuse angle, the strike is considered a head-on strike. When multiple drones have caused effective strikes on the blue team's target, and the time difference between the strikes of the two drones is less than 2 seconds, it is considered a coordinated strike.

[0058] (3) Collaborative round-up

[0059] Set multiple drones as the red team and ground remote-controlled unmanned vehicles as the blue team targets; coordinate the multiple drones of the red team to carry out time-coordinated strikes on the blue team targets running the game escape strategy in the field unit, and the blue team targets can autonomously avoid the strikes of the red team drones; Figure 5 The red box in the middle is the take-off position of the red team’s drone, and the blue circle is the initial position of the blue team’s target (this position is closer to the CD of the opposite side of the unit site than the drone’s take-off position).

[0060] Scoring criteria:

[0061] To simplify the calculation, the horizontal projection method is used, and two parts are set: effective strike and coordinated strike. The effective strike is determined by the distance between the position of the red drone and the blue target; when multiple drones of the red side have caused effective strikes on the blue target, and the time difference between the strikes of the two drones is less than 2 seconds, it is determined to constitute a coordinated strike.

[0062] (4) Tit for tat

[0063] Tit-for-tat is to divide multiple drones into two confrontation clusters, requiring them to attack and defend key points as the red team and the blue team respectively within a specified time; set up a confrontation area in the field unit, such as Figure 6 As shown, the confrontation area is a circular area with a radius of 12 meters. The red and blue sides should have at least 2 drones. The red side is the defender, and its base is located in the center of the confrontation area. The blue side is the attacker. Initially, the red side is located in a warning area with a radius of 6 meters centered on the base, and the blue side's drones are randomly distributed outside the warning area and within the confrontation area. The blue side's confrontation goal is to break through the red side's interception, destroy the red side's base, and successfully survive (escape from the confrontation area), while the red side's confrontation goal is to "destroy" the enemy's drones and protect its own base.

[0064] Scoring criteria:

[0065] To simplify the calculation, horizontal projection is used, and three parts are set: successful interception, successful penetration, and survival after penetration. When the defender shoots down the attacker's drone, it is considered a successful interception. When the horizontal distance between the attacker's drone and the defender's base is less than 1 meter, the attacker's drone is considered to have successfully penetrated. Each attacker's drone can only be counted as having successfully penetrated once. Judgment of survival after penetration: When a drone of the attacker successfully penetrates and is not hit before escaping from the confrontation area, it is considered to have survived the penetration.

[0066] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A simulation demonstration and evaluation system for drone swarm game algorithm, characterized in that: It includes site unit, algorithm verification platform, simulated confrontation target, and demonstration evaluation unit, among which: The venue unit is selected indoors or in an open area, and guardrails are arranged around the venue unit; the red team and the blue team are arranged according to the demonstration subjects set by the competition, such as fixed-point strike, route interception, coordinated round-up, and tit-for-tat; The UAV is selected as the algorithm verification platform for the swarm game algorithm; the ground remote-controlled unmanned vehicle is selected as the simulated confrontation target; The demonstration and evaluation unit includes a positioning module, a communication module and scoring software. The positioning module adopts the UWB positioning system, including a positioning base station and a UWB tag. The communication module adopts a Wi-Fi device. The scoring software is used to record all data of the algorithm verification platform in each demonstration subject and score according to the preset scoring criteria. At the same time, real-time demonstration and playback of the corresponding subjects can be carried out based on the recorded data.

2. The simulation demonstration and evaluation system of the drone swarm game algorithm according to claim 1 is characterized in that: The unit area of ​​the site shall not be less than 50 meters in length and 50 meters in width, and shall be constructed with square tube trusses; the take-off and landing area, assembly and waiting area, no-fly / avoidance area, and confrontation / mission area of ​​the drone shall be set up according to the requirements of the mission scenario.

3. The simulation demonstration and evaluation system of the drone swarm game algorithm according to claim 1 is characterized in that: The drone should be a pure electric quad-rotor drone. The drone must be fully autonomously controlled throughout the competition, and the flight speed should not exceed 5m / s. The drone is equipped with a positioning module, and the location information recorded by the module will be transmitted to the competition venue LAN. The drone obtains the real-time location of its own drone, other drones and unmanned vehicles published in the competition venue LAN by subscribing to ROS messages in the competition venue LAN; the maximum movement speed of the ground remote-controlled unmanned vehicle is not less than 3m / s, and it can carry no less than two UWB tags at the same time; it has omnidirectional driving function, program-controlled driving and adjustable speed function; at the same time, it can be equipped with a positioning module to report location information to the competition venue LAN in real time.

4. The simulation demonstration and evaluation system of the drone swarm game algorithm according to claim 1 is characterized in that: The UWB positioning system adopts 2D positioning. The positioning base station of the UWB positioning system is deployed outside the site unit, on road lamp poles, building roofs, container tops or constructed poles. The UWB positioning system can meet the needs of indoor and outdoor use, and can support no less than 5 teams to use it at the same time. The positioning coverage area is not less than 100 meters × 100 meters.

5. The simulation demonstration and evaluation system of the drone swarm game algorithm according to claim 1 is characterized in that: The time data recorded by the scoring software should be accurate to 1 second, the position calculation should be accurate to 1mm, and the angle calculation should be accurate to 1°, and it should be able to calculate the coordinated attack angle and direction.

6. The simulation demonstration and evaluation system of the drone swarm game algorithm according to claim 1 is characterized in that: The fixed-point strike is a direct collision and coordinated strike by multiple drones of the Red side against the Blue side's target at a fixed position in the field unit; the Blue side's target position is randomly set within the preset range of the field unit; Scoring criteria: There are two parts of scoring: effective strike and coordinated strike, among which coordinated strike is further divided into time coordination and angle coordination. The effective strike is determined by the distance between the red team's drone position and the blue team's target position. When multiple red team drones have caused effective strikes on the blue team's target, it is determined whether time coordination has been constituted by calculating whether the time between the previous and subsequent strikes meets the preset coordination time. When multiple red team drones constitute time coordination, the angle formed by the velocity vectors of two of the drones that constitute time coordination when they strike the blue team's target is recorded as the coordinated strike angle, and it is determined whether the coordinated strike angle meets the preset coordination angle requirements, thereby determining whether the two drones constitute angle coordination.

7. The simulation demonstration and evaluation system of the drone swarm game algorithm according to claim 1 is characterized in that: Route interception is a direct collision and coordinated attack by multiple drones of the Red side on the Blue side targets moving along a preset trajectory in the field unit; Scoring criteria: It is divided into three parts: effective strike, head-on strike and coordinated strike. The effective strike is determined by the distance between the red team's drone position and the blue team's target position. When a drone effectively strikes the blue team's target, the angle between the drone's speed vector and the blue team's target speed vector is calculated. When the angle is obtuse, the strike is considered a head-on strike. When multiple drones effectively strike the blue team's target, and the time difference between the strikes of the two drones is less than the preset time, it is considered a coordinated strike.

8. The simulation demonstration and evaluation system of the drone swarm game algorithm according to claim 1 is characterized in that: Coordinated capture is a time-coordinated attack by multiple drones of the Red side on the Blue side targets that are running the game escape strategy in the field unit. The Blue side targets can autonomously avoid the attacks of the Red side drones; Scoring criteria: It is set up in two parts: effective strike and coordinated strike. The effective strike is determined by the distance between the red side's drone position and the blue side's target position. When multiple drones of the red side have caused effective strikes on the blue side's target, and the time difference between the strikes of the two drones is less than the preset time, it is determined to have constituted a coordinated strike.

9. The simulation demonstration and evaluation system of the drone swarm game algorithm according to claim 1 is characterized in that: Tit-for-tat is to divide multiple drones into two confrontation clusters, and require them to attack and defend key points as the red team and the blue team respectively within the specified time; set up a confrontation area in the field unit, the red team is the defender, and its base is located in the center of the confrontation area, and the blue team is the attacker; initially, the red team is in the warning area, and the blue team's drones are randomly distributed outside the warning area; the blue team's confrontation goal is to break through the red team's interception, destroy the red team's base, and successfully escape the confrontation area, and the red team's confrontation goal is to destroy the enemy's drones and protect its own base; Scoring criteria: Set up three parts: successful interception, successful penetration, and survival after penetration; when the defender shoots down the attacker's drone, it is considered a successful interception; when the horizontal distance between the attacker's drone and the defender's base is less than the preset distance, the attacker's drone is considered to have successfully penetrated, and each attacker's drone can only be recorded as a successful penetration once; judgment of survival after penetration: when a drone of the attacker successfully penetrates and is not hit before escaping from the confrontation area, it is considered to have survived the penetration.

10. The simulation demonstration and evaluation system of the drone swarm game algorithm according to claim 9 is characterized in that: In the tit-for-tat demonstration subject, the confrontation area is a circular area with a radius of 12 meters. The red and blue sides should have at least 2 drones, and the blue drones are randomly distributed in the area 6 meters outside the red base.

Images