Method, device and equipment for evaluating the collaborative formation capability of drone swarms

Abstract

The present invention discloses a method, device and equipment for evaluating the collaborative formation capability of a drone cluster, and relates to the field of drone technology. The method comprises: obtaining a test task for a drone cluster, the drone cluster comprises at least two drones, and the test task comprises at least one expected formation; running the test task; wherein, during the running of the test task, the drone cluster flies in accordance with at least one expected formation in sequence; during the flight of the drone cluster, the centroid of the drone cluster is obtained, and a formation coordinate system is constructed with the centroid as the origin; wherein the formation coordinate system is a three-dimensional coordinate system, and the Z axis of the formation coordinate system is parallel to the gravity direction of the drone; based on at least one expected formation and the formation coordinates of each drone, the collaborative formation capability of the drone cluster is evaluated, and the formation coordinates of the drone refer to the coordinates of the drone in the formation coordinate system. The present invention realizes a flexible evaluation of the collaborative formation capability of the drone cluster.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles (UAVs), and in particular to a method, device and equipment for evaluating the collaborative formation capability of a UAV cluster. Background Art

[0002] Multi-UAV collaborative formation means that the UAVs have computing and decision-making capabilities, can independently undertake part of the system tasks, perceive and interact with surrounding information, coordinate with each other, gradually optimize during movement, make the results consistent, and complete the entire formation control task.

[0003] The coordinated formation of multiple drones is a manifestation of a self-organizing system. It can be simply described as multiple drones distributed in various locations. Drones can only sense drones within their own communication range and cannot obtain the location information of other drones. Multiple drones move toward the target position in the formation by exchanging information with their neighbors until all drones reach the target position to form a target formation. At present, there are many studies on single key technologies of drone clusters (clusters composed of multiple drones), but there is a lack of research on the evaluation of the coordinated formation capabilities of drone clusters.

[0004] In the study of UAV formation, Desai et al. from Drexel University proposed two types of feedback controllers to maintain the formation of mobile robot teams. The two controllers correspond to two types of two-dimensional plane descriptions of formations. One is the l-ψ method, which uses the relative distance and angle between each UAV in the formation and the formation reference point to describe the formation; the other is the ll method, which uses the distance difference between the position of each UAV and the formation reference point on the coordinate axis to describe the formation.

[0005] However, both of the above methods are based on two-dimensional space, and require the drone cluster to be on the same plane to describe the formation in order to evaluate the coordinated formation capability of the drone cluster. The evaluation implementation method is not flexible enough. Summary of the invention

[0006] In order to overcome the shortcomings of the prior art, the present invention proposes a method, device and equipment for evaluating the collaborative formation capability of a drone cluster. The method constructs a three-dimensional formation coordinate system based on the centroid of the drone cluster to describe the positions of the drones, breaking the spatial limitations of the collaborative formation capability evaluation and realizing flexible evaluation of the collaborative formation capability of the drone cluster.

[0007] The technical solution of the present invention includes:

[0008] A method for evaluating the collaborative formation capability of a drone cluster, the method comprising:

[0009] Acquire a test task for a drone cluster, the drone cluster including at least two drones, and the test task including at least one expected formation;

[0010] Running the test task; wherein, during the running of the test task, the drone cluster flies in sequence according to the at least one expected formation;

[0011] During the flight of the drone cluster, the centroid of the drone cluster is obtained, and a formation coordinate system is constructed with the centroid as the origin; wherein the formation coordinate system is a three-dimensional coordinate system, and the Z axis of the formation coordinate system is parallel to the gravity direction of the drone;

[0012] Based on the at least one expected formation shape and the formation coordinates of each of the drones, the collaborative formation capability of the drone cluster is evaluated, where the formation coordinates of the drones refer to the coordinates of the drones in the formation coordinate system.

[0013] Optionally, obtaining the centroid of the drone cluster includes:

[0014] For each drone in the drone cluster, the coordinates of the center of the drone in the world coordinate system are used as the actual coordinates of the drone;

[0015] Obtaining the actual coordinate values ​​of the same dimension in the actual coordinates of all drones in the drone cluster, and calculating the average of the actual coordinate values ​​of the corresponding dimension;

[0016] The average of the actual coordinate values ​​of each dimension is taken as the actual coordinate value of the centroid in the corresponding dimension, and the actual coordinate values ​​of the centroid in each dimension are combined into the actual coordinates of the centroid.

[0017] Optionally, after constructing the formation coordinate system with the centroid as the origin, the method further includes:

[0018] For a target UAV in the UAV cluster, calculating a difference between the actual coordinate values ​​of the target UAV and the actual coordinates of the centroid in the same dimension;

[0019] The actual coordinate value difference of each dimension is used as the coordinate value of the corresponding dimension of the target UAV in the formation coordinate system, and the coordinate values ​​of each dimension of the target UAV in the formation coordinate system are combined into the formation coordinates of the target UAV.

[0020] Optionally, the evaluating the collaborative formation capability of the drone cluster based on the at least one expected formation shape and the formation coordinates of each of the drones includes:

[0021] When the drone cluster flies according to each of the at least one expected formation formations, respectively obtaining the formation coordinates of the reference drones in the corresponding expected formation formations;

[0022] Performing weighted summation processing on the coordinate values ​​of all dimensions in the formation coordinates of each drone to obtain the formation position quantity of the corresponding drone, and sorting the formation position quantities of all drones in the drone cluster according to the target order to obtain the drone formation position order;

[0023] Performing weighted summation processing on the coordinate values ​​of all dimensions in the formation coordinates of each reference UAV to obtain the formation position quantity of the corresponding reference UAV, and sorting the formation position quantities of all reference UAVs in the corresponding expected formation formation according to the target order to obtain the reference UAV formation position order;

[0024] Matching each drone with a ranking in the drone formation position sequence with a reference drone with the same ranking in the reference drone formation position sequence;

[0025] Based on the formation coordinates of each of the drones and the formation coordinates of the reference drones corresponding to the drones, the collaborative formation capability of the drone cluster is evaluated.

[0026] Optionally, the evaluating the collaborative formation capability of the drone cluster based on the formation coordinates of each of the drones and the formation coordinates of a reference drone corresponding to the drone includes:

[0027] When the drone cluster flies according to the target expected formation formation in the at least one expected formation formation, calculating the coordinate error value between the formation coordinates of each of the drones and the formation coordinates of the reference drone corresponding to the drone;

[0028] If the coordinate error values ​​corresponding to all drones in the drone cluster are less than the error value threshold, the formation stabilization time is started;

[0029] If the timing duration of the formation stabilization time is greater than the preset duration, determining that the drone cluster forms the target expected formation;

[0030] Based on the moment when the drone cluster forms the target expected formation and the start moment when the drone cluster flies according to the target expected formation, the formation time taken by the drone cluster to form the target expected formation is calculated; wherein the formation time is inversely proportional to the coordinated formation capability of the drone cluster;

[0031] Start the formation holding time timing; during the formation holding time timing process, if the coordinate error value corresponding to the drone in the drone cluster is greater than or equal to the error value threshold, end the formation holding time timing; use the timing duration of the formation holding time as the formation holding time for the drone cluster to form the target expected formation formation; wherein the formation holding time is positively proportional to the collaborative formation capability of the drone cluster.

[0032] Optionally, when the drone cluster flies according to each of the at least one expected formation formations, before respectively acquiring the formation coordinates of the reference drones in the corresponding expected formation formations, the method further includes:

[0033] Determine whether any drone in the drone cluster has a malfunction;

[0034] If a UAV fails, the reference UAV with the same number as the failed UAV is deleted from the corresponding expected formation, and the corresponding expected formation and the formation coordinates of each reference UAV in the corresponding expected formation are updated based on the remaining reference UAVs in the corresponding expected formation;

[0035] Based on the remaining drones in the drone cluster except the malfunctioning drone, the centroid of the drone cluster and the formation coordinate system are updated, and the formation coordinates of the remaining drones are updated according to the updated centroid and formation coordinate system.

[0036] Optionally, after obtaining the centroid of the drone cluster and constructing a formation coordinate system with the centroid as the origin, the method further includes:

[0037] Selecting a reference drone from the drone cluster based on the formation coordinates of each of the drones;

[0038] Calculating polar angles between the remaining drones in the drone cluster, except the reference drone, and the reference drone;

[0039] According to the value of the polar angle, the remaining drones except the reference drone are sorted in order from small to large, so as to obtain the polar angle position order of the drones;

[0040] Starting from the reference UAV, recursively traverse three adjacent UAVs in the polar angle position sequence of the UAVs counterclockwise, and calculate the rotation directions of two vectors formed by the middle UAV of the three adjacent UAVs as the origin;

[0041] If the rotation direction is less than or equal to 180 degrees, the middle UAV is used as the vertex of the outer polygon of the projection of the airspace occupied by the UAV cluster;

[0042] If the rotation direction is greater than 180 degrees, the middle drone is deleted from the drone polar angle position sequence, and the three adjacent drones in the drone polar angle position sequence are continuously recursively traversed counterclockwise;

[0043] Until all the drones in the polar angle position sequence of the drones are traversed, all vertices of the outer polygon of the projection of the airspace occupied by the drone cluster are obtained;

[0044] Based on the plane coordinates of the drones corresponding to all vertices, calculate the airspace projection area occupied by the drone cluster;

[0045] The ratio of the total number of drones in the drone cluster to the projected area of ​​the airspace occupied by the drone cluster is used as the formation density of the drone cluster; wherein the formation density is inversely proportional to the collaborative formation capability of the drone cluster.

[0046] Optionally, after obtaining the centroid of the drone cluster and constructing a formation coordinate system with the centroid as the origin, the method further includes:

[0047] Obtain the number of collision drones in the drone cluster within the target duration;

[0048] Based on the target duration, the number of the colliding drones and the total number of drones in the drone cluster, the formation time mutual interference probability density is calculated; wherein the formation time mutual interference probability density is inversely proportional to the cooperative formation capability of the drone cluster;

[0049] Based on the projected area of ​​the airspace occupied by the drone cluster, the number of the colliding drones and the total number of drones in the drone cluster, the probability density of mutual interference in the formation space is calculated; wherein the probability density of mutual interference in the formation space is inversely proportional to the collaborative formation capability of the drone cluster.

[0050] A device for evaluating the collaborative formation capability of a drone cluster, the device comprising:

[0051] A task acquisition module, used to acquire a test task for a drone cluster, wherein the drone cluster includes at least two drones, and the test task includes at least one expected formation;

[0052] A task running module, used to run the test task; wherein, during the running of the test task, the drone cluster flies in sequence according to the at least one expected formation;

[0053] A coordinate system construction module, used to obtain the centroid of the drone cluster during the flight of the drone cluster, and to construct a formation coordinate system with the centroid as the origin; wherein the formation coordinate system is a three-dimensional coordinate system, and the Z axis of the formation coordinate system is parallel to the gravity direction of the drone;

[0054] A capability assessment module is used to assess the collaborative formation capability of the drone cluster based on the at least one expected formation shape and the formation coordinates of each of the drones, where the formation coordinates of the drones refer to the coordinates of the drones in the formation coordinate system.

[0055] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the collaborative formation capability assessment method for a drone cluster as described above.

[0056] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the computer program is loaded and executed by the processor to implement the collaborative formation capability assessment method of the above-mentioned drone cluster.

[0057] A computer program product, when the computer program product is run on a computer device, enables the computer device to execute the collaborative formation capability assessment method of the drone cluster as described above.

[0058] Compared with the prior art, the method proposed in the present invention has the following advantages and effects:

[0059] 1. Construct a formation coordinate system with the centroid of the drone cluster as the origin. When evaluating the collaborative formation capability of the drone cluster, based on the coordinates of each drone in the formation coordinate system, on the one hand, it can avoid the situation where the test evaluation cannot be carried out normally due to the failure of the drone as the origin of the coordinate system; on the other hand, the formation information can be dynamically adjusted based on the change in the number of drones in the drone cluster, effectively avoiding the situation where the test evaluation fails due to damage or disconnection of a drone. In addition, the formation coordinate system constructed by the present invention is a three-dimensional coordinate system. The position of the drone can be described without all drones in the drone cluster flying to the same two-dimensional plane, breaking the spatial limitation of the drone position description, and facilitating the continuous and real-time evaluation of the collaborative formation capability during the flight of the drone cluster, which is more in line with the evaluation of the collaborative formation capability under actual conditions. Through the use of a three-dimensional formation coordinate system, the present invention simplifies the parameters required for the evaluation of collaborative formation capability, and at the same time, the evaluation speed is stabilized at 1000FPS, which is sufficient to meet the needs of continuous and real-time evaluation;

[0060] 2. The formation position quantity is calculated using the coordinate values ​​of the formation coordinates in all dimensions, and the corresponding relationship between the drones in the drone cluster and the reference drones in the expected formation formation is obtained through the formation position quantity, which improves the accuracy of the collaborative formation capability assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A flowchart of a method for evaluating the collaborative formation capability of a drone cluster provided by an embodiment of the present invention.

[0062] Figure 2 A schematic diagram of a desired formation provided by an embodiment of the present invention.

[0063] Figure 3 A schematic diagram comparing a formation and an expected formation provided by an embodiment of the present invention.

[0064] Figure 4 A schematic diagram of a collaborative formation capability evaluation system provided by an embodiment of the present invention.

[0065] Figure 5 A schematic diagram of the projection of the airspace occupied by a drone cluster provided by an embodiment of the present invention.

[0066] Figure 6 A block diagram of a device for evaluating the collaborative formation capability of a drone cluster provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0067] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only specific embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0068] The collaborative formation capability evaluation method of the UAV cluster in the present invention is as follows: Figure 1 As shown, the process includes the following steps (steps 1 to 4).

[0069] Step 1: Obtain a test task for a drone cluster, where the drone cluster includes at least two drones, and the test task includes at least one expected formation.

[0070] Usually, the coordinated flight of multiple UAVs can reasonably allocate targets to UAVs in different positions to complete. This requires the UAV cluster to form a certain formation. The speed, quality, and stability duration of the formation formed by the UAV cluster also affect the execution quality of subsequent reconnaissance, confrontation, attack and other tasks. Therefore, the coordinated formation capability of the UAV cluster is also an important part of reflecting the coordinated capability of the UAV cluster.

[0071] In order to evaluate the cooperative formation capability of the drone cluster, the computer device first obtains a test task for the drone cluster, and the test task includes at least one expected formation. The present invention does not limit the specific method of obtaining the test task. Optionally, the computer device responds to the user's selection and input and other operation instructions to obtain the expected formation formation combination corresponding to the operation instruction as a test task; or, the computer device randomly obtains at least one expected formation formation combination from a preset database as a test task; or, the computer device obtains the expected formation formation combination matching the type of the drone cluster, the cooperative task (such as reconnaissance, confrontation, attack, etc.), the number of drones, etc. from a preset database as a test task.

[0072] For example, Figure 2 As shown in the figure, the drone cluster includes 8 drones, and the test task for the drone cluster obtained by the computer device includes 4 expected formations, namely: the cross shown in (a), the herringbone shown in (b), the arrow shape shown in (c), and the diamond shape shown in (d). Of course, in actual application, the expected formation can also be a three-dimensional formation, that is, the drones in the expected formation can be in different planes. Figure 2 For the sake of convenience of description only, the description is made by taking the expected formation as a two-dimensional formation as an example. It should be understood that this does not constitute a limitation of the present invention.

[0073] Step 2: Run the test task; during the test task running process, the drone cluster flies in turn according to at least one expected formation.

[0074] After obtaining the test task for the drone cluster, the computer device runs the test task to obtain test data in real time to evaluate the collaborative formation capability of the drone cluster. Among them, since the test task for the drone cluster includes at least one expected formation, during the running of the test task, the drone cluster autonomously controls and makes decisions to sequentially form at least one expected formation. Optionally, the computer device can divide the test task into a corresponding number of sub-test tasks according to the number of expected formations, each sub-test task includes an expected formation, and the subsequent computer device runs the sub-test tasks in sequence, so that the drone cluster sequentially forms the expected formation in the sub-test tasks.

[0075] The present invention does not limit the running order of the above-mentioned sub-test tasks or the formation order of the expected formation formation. In one example, the computer device randomly obtains and runs the sub-test task, or the drone cluster randomly flies in formation according to the expected formation formation during the computer device running the test task; in another example, the computer device obtains and runs the corresponding sub-test tasks in sequence according to the order of the expected formation formation from simple to complex, or the drone cluster randomly flies in formation according to the order of the expected formation formation from simple to complex during the computer device running the test task.

[0076] Step 3: During the flight of the drone cluster, obtain the centroid of the drone cluster and construct a formation coordinate system with the centroid as the origin; wherein the formation coordinate system is a three-dimensional coordinate system, and the Z axis of the formation coordinate system is parallel to the gravity direction of the drone.

[0077] In order to continuously obtain the position of each drone in real time and realize the dynamic description of the formation of the drone cluster, during the flight of the drone cluster, the computer equipment obtains the centroid of the drone cluster in real time and constructs a formation coordinate system with the centroid as the origin. Among them, the centroid of the drone cluster refers to the geometric center of the drone cluster. During the flight of the drone cluster, the centroid may coincide with the center of a drone in the drone cluster. Compared with constructing a coordinate system with the center of a drone as the origin to describe the position of each drone, the present invention constructs a formation coordinate system with the centroid of the drone cluster as the origin. On the one hand, it can avoid the situation where the test evaluation cannot be carried out normally due to the failure of the drone as the origin of the coordinate system; on the other hand, it can dynamically adjust the formation information based on the change in the number of drones in the drone cluster, effectively avoiding the failure of the test evaluation due to the damage or disconnection of a drone.

[0078] In addition, in order to describe the position of the drone at any time during the flight of the drone cluster, in the present invention, the formation coordinate system constructed by the computer device is a three-dimensional coordinate system, so that the position of the drone can be described without all the drones in the drone cluster flying to the same two-dimensional plane, breaking the spatial limitation of the drone position description. Among them, the Z axis of the formation coordinate system is parallel to the gravity direction of the drone, that is, the Z axis of the formation coordinate system is vertically downward or vertically upward; the plane formed by the X axis and Y axis of the formation coordinate system is perpendicular to the Z axis of the formation coordinate system, optionally, the X axis of the formation coordinate system is the flight forward direction of the drone cluster, and the Y axis is perpendicular to the flight forward direction of the drone cluster, or the Y axis of the formation coordinate system is the flight forward direction of the drone cluster, and the X axis is perpendicular to the flight forward direction of the drone cluster.

[0079] In one example, the above-mentioned process of obtaining the centroid of the drone cluster includes the following sub-steps (steps 3.1 to 3.3).

[0080] Step 3.1: For each drone in the drone cluster, the coordinates of the center of the drone in the world coordinate system are used as the actual coordinates of the drone.

[0081] Drones are usually large in size and complex in shape. In order to abstractly describe the location of a drone, it is necessary to obtain spatial points related to the drone to determine the coordinates of the drone. Since the formation coordinate system uses the centroid of the drone cluster as the coordinate origin, in order to ensure the accuracy of the calculation, the computer device calculates the centroid of the drone cluster through the center of each drone in the drone cluster. Based on this, for each drone, the computer device uses the coordinates of the center of the drone in the world coordinate system as the actual coordinates of the drone. Among them, the center of the drone can also be called the centroid of the drone, which refers to the geometric center of the drone, which may coincide with the center of mass or center of gravity of the drone.

[0082] Step 3.2: Obtain the actual coordinate values ​​of the same dimension in the actual coordinates of all drones in the drone cluster, and calculate the mean of the actual coordinate values ​​of the corresponding dimension.

[0083] For each dimension of the actual coordinates, the computer device obtains the actual coordinate values ​​of all drones in the drone cluster in that dimension from the actual coordinates, and calculates the mean of the actual coordinate values ​​of that dimension through averaging processing.

[0084] Step 3.3: Take the mean of the actual coordinate values ​​of each dimension as the actual coordinate value of the centroid in the corresponding dimension.

[0085] The average of the actual coordinate values ​​of all drones in the drone cluster in each dimension is the actual coordinate value of the centroid of the drone cluster in the corresponding dimension. The actual coordinate values ​​of the centroid in each dimension are combined to form the actual coordinates of the centroid.

[0086] For example, the number of drones in the drone cluster is N (N is a positive integer greater than 1), and F xyz To represent the actual coordinates of each drone in the drone cluster, where x i ,y i and z i represents the actual coordinate value of the center of the i-th (i is a positive integer less than or equal to N) drone in the drone cluster in each dimension, then F xyz The expression formula is as follows.

[0087]

[0088] Based on this, the calculation formula for the mean of the actual coordinate values ​​of the three dimensions in the actual coordinates is as follows:

[0089]

[0090]

[0091]

[0092] Assume that through C xyz To represent the actual coordinates of the centroid of the drone cluster, then C xyz The expression formula is as follows.

[0093]

[0094] After calculating the actual coordinates of the centroid of the drone cluster, the coordinates of each drone in the formation coordinate system with the centroid as the origin can be calculated by comparing the actual coordinates of the centroid with the actual coordinates of each drone. That is, after constructing the formation coordinate system with the centroid as the origin, it also includes: for the target drone in the drone cluster, calculating the difference in the actual coordinate values ​​of the same dimension between the target drone and the actual coordinates of the centroid; using the difference in the actual coordinate values ​​of each dimension as the coordinate value of the corresponding dimension of the target drone in the formation coordinate system. Among them, the coordinate values ​​of each dimension of the target drone in the formation coordinate system are combined into the formation coordinates of the target drone, and the formation coordinates of the target drone refer to the coordinates of the target drone in the formation coordinate system.

[0095] For example, it is assumed that the actual coordinates of each drone in the drone cluster are the above F xyz The centroid of the drone cluster is C xyz ,pass To represent the formation coordinates of each drone in the drone cluster, through x i -c x ,y i -c y and z i -c z represents the coordinate values ​​of each dimension of the i-th UAV in the UAV cluster in the formation coordinate system, then The expression formula is as follows.

[0096]

[0097] Step 4: Based on at least one expected formation and the formation coordinates of each UAV, evaluate the coordinated formation capability of the UAV cluster; the formation coordinates of the UAV refer to the coordinates of the UAV in the formation coordinate system.

[0098] The formation coordinates of each drone can reflect the formation of the drone cluster. During the flight of the drone cluster, by comparing the formation of the drone cluster with at least one expected formation in the test task, the duration of the drone cluster forming the expected formation, the duration of the expected formation after the formation, the error between the expected formation and the expected formation after the formation, the number of expected formations that can be formed, etc. can be obtained. The coordinated formation capability of the drone cluster is evaluated through these data. The present invention does not limit the specific comparison method of the formation formation of the drone cluster and the expected formation formation. Optionally, the computer device can first calculate the formation formation of the drone cluster based on the formation coordinates of each drone, and then compare the shape, size, etc. of the formation formation of the drone cluster with the expected formation formation; or, the computer device can first calculate the formation coordinates of each reference drone in the expected formation formation according to the expected formation formation, and then compare the formation coordinates of each drone with the formation coordinates of each reference drone.

[0099] In one example, the above step 4 includes the following sub-steps (steps 4.1 to 4.5).

[0100] Step 4.1: When the drone cluster flies according to each expected formation formation in at least one expected formation formation, the formation coordinates of the reference drones in the corresponding expected formation formation are respectively obtained.

[0101] The present invention refers to the drones in the expected formation as reference drones to distinguish them from the drones in the drone cluster. In order to compare the expected formation with the formation of the drone cluster, on the one hand, during the flight of the drone cluster, the computer device acquires the formation coordinates of each drone in real time (including at intervals of 0.01 seconds, 0.1 seconds, 1 second, etc.); on the other hand, when the drone cluster flies according to a certain expected formation (that is, the drone cluster cooperates to form a certain expected formation), the computer device also needs to acquire the formation coordinates of each reference drone in the expected formation. Optionally, the formation coordinates of each reference drone can be pre-set and stored and directly obtained by the computer device, such as the formation coordinates of each reference drone in the expected formation and the expected formation are preset in the database together, and when the computer device obtains the expected formation from the preset database, it also obtains the formation coordinates of each reference drone in the expected formation; or, the formation coordinates of each reference drone are obtained by calculation by the computer device, such as when the computer device runs a test task according to a certain expected formation, or when it is about to run a test task according to a certain expected formation, it calculates the formation coordinates of each reference drone in the expected formation. Among them, the calculation method of the formation coordinates of each reference drone is similar to the calculation method of the formation coordinates of each drone, that is, first obtain the centroid of the expected formation and construct a formation coordinate system with the centroid as the origin, and then calculate the coordinates of each reference drone in the formation coordinate system. The specific calculation process can refer to the calculation method of the formation coordinates of each drone, which will not be repeated here.

[0102] In the actual test process, a drone cluster may have a situation where a drone is damaged or disconnected. In this case, if the test task is still performed according to the expected formation obtained for the initial drone cluster, the drone cluster will not be able to form the expected formation, and the computer equipment will not be able to evaluate the collaborative formation capabilities of the remaining drones in the drone cluster, which will lead to evaluation failure. Based on this, optionally, in order to improve the success rate and accuracy of the evaluation, before the above step 4.1, it also includes: judging whether there is a drone failure in the drone cluster (including drone damage, disconnection, failure, etc.); if there is a drone failure, deleting the reference drone with the same number as the failed drone from the corresponding expected formation, and based on the remaining reference drones in the corresponding expected formation, updating the corresponding expected formation and the formation coordinates of each reference drone in the corresponding expected formation; based on the remaining drones in the drone cluster except the failed drone, updating the centroid and formation coordinate system of the drone cluster, and updating the formation coordinates of the remaining drones according to the updated centroid and formation coordinate system.

[0103] Step 4.2: Perform weighted summation on the coordinate values ​​of all dimensions in the formation coordinates of each drone to obtain the formation position quantity of the corresponding drone, and sort the formation position quantities of all drones in the drone cluster according to the target order to obtain the drone formation position order.

[0104] In actual testing, there may be a situation where the UAV cluster’s formation is consistent with the expected formation in terms of shape and size, but the numbers of the UAVs at two locations are opposite to those of the reference UAVs. Usually, in this case, the UAV cluster is also considered to have formed the expected formation. For example, Figure 3 As shown in the figure, a UAV cluster includes 4 UAVs and the expected formation is a diamond. Figure 3 In the figure, (1) is a schematic diagram of the desired formation. Assume that during the flight of the drone cluster, the drone cluster forms the formation shown in (2). The formation shown in (2) is consistent with the desired formation shown in (1) in terms of shape and size. However, the drones numbered B and C in the formation shown in (2) are in the opposite position to the reference drones numbered B and C in the desired formation shown in (1). If the correspondence between the drones in the drone cluster and the reference drones in the desired formation is obtained by numbering when comparing the formation of the drone cluster with the desired formation, in the above case, the computer device will think that the drone cluster has not formed the desired formation, thereby affecting the accuracy of the evaluation.

[0105] In order to improve the accuracy of collaborative formation capability assessment, in the present invention, the computer device calculates the formation position quantity using the coordinate values ​​of the formation coordinates in all dimensions, and obtains the correspondence between the drones in the drone cluster and the reference drones in the expected formation formation through the formation position quantity.

[0106] Based on this, on the one hand, the computer device performs weighted summation processing on the coordinate values ​​of all dimensions in the formation coordinates of each drone in the drone cluster to obtain the formation position quantity of the corresponding drone. Optionally, when performing the weighted summation processing of the coordinate values, the weights of the coordinate values ​​of different dimensions can be determined according to the expected formation formation. In order to enable the formation position quantity to accurately reflect and distinguish the position characteristics between different drones, the weight corresponding to the dimension that has a greater impact on the shape of the expected formation formation can be appropriately increased. For example, for a triangular formation flying along the x-axis, the x-axis may not be able to effectively distinguish two drones flying side by side, and the y-axis has a more obvious position feature than the x-axis, then the weight corresponding to the y-axis can be set larger (such as setting the weight corresponding to the y-axis to 0.5, the weight corresponding to the x-axis to 0.2, and the weight corresponding to the z-axis to 0.3). After calculating the formation position quantities of all drones in the drone cluster, the computer device sorts the formation position quantities according to the target order to obtain the drone formation position order. Among them, the target order is the formation position quantity from large to small, or the formation position quantity from small to large.

[0107] Step 4.3: Perform weighted summation on the coordinate values ​​of all dimensions in the formation coordinates of each reference UAV to obtain the formation position quantity of the corresponding reference UAV, and sort the formation position quantities of all reference UAVs in the corresponding expected formation formation according to the target order to obtain the reference UAV formation position order.

[0108] On the other hand, the computer device performs weighted summation processing on the coordinate values ​​of all dimensions in the formation coordinates of each reference drone in the expected formation formation to obtain the formation position quantity of the corresponding reference drone. It should be understood that the weights of the coordinate values ​​of each dimension when the coordinate values ​​are weighted summed in step 4.3 are the same as the weights of the coordinate values ​​of each dimension when the coordinate values ​​are weighted summed in the above step 4.2. After calculating the formation position quantities of all reference drones in the expected formation formation, the computer device sorts the formation position quantities according to the target order to obtain the reference drone formation position order. It should be understood that the target order based on which the formation position quantities are sorted in step 4.3 is also the same as the target order based on which the formation position quantities are sorted in step 4.2.

[0109] It should be noted that the present invention does not limit the execution order of the above steps 4.2 and 4.3. Optionally, step 4.2 is executed before step 4.3, or after step 4.3, or simultaneously with step 4.3. The above embodiment is only for the convenience of description, and step 4.2 is described before step 4.3. It should be understood that this does not constitute a limitation of the present invention.

[0110] Step 4.4: Match each ranked drone in the drone formation position sequence with the reference drone of the same ranking in the reference drone formation position sequence.

[0111] By comparing the ranking of the drone formation position sequence and the reference drone formation position sequence, the corresponding relationship between each drone in the drone cluster and each reference drone in the expected formation formation can be determined. Specifically, the computer device establishes a corresponding relationship between drones and reference drones that are in the same ranking in the drone formation position sequence and the reference drone formation position sequence.

[0112] Step 4.5: Based on the formation coordinates of each UAV and the formation coordinates of the reference UAV corresponding to the UAV, the coordinated formation capability of the UAV cluster is evaluated.

[0113] By comparing the formation coordinates of each UAV and the formation coordinates of the reference UAV corresponding to the UAV, the computer equipment can obtain the time it takes for the UAV cluster to form the expected formation, the time it maintains the expected formation after forming, the error between the expected formation and the expected formation after forming, the number of expected formations that can be formed, etc. These data can be used to evaluate the collaborative formation capability of the UAV cluster.

[0114] In one example, the above step 4.5 includes: when the drone cluster flies in accordance with the target expected formation formation in at least one expected formation formation, calculating the coordinate error value between the formation coordinates of each drone and the formation coordinates of the reference drone corresponding to the drone; if the coordinate error values ​​corresponding to all drones in the drone cluster are less than the error value threshold, starting the formation stabilization time timing; if the timing duration of the formation stabilization time is greater than the preset duration, determining that the drone cluster forms the target expected formation formation; based on the moment when the drone cluster forms the target expected formation formation and the starting moment when the drone cluster flies in accordance with the target expected formation formation, calculating the formation time taken for the drone cluster to form the target expected formation formation.

[0115] Wherein, the calculation method of the coordinate error value includes but is not limited to: the Euclidean distance, Manhattan distance, Chebyshev distance, etc. between the two formation coordinates. The error value threshold can be flexibly set according to the different requirements for the coordinated formation capability of the drone cluster. For example, in the case of strict requirements for the coordinated formation capability, a smaller error value threshold is pre-set. When the coordinate error values ​​corresponding to all drones in the drone cluster are less than the error value threshold, it is considered that the drone cluster forms the desired formation formation. However, in order to avoid the drone cluster only forming the desired formation formation by chance, the present invention also sets a formation stabilization time timing, such as starting the formation stabilization time timer. Only when the timing duration of the formation stabilization time is greater than the preset duration, it is determined that the drone cluster actually forms the desired formation formation, and the formation stabilization time timing is stopped. Wherein, the preset duration can also be flexibly set according to the different requirements for the coordinated formation capability of the drone cluster. For example, in the case of strict requirements for the coordinated formation capability, a larger preset duration is pre-set. Afterwards, the computer device calculates the time difference between the moment when the drone cluster actually forms the desired formation formation and the starting moment when the drone cluster flies according to the desired formation formation, and the formation time taken for the drone cluster to form the desired formation formation can be obtained. Among them, the formation time is inversely proportional to the coordinated formation capability of the drone cluster. That is, the longer the formation time is, the lower the coordinated formation capability is; the shorter the formation time is, the higher the coordinated formation capability is.

[0116] For example, taking the drone swarm flying in the target expected formation as an example, the error value threshold is set to 2 meters and the preset duration is 3 seconds. If the coordinate error values ​​of all drones in the drone swarm are less than 2 meters and the formation is maintained for more than 3 seconds, it is considered that the drone swarm actually forms the target expected formation. Assume that the time when the drone swarm actually forms the target expected formation is T s The starting time for the drone cluster to fly in the target formation is T I , then the formation time for the drone cluster to form the target expected formation is T PS The calculation formula is as follows.

[0117] T PS =T S -T I

[0118] One point that needs to be explained is that during the timing of the formation stabilization time (the timing duration of the formation stabilization time is less than or equal to the preset duration), if there is a coordinate error value corresponding to a drone in the drone cluster that is greater than or equal to the error value threshold, the timing condition of the formation stabilization time is reset. In other words, the computer device re-judges whether the coordinate value errors corresponding to all drones are less than the error value threshold. If so, the formation stabilization time timing is restarted.

[0119] In another example, after the above example determines that the drone cluster forms the target desired formation, it also includes: starting the formation holding time timing; during the formation holding time timing process, if the coordinate error value corresponding to the drone in the drone cluster is greater than or equal to the error value threshold, then ending the formation holding time timing; using the timing duration of the formation holding time as the formation holding time for the drone cluster to form the target desired formation. Among them, the formation holding time is in positive proportion to the coordinated formation capability of the drone cluster, that is, the longer the formation holding time, the higher the coordinated formation capability; the shorter the formation holding time, the lower the coordinated formation capability.

[0120] For example, it is assumed that the time when the drone cluster actually forms the target desired formation is T S , computer equipment in T S The formation holding time is started at time T. During the formation holding time, if it is detected that the coordinate error value of a drone in the drone cluster is greater than or equal to the error value threshold, the formation holding time is terminated. It is further assumed that the end time of the formation holding time is T E , then the formation of the target expected formation of the UAV cluster is maintained for a formation duration T FK The calculation formula is as follows.

[0121] T FK =T E -T S

[0122] During the flight of the drone cluster, the computer device can also calculate the formation error of the drone cluster in real time. For example, assuming that the number of drones in the drone cluster is N, the formation coordinates of the i-th drone are The computer device obtains the expected formation currently running in the test task. The formation coordinates of the i-th reference drone in the expected formation are: Then the formation error E of the drone cluster is p The real-time calculation formula is as follows. Furthermore, the computer device can also perform statistical processing (such as averaging, weighted averaging, summing, summing after assignment conversion, etc.) on the formation error calculated during the entire test task operation to obtain a formation error statistic value, wherein the formation error statistic value is inversely proportional to the collaborative formation capability of the UAV cluster, that is, the smaller the formation error statistic value, the higher the collaborative formation capability; the larger the formation error statistic value, the lower the collaborative formation capability.

[0123]

[0124] In addition, during the calculation process of the above formation time consumption, the computer device can also simultaneously count the formation support number of the drone cluster. Exemplarily, the computer device runs the test task in accordance with at least one expected formation formation in turn, and initializes the formation support number (that is, the formation support number is set to zero) when starting to run the test task; when running the test task according to each expected formation formation, the drone cluster autonomously controls and makes decisions to form the corresponding expected formation formation; if it is determined that the drone cluster has actually formed the corresponding expected formation formation (the determination method can be found in the introduction of the above formation time consumption), the formation support number is added by one; until the test task is completed, the formation support number is output. Among them, the formation support number is in positive proportion to the collaborative formation capability of the drone cluster, that is, the larger the formation support number, the higher the collaborative formation capability; the smaller the formation support number, the lower the collaborative formation capability. Optionally, the computer device may set a maximum operating time of the expected formation. If the drone cluster does not form a corresponding expected formation within the maximum operating time, it is determined that the drone cluster does not support the corresponding expected formation, and the test task continues to run according to the next expected formation. The maximum operating times of different expected formations may be the same or different. For example, different maximum operating times may be set according to the complexity of the expected formation, and a longer maximum operating time may be set for a more complex expected formation.

[0125] In order to enrich the evaluation methods of collaborative formation capability and achieve more complete and accurate capability evaluation, the present invention also provides a fine-grained evaluation system, which further divides the collaborative formation capability into collaborative capability and formation capability. Among them, the formation capability focuses on whether the drone cluster supports the expected formation formation, and the collaborative capability focuses on the collaborative cooperation between drones (such as whether the drones collide). Figure 4 As shown in the figure, the formation capability can be evaluated by the above-mentioned formation time, formation holding time, formation error, formation support number, etc., and the coordination capability can be evaluated by formation density, formation time mutual interference probability density, formation space mutual interference probability density. Below, the calculation of each evaluation index in the coordination capability is introduced and explained.

[0126] In one example, the above step 3 further includes the following steps (steps 5.1 to 5.8).

[0127] Step 5.1: Based on the formation coordinates of each drone, select a benchmark drone from the drone cluster.

[0128] The selection of the reference drone can be based on any coordinate axis of the formation coordinate system, and the drone with the smallest coordinate value on the coordinate axis is used as the reference drone. If there are two or more drones with the smallest coordinate value on the coordinate axis, then a drone with the smallest coordinate value on the other two coordinate axes is selected as the reference drone. For example, the drone with the smallest y-axis coordinate value of the formation coordinate in the drone cluster is selected as the reference drone. If there are two or more drones with the smallest y-axis coordinate value, then the drone with the smallest x-axis coordinate value is selected as the reference drone.

[0129] Step 5.2: Calculate the polar angles between the remaining drones in the drone cluster, except the benchmark drone, and the benchmark drone.

[0130] The polar angle refers to the angle between the line from any point on the plane to the pole in the polar coordinate system and the polar axis. In this example, each drone in the drone cluster is abstracted as a point in the plane, and the pole corresponds to the above-mentioned reference drone. The computer device calculates the polar angle between the plane points corresponding to the remaining drones in the drone cluster and the pole corresponding to the reference drone.

[0131] Step 5.3: Sort the remaining drones except the reference drone in the order of polar angle values ​​from small to large to obtain the polar angle position order of the drones.

[0132] According to the polar angle values, the computer device sorts the remaining drones in the drone cluster in ascending order to obtain the polar angle position sequence of the drones. Optionally, in the polar angle position sequence, the computer device numbers the remaining drones starting from 1 to facilitate subsequent traversal processing.

[0133] Step 5.4: Starting from the reference UAV, recursively traverse the three adjacent UAVs in the order of the UAV polar angle positions counterclockwise, and calculate the rotation directions of the two vectors composed of the middle UAV of the three adjacent UAVs as the origin.

[0134] The computer device starts from the reference drone and traverses the three adjacent drones in the order of the drone polar angle position in a counterclockwise manner, that is, it starts from the three drones with the smallest polar angle values ​​(drones numbered 1, 2, and 3 in the order of the polar angle position). During each traversal, the computer device constructs two vectors with the middle drone of the three adjacent drones as the origin, that is, one vector is constructed based on the middle drone and the first drone in the order of the polar angle position, and the other vector is constructed based on the middle drone and the last drone in the order of the polar angle position. Afterwards, the computer device can calculate the rotation direction of the two vectors by vector cross multiplication.

[0135] Step 5.5: If the rotation direction is less than or equal to 180 degrees, the middle UAV is used as the vertex of the projection polygon of the airspace occupied by the UAV cluster; if the rotation direction is greater than 180 degrees, the middle UAV is deleted from the UAV polar position sequence, and the three adjacent UAVs in the UAV polar position sequence are continued to be recursively traversed counterclockwise.

[0136] When calculating the rotation direction of two vectors through vector cross multiplication, if the cross product result is positive, indicating that the rotation direction of the two vectors is less than or equal to 180 degrees, the middle UAV is used as the vertex of the outer polygon of the projection of the airspace occupied by the UAV cluster, and is added to the convex hull vertex stack for storage; if the cross product result is negative, indicating that the rotation direction of the two vectors is greater than 180 degrees, the middle UAV is located in the outer polygon of the projection of the airspace occupied by the UAV cluster, the middle UAV is deleted from the UAV polar angle position order, and the counterclockwise recursive traversal continues.

[0137] Step 5.6: Until all drones in the order of the polar angle positions are traversed, all vertices of the projection outsourcing polygon of the airspace occupied by the drone cluster are obtained.

[0138] Until all drones in the drone polar angle position sequence are traversed, the computer device uses the drones in the convex hull vertex stack as all vertices of the projection outer polygon of the airspace occupied by the drone cluster. It should be understood that the first drone and the last drone in the drone polar angle position sequence, as well as the reference drone, are saved in the convex hull vertex stack by default.

[0139] For example, Figure 5 As shown in the figure, the reference drone is U0, and the remaining drones in the order of the polar angle positions are U1 to U 13 , the computer equipment starts from U1, U2, and U3 and traverses counterclockwise; among them, the vector and If the cross product result is negative, then remove UAV U2 from the order of UAV polar angle positions, and continue to traverse U1, U3, and U4 counterclockwise; where the vector and If the cross product result is positive, UAV U3 is used as the vertex of the outer polygon of the airspace projection occupied by the UAV cluster, and U3, U4, U5, etc. are traversed counterclockwise until all UAVs in the order of the polar angle position of the UAV are traversed, and all the vertices of the outer polygon of the airspace projection occupied by the UAV cluster are obtained as U0, U1, U3, U4, U8, U 10 , U 11 , U 13 .

[0140] Step 5.7: Based on the plane coordinates of the drones corresponding to all vertices, calculate the projected area of ​​the airspace occupied by the drone cluster.

[0141] In the process of calculating the polar angle, the computer device has abstracted each drone as a point in the plane, so that the drone formation coordinates are converted into plane coordinates. After all vertices are obtained through step 5.6, the airspace projection area occupied by the drone cluster is calculated based on the plane coordinates of the drones corresponding to all vertices. For example, the airspace projection area occupied by the drone cluster R F The calculation is done by the Shoelace formula (also known as the shoelace formula or the Gaussian area formula), as shown below. The number of all vertices is M (M is an integer greater than or equal to 3), x k refers to the x-axis coordinate of the kth vertex among all vertices, y k+1 and k-1 They are the y-axis coordinates of the k+1th vertex and the k-1th vertex among all vertices respectively.

[0142]

[0143] Step 5.8: The ratio of the total number of drones in the drone cluster to the projected area of ​​the airspace occupied by the drone cluster is used as the formation density of the drone cluster; among which, the formation density is inversely proportional to the coordinated formation capability of the drone cluster.

[0144] The computer equipment uses the total number of drones in the drone cluster to divide by the projected area of ​​the airspace occupied by the drone cluster to obtain the formation density of the drone cluster. The formation density can also reflect the coordinated formation capability of the drone cluster. The smaller the formation density, the closer the distance between the drones in the drone cluster, and the easier it is for the drones to collide. Based on this, the formation density is inversely proportional to the coordinated formation capability of the drone cluster. That is, the larger the formation density, the lower the coordinated formation capability; the smaller the formation density, the higher the coordinated formation capability. For example, the formation density D F The calculation formula is as follows.

[0145]

[0146] In another example, the above step 3 further includes the following steps (steps 6.1 to 6.3).

[0147] Step 6.1: Obtain the number of collision drones in the drone cluster within the target duration.

[0148] The computer device can count the collisions of the drones within a preset target time. If each drone is bound to a collision sensor, the sensor data can be directly read to obtain the number of collision drones in the drone cluster; otherwise, the Euclidean distance between the drones can be used to determine whether a collision has occurred. Optionally, if the distance between two drones is less than the radius of the drone, the two drones are considered to have collided.

[0149] For example, the number of collision drones at time t within the target duration is taken as an example. Assume that the radius of the drone is R UAV , the number of drones in the drone cluster is N, and the Euclidean distance between the i-th drone and the j-th drone is E ij , then the Euclidean distance between any two drones in the drone cluster (including the drone and itself) can be obtained through the distance matrix D t Based on this, the statistical spacing matrix D t The value is not zero and is less than the radius R of the drone UAV The number of C t , the number C t Equal to the number of collision drones.

[0150]

[0151] C t =NUM(D, <R UAV )

[0152] Step 6.2: Based on the target duration, the number of colliding drones, and the total number of drones in the drone cluster, calculate the probability density of formation time mutual interference; among them, the probability density of formation time mutual interference is inversely proportional to the coordinated formation capability of the drone cluster.

[0153] The ratio of the number of colliding drones to the total number of drones in a drone cluster at a certain moment reflects the probability of a drone collision at that moment. The computer device averages the probability of a drone collision at each moment within the target duration, and can calculate the formation time mutual interference probability density of the drone cluster. Among them, the formation time mutual interference probability density is inversely proportional to the collaborative formation capability of the drone cluster, that is, the greater the formation time mutual interference probability density, the lower the collaborative formation capability; the smaller the formation time mutual interference probability density, the higher the collaborative formation capability. For example, assuming that the target duration is T (T is a positive integer), the number of colliding drones at the tth moment within the target duration is C. t , the number of drones in the drone cluster is N, then the probability density of formation time mutual interference of the drone cluster within the target duration is P FS The calculation formula is as follows.

[0154]

[0155] Step 6.3: Based on the projected area of ​​the airspace occupied by the drone cluster, the number of colliding drones, and the total number of drones in the drone cluster, calculate the probability density of mutual interference in the formation space; among them, the probability density of mutual interference in the formation space is inversely proportional to the coordinated formation capability of the drone cluster.

[0156] The computer device sums the probability of collision of drones at each moment within the target duration, and calculates the ratio of the summation result to the projected area of ​​the airspace occupied by the drone cluster, to obtain the probability density of mutual interference in the formation space of the drone cluster. Among them, the probability density of mutual interference in the formation space is inversely proportional to the collaborative formation capability of the drone cluster, that is, the smaller the probability density of mutual interference in the formation space, the higher the collaborative formation capability; the larger the probability density of mutual interference in the formation space, the lower the collaborative formation capability. For example, assuming that the target duration is T (T is a positive integer), the number of colliding drones at the tth moment within the target duration is C t , the number of drones in the drone cluster is N, and the projected area of ​​the airspace occupied by the drone cluster is R F , then the probability density P of the formation space mutual interference of the UAV cluster within the target duration FK The calculation formula is as follows.

[0157]

[0158] It should be noted that in the present invention, when the computer device evaluates the cooperative formation capability of the drone cluster from one or more evaluation indicators such as formation time consumption, formation holding time, formation error, formation support number, formation density, formation time mutual interference probability density, formation space mutual interference probability density, etc., it can independently evaluate the cooperative formation capability with each evaluation indicator as shown in the above embodiment, or it can jointly evaluate the cooperative formation capability with multiple evaluation indicators, and the present invention does not limit this. Exemplarily, on the one hand, the computer device calculates the formation capability evaluation value of the drone cluster based on the evaluation values ​​of the four evaluation indicators of formation time consumption, formation holding time, formation error and formation support number, and on the other hand, calculates the cooperative capability evaluation value of the drone cluster based on the evaluation values ​​of the three evaluation indicators of formation density, formation time mutual interference probability density and formation space mutual interference probability density, and then combines the formation capability evaluation value and the cooperative capability evaluation value to evaluate the cooperative formation capability of the drone cluster.

[0159] In summary, the present invention constructs a formation coordinate system with the centroid of the drone cluster as the origin. When evaluating the collaborative formation capability of the drone cluster, based on the coordinates of each drone in the formation coordinate system, on the one hand, it can avoid the situation where the test evaluation cannot be carried out normally due to the failure of the drone as the origin of the coordinate system; on the other hand, it can dynamically adjust the formation information based on the change in the number of drones in the drone cluster, effectively avoiding the situation where the test evaluation fails due to the damage or disconnection of a drone. In addition, the formation coordinate system constructed by the present invention is a three-dimensional coordinate system, which can describe the position of the drone without all drones in the drone cluster flying to the same two-dimensional plane, breaking the spatial limitation of the drone position description, and helping to realize the continuous and real-time evaluation of the collaborative formation capability during the flight of the drone cluster, which is more in line with the evaluation of the collaborative formation capability under actual conditions. The present invention simplifies the parameters required for the evaluation of the collaborative formation capability through the use of a three-dimensional formation coordinate system, and the evaluation speed is stabilized at 1000FPS, which is sufficient to meet the continuous and real-time evaluation requirements.

[0160] In addition, during the actual test process, there may be a situation where the formation of the drone cluster is consistent with the expected formation in terms of shape and size, but the numbers of the drones in two positions are opposite to the numbers of the reference drones. Usually, in this case, the drone cluster is also considered to have formed the expected formation. If the correspondence between the drones in the drone cluster and the reference drones in the expected formation is obtained by numbering when comparing the formation of the drone cluster and the expected formation, then in the above case, it will cause the computer device to believe that the drone cluster has not formed the expected formation, thereby affecting the accuracy of the evaluation. In the present invention, the computer device calculates the formation position quantity using the coordinate values ​​of the formation coordinates in all dimensions, and obtains the correspondence between the drones in the drone cluster and the reference drones in the expected formation through the formation position quantity, thereby improving the accuracy of the collaborative formation capability evaluation.

[0161] refer to Figure 6 , which shows a block diagram of a cooperative formation capability evaluation device for a drone cluster provided by an embodiment of the present invention. The device may be a computer device, or may be set in a computer device. Figure 6 As shown, the device 600 includes: a task acquisition module 610, a task execution module 620, a coordinate system construction module 630 and a capability evaluation module 640.

[0162] The task acquisition module 610 is used to acquire a test task for a drone cluster, where the drone cluster includes at least two drones, and the test task includes at least one expected formation.

[0163] The task running module 620 is used to run the test task; wherein, during the running of the test task, the drone cluster flies in turn according to the at least one expected formation.

[0164] The coordinate system construction module 630 is used to obtain the centroid of the drone cluster during the flight of the drone cluster, and to construct a formation coordinate system with the centroid as the origin; wherein the formation coordinate system is a three-dimensional coordinate system, and the Z axis of the formation coordinate system is parallel to the gravity direction of the drone.

[0165] The capability evaluation module 640 is used to evaluate the collaborative formation capability of the drone cluster based on the at least one expected formation formation and the formation coordinates of each of the drones, where the formation coordinates of the drones refer to the coordinates of the drones in the formation coordinate system.

[0166] Optionally, the coordinate system construction module 630 is used to: for each drone in the drone cluster, use the coordinates of the center of the drone in the world coordinate system as the actual coordinates of the drone; obtain the actual coordinate values ​​of the same dimension in the actual coordinates of all drones in the drone cluster, and calculate the average of the actual coordinate values ​​of the corresponding dimension; use the average of the actual coordinate values ​​of each dimension as the actual coordinate value of the centroid in the corresponding dimension, and combine the actual coordinate values ​​of the centroid in each dimension into the actual coordinates of the centroid.

[0167] Optionally, the device 600 further includes a formation coordinate calculation module ( Figure 6 (not shown) is used for: calculating the difference between the actual coordinate values ​​of the same dimension in the actual coordinates of the target UAV and the centroid for the target UAV in the UAV cluster; using the difference between the actual coordinate values ​​of each dimension as the coordinate value of the corresponding dimension of the target UAV in the formation coordinate system, and combining the coordinate values ​​of each dimension of the target UAV in the formation coordinate system into the formation coordinates of the target UAV.

[0168] Optionally, the capability evaluation module 640 is used to: when the drone cluster flies according to each expected formation formation in the at least one expected formation formation, respectively obtain the formation coordinates of the reference drones in the corresponding expected formation formation; perform weighted summation processing on the coordinate values ​​of all dimensions in the formation coordinates of each drone to obtain the formation position quantity of the corresponding drone, and sort the formation position quantities of all drones in the drone cluster according to the target order to obtain the drone formation position order; perform weighted summation processing on the coordinate values ​​of all dimensions in the formation coordinates of each reference drone to obtain the formation position quantity of the corresponding reference drone, and sort the formation position quantities of all reference drones in the corresponding expected formation formation according to the target order to obtain the reference drone formation position order; correspond each ranked drone in the drone formation position order to a reference drone with the same ranking in the reference drone formation position order; and evaluate the collaborative formation capability of the drone cluster based on the formation coordinates of each of the drones and the formation coordinates of the reference drones corresponding to the drones.

[0169] Optionally, the capability evaluation module 640 is used to: calculate the coordinate error value between the formation coordinates of each of the drones and the formation coordinates of the reference drone corresponding to the drone when the drone cluster flies in accordance with the target expected formation in the at least one expected formation; if the coordinate error values ​​corresponding to all drones in the drone cluster are less than the error value threshold, start the formation stabilization time timing; if the timing duration of the formation stabilization time is greater than a preset duration, determine that the drone cluster forms the target expected formation; based on the moment when the drone cluster forms the target expected formation, determine the formation stabilization time when the drone cluster forms the target expected formation. At the starting moment of the expected formation flight, the formation time taken by the UAV cluster to form the target expected formation is calculated; wherein the formation time is inversely proportional to the coordinated formation capability of the UAV cluster; the formation holding time timing is started; during the formation holding time timing, if the coordinate error value corresponding to the UAV in the UAV cluster is greater than or equal to the error value threshold, the formation holding time timing is ended; the timing duration of the formation holding time is used as the formation holding time for the UAV cluster to form the target expected formation; wherein the formation holding time is directly proportional to the coordinated formation capability of the UAV cluster.

[0170] Optionally, the device 600 further includes a fault monitoring module ( Figure 6), which is used to: determine whether there is a malfunctioning drone in the drone cluster; if there is a malfunctioning drone, delete the reference drone with the same number as the malfunctioning drone from the corresponding expected formation formation, and update the corresponding expected formation formation and the formation coordinates of each reference drone in the corresponding expected formation formation based on the remaining reference drones in the corresponding expected formation formation; update the centroid of the drone cluster and the formation coordinate system based on the remaining drones in the drone cluster except the malfunctioning drone, and update the formation coordinates of the remaining drones according to the updated centroid and formation coordinate system.

[0171] Optionally, the capability evaluation module 640 is used to: select a reference drone from the drone cluster based on the formation coordinates of each of the drones; calculate the polar angles between the remaining drones in the drone cluster except the reference drone and the reference drone; sort the remaining drones except the reference drone in the order of the polar angle values ​​from small to large to obtain a polar angle position sequence of the drones; starting from the reference drone, recursively traverse the three adjacent drones in the polar angle position sequence of the drones counterclockwise, and calculate the rotation direction of two vectors composed of the middle drone of the three adjacent drones as the origin; if the rotation direction is less than or equal to 180 degrees, the middle drone is taken as the first drone in the drone cluster. The vertices of the outer polygon of the projection of the airspace occupied by the group; if the rotation direction is greater than 180 degrees, the middle drone is deleted from the polar angular position sequence of the drones, and the three adjacent drones in the polar angular position sequence of the drones are continued to be recursively traversed counterclockwise; until all drones in the polar angular position sequence of the drones are traversed, all vertices of the outer polygon of the projection of the airspace occupied by the drone cluster are obtained; based on the plane coordinates of the drones corresponding to all vertices, the projection area of ​​the airspace occupied by the drone cluster is calculated; the ratio of the total number of drones in the drone cluster to the projection area of ​​the airspace occupied by the drone cluster is used as the formation density of the drone cluster; wherein the formation density is inversely proportional to the collaborative formation capability of the drone cluster.

[0172] Optionally, the capability assessment module 640 is used to: obtain the number of colliding drones in the drone cluster within a target duration; calculate the formation time mutual interference probability density based on the target duration, the number of colliding drones and the total number of drones in the drone cluster; wherein the formation time mutual interference probability density is inversely proportional to the collaborative formation capability of the drone cluster; calculate the formation space mutual interference probability density based on the projected area of ​​the airspace occupied by the drone cluster, the number of colliding drones and the total number of drones in the drone cluster; wherein the formation space mutual interference probability density is inversely proportional to the collaborative formation capability of the drone cluster.

[0173] For the detailed description of the implementation process, beneficial effects, etc. of the device module, please refer to the introduction of the above method embodiment, which will not be elaborated here.

[0174] In an exemplary embodiment, a computer device is also provided, which includes a memory and a processor, wherein a computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the above-mentioned method for evaluating the collaborative formation capability of a drone cluster.

[0175] In an exemplary embodiment, a computer-readable storage medium is also provided, on which a computer program is stored. When the computer program is executed by a processor, the method for evaluating the collaborative formation capability of a drone cluster as described above is implemented.

[0176] In an exemplary embodiment, a computer program product is also provided. When the computer program product is executed on a computer device, the computer device executes the collaborative formation capability assessment method of the drone cluster as described above.

[0177] The above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit the same. A person skilled in the art may modify or make equivalent substitutions for the technical solutions of the present invention without departing from the spirit and scope of the present invention. The protection scope of the present invention shall be subject to the claims.

Claims

1. A method for evaluating the collaborative formation capability of a drone swarm, characterized in that: The method comprises: Acquire a test task for a drone cluster, the drone cluster including at least two drones, and the test task including at least one expected formation; Running the test task; wherein, during the running of the test task, the drone cluster flies in sequence according to the at least one expected formation; During the flight of the drone cluster, the centroid of the drone cluster is obtained, and a formation coordinate system is constructed with the centroid as the origin; wherein the formation coordinate system is a three-dimensional coordinate system, and the Z axis of the formation coordinate system is parallel to the gravity direction of the drone. During the flight of the drone cluster, the centroid of the drone cluster is obtained, including: For each drone in the drone cluster, the coordinates of the center of the drone in the world coordinate system are used as the actual coordinates of the drone; Obtaining the actual coordinate values ​​of the same dimension in the actual coordinates of all drones in the drone cluster, and calculating the average of the actual coordinate values ​​of the corresponding dimension; The average of the actual coordinate values ​​of each dimension is used as the actual coordinate value of the centroid in the corresponding dimension, and the actual coordinate values ​​of the centroid in each dimension are combined into the actual coordinates of the centroid; Based on the at least one expected formation and the formation coordinates of each of the drones, evaluating the coordinated formation capability of the drone cluster, where the formation coordinates of the drones refer to the coordinates of the drones in the formation coordinate system; wherein, based on the at least one expected formation and the formation coordinates of each of the drones, evaluating the coordinated formation capability of the drone cluster includes: When the drone cluster flies according to each of the at least one expected formation formations, respectively obtaining the formation coordinates of the reference drones in the corresponding expected formation formations; Performing weighted summation processing on the coordinate values ​​of all dimensions in the formation coordinates of each drone to obtain the formation position quantity of the corresponding drone, and sorting the formation position quantities of all drones in the drone cluster according to the target order to obtain the drone formation position order; Performing weighted summation processing on the coordinate values ​​of all dimensions in the formation coordinates of each reference UAV to obtain the formation position quantity of the corresponding reference UAV, and sorting the formation position quantities of all reference UAVs in the corresponding expected formation formation according to the target order to obtain the reference UAV formation position order; Matching each drone with a ranking in the drone formation position sequence with a reference drone with the same ranking in the reference drone formation position sequence; Based on the formation coordinates of each of the drones and the formation coordinates of a reference drone corresponding to the drone, the coordinated formation capability of the drone cluster is evaluated; wherein, based on the formation coordinates of each of the drones and the formation coordinates of a reference drone corresponding to the drone, the coordinated formation capability of the drone cluster is evaluated, including: When the drone cluster flies according to the target expected formation formation in the at least one expected formation formation, calculating the coordinate error value between the formation coordinates of each of the drones and the formation coordinates of the reference drone corresponding to the drone; If the coordinate error values ​​corresponding to all drones in the drone cluster are less than the error value threshold, the formation stabilization time is started; If the timing duration of the formation stabilization time is greater than the preset duration, determining that the drone cluster forms the target expected formation; Based on the moment when the drone cluster forms the target expected formation and the start moment when the drone cluster flies according to the target expected formation, the formation time taken by the drone cluster to form the target expected formation is calculated; wherein the formation time is inversely proportional to the coordinated formation capability of the drone cluster; Start the formation holding time timing; during the formation holding time timing process, if the coordinate error value corresponding to the drone in the drone cluster is greater than or equal to the error value threshold, end the formation holding time timing; use the timing duration of the formation holding time as the formation holding time for the drone cluster to form the target expected formation formation; wherein the formation holding time is positively proportional to the collaborative formation capability of the drone cluster.

2. The method according to claim 1, characterized in that After the formation coordinate system is constructed with the centroid as the origin, the method further includes: For a target UAV in the UAV cluster, calculating a difference between the actual coordinate values ​​of the target UAV and the actual coordinates of the centroid in the same dimension; The actual coordinate value difference of each dimension is used as the coordinate value of the corresponding dimension of the target UAV in the formation coordinate system, and the coordinate values ​​of each dimension of the target UAV in the formation coordinate system are combined into the formation coordinates of the target UAV.

3. The method according to claim 1, characterized in that When the drone cluster flies according to each of the at least one expected formation formations, before respectively acquiring the formation coordinates of the reference drones in the corresponding expected formation formations, the method further includes: Determine whether any drone in the drone cluster has a malfunction; If a UAV fails, the reference UAV with the same number as the failed UAV is deleted from the corresponding expected formation, and the corresponding expected formation and the formation coordinates of each reference UAV in the corresponding expected formation are updated based on the remaining reference UAVs in the corresponding expected formation; Based on the remaining drones in the drone cluster except the malfunctioning drone, the centroid of the drone cluster and the formation coordinate system are updated, and the formation coordinates of the remaining drones are updated according to the updated centroid and formation coordinate system.

4. The method according to any one of claims 1 to 3, characterized in that: After obtaining the centroid of the drone cluster and constructing a formation coordinate system with the centroid as the origin, the method further includes: Selecting a reference drone from the drone cluster based on the formation coordinates of each of the drones; Calculating polar angles between the remaining drones in the drone cluster, except the reference drone, and the reference drone; According to the value of the polar angle, the remaining drones except the reference drone are sorted in order from small to large, so as to obtain the polar angle position order of the drones; Starting from the reference UAV, recursively traverse three adjacent UAVs in the polar angle position sequence of the UAVs counterclockwise, and calculate the rotation directions of two vectors formed by the middle UAV of the three adjacent UAVs as the origin; If the rotation direction is less than or equal to 180 degrees, the middle UAV is used as the vertex of the outer polygon of the projection of the airspace occupied by the UAV cluster; If the rotation direction is greater than 180 degrees, the middle drone is deleted from the drone polar angle position sequence, and the three adjacent drones in the drone polar angle position sequence are continuously recursively traversed counterclockwise; Until all the drones in the polar angle position sequence of the drones are traversed, all vertices of the outer polygon of the projection of the airspace occupied by the drone cluster are obtained; Based on the plane coordinates of the drones corresponding to all vertices, calculate the airspace projection area occupied by the drone cluster; The ratio of the total number of drones in the drone cluster to the projected area of ​​the airspace occupied by the drone cluster is used as the formation density of the drone cluster; wherein the formation density is inversely proportional to the collaborative formation capability of the drone cluster.

5. The method according to any one of claims 1 to 3, characterized in that: After obtaining the centroid of the drone cluster and constructing a formation coordinate system with the centroid as the origin, the method further includes: Obtain the number of collision drones in the drone cluster within the target duration; Based on the target duration, the number of the colliding drones and the total number of drones in the drone cluster, the formation time mutual interference probability density is calculated; wherein the formation time mutual interference probability density is inversely proportional to the cooperative formation capability of the drone cluster; Based on the projected area of ​​the airspace occupied by the drone cluster, the number of the colliding drones and the total number of drones in the drone cluster, the probability density of mutual interference in the formation space is calculated; wherein the probability density of mutual interference in the formation space is inversely proportional to the collaborative formation capability of the drone cluster.

6. A device for evaluating the collaborative formation capability of a drone swarm, characterized in that: The device comprises: A task acquisition module, used to acquire a test task for a drone cluster, wherein the drone cluster includes at least two drones, and the test task includes at least one expected formation; A task running module, used to run the test task; wherein, during the running of the test task, the drone cluster flies in sequence according to the at least one expected formation; A coordinate system construction module is used to obtain the centroid of the drone cluster during the flight of the drone cluster, and to construct a formation coordinate system with the centroid as the origin; wherein the formation coordinate system is a three-dimensional coordinate system, and the Z axis of the formation coordinate system is parallel to the gravity direction of the drone. The centroid of the drone cluster is obtained during the flight of the drone cluster, including: For each drone in the drone cluster, the coordinates of the center of the drone in the world coordinate system are used as the actual coordinates of the drone; Obtaining the actual coordinate values ​​of the same dimension in the actual coordinates of all drones in the drone cluster, and calculating the average of the actual coordinate values ​​of the corresponding dimension; The average of the actual coordinate values ​​of each dimension is used as the actual coordinate value of the centroid in the corresponding dimension, and the actual coordinate values ​​of the centroid in each dimension are combined into the actual coordinates of the centroid; A capability evaluation module is used to evaluate the coordinated formation capability of the drone cluster based on the at least one expected formation formation and the formation coordinates of each of the drones, where the formation coordinates of the drones refer to the coordinates of the drones in the formation coordinate system; wherein the evaluating the coordinated formation capability of the drone cluster based on the at least one expected formation formation and the formation coordinates of each of the drones includes: When the drone cluster flies according to each of the at least one expected formation formations, respectively obtaining the formation coordinates of the reference drones in the corresponding expected formation formations; Performing weighted summation processing on the coordinate values ​​of all dimensions in the formation coordinates of each drone to obtain the formation position quantity of the corresponding drone, and sorting the formation position quantities of all drones in the drone cluster according to the target order to obtain the drone formation position order; Performing weighted summation processing on the coordinate values ​​of all dimensions in the formation coordinates of each reference UAV to obtain the formation position quantity of the corresponding reference UAV, and sorting the formation position quantities of all reference UAVs in the corresponding expected formation formation according to the target order to obtain the reference UAV formation position order; Matching each drone with a ranking in the drone formation position sequence with a reference drone with the same ranking in the reference drone formation position sequence; Based on the formation coordinates of each of the drones and the formation coordinates of a reference drone corresponding to the drone, the coordinated formation capability of the drone cluster is evaluated; wherein, based on the formation coordinates of each of the drones and the formation coordinates of a reference drone corresponding to the drone, the coordinated formation capability of the drone cluster is evaluated, including: When the drone cluster flies according to the target expected formation formation in the at least one expected formation formation, calculating the coordinate error value between the formation coordinates of each of the drones and the formation coordinates of the reference drone corresponding to the drone; If the coordinate error values ​​corresponding to all drones in the drone cluster are less than the error value threshold, the formation stabilization time is started; If the timing duration of the formation stabilization time is greater than the preset duration, determining that the drone cluster forms the target expected formation; Based on the moment when the drone cluster forms the target expected formation and the start moment when the drone cluster flies according to the target expected formation, the formation time taken by the drone cluster to form the target expected formation is calculated; wherein the formation time is inversely proportional to the coordinated formation capability of the drone cluster; Start the formation holding time timing; during the formation holding time timing process, if the coordinate error value corresponding to the drone in the drone cluster is greater than or equal to the error value threshold, end the formation holding time timing; use the timing duration of the formation holding time as the formation holding time for the drone cluster to form the target expected formation formation; wherein the formation holding time is positively proportional to the collaborative formation capability of the drone cluster.

7. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the method according to any one of claims 1 to 5.

Images