A distributed UAV swarm formation and collaborative control method

Abstract

The present application belongs to the technical field of drone cluster control, and specifically relates to a distributed drone cluster formation and collaborative control method. The method includes step S1, discretizing the horizontal space of the drone cluster formation into multiple grids corresponding to the formation configuration matrix, and setting the element value of the formation configuration matrix corresponding to the grid to a non-zero value according to the requirements of the flying drone; step S2, sending a formation flight instruction carrying a formation configuration number to each drone, and the formation configuration number is associated with the corresponding formation configuration matrix; step S3, each drone determines the position of the drone according to the formation configuration matrix; step S4, each drone makes an autonomous decision to fly according to the target position and set speed of the drone cluster formation. The present application solves the problem of drone cluster dependence on the computing and communication capabilities of the central planning node, so that the drone cluster has good robustness, flexibility and scalability.

Description

Technical Field

[0001] The present application belongs to the field of UAV cluster control technology, and in particular relates to a distributed UAV cluster formation and collaborative control method. Background Art

[0002] The collaborative execution of tasks by drone swarms can break through the limitations of a single drone in terms of perception, decision-making and behavioral capabilities, have stronger survivability and a wider range of mission execution, and can complete planned flight missions with higher quality, higher probability and shorter time. It is an inevitable trend in the current development of drone technology.

[0003] UAV swarm formation flight is one of the basic tasks of UAV swarms. In the formation, UAVs form corresponding formations according to certain rules, reconstruct, maintain and switch the formation according to the actual flight situation, and complete the flight mission in a coordinated manner. Among them, the efficient, flexible and autonomous transformation of formations of different configurations by UAV swarms according to the needs of the mission determines whether the swarm collaborative mission can be effectively completed. Whether it is military fields such as UAV collaborative reconnaissance, electronic countermeasures, communication relay, and collaborative attack, or civilian fields such as flight performances and ground observation, they all rely on the efficient, flexible and precise formation flight capabilities of UAV swarms.

[0004] In the prior art, the UAV swarm formation task setting and control methods can be mainly divided into three categories:

[0005] (1) The ground station or a certain UAV is used as a centralized planning node. The formation task allocation and trajectory planning are all performed on this centralized planning node, forming a detailed flight waypoint or instruction sequence for each UAV, which is sent to other UAVs in turn. Other UAVs only fly along the waypoints or instructions provided by the planning node.

[0006] For example, in the Chinese invention patent with publication number CN109445459A, a method and device for controlling the formation of drone clusters is disclosed. According to the position of the formation reference point and the position of the target reference point, the velocity vector, the acceleration vector, the kinematic equation and the motion type, the acceleration control function of each internal member of the cluster and the velocity control function of each internal member at each formation reference point are used to control each internal member to follow the formation reference point toward the target reference point in the default formation. This type of method strictly relies on the reliability of the centralized planning node and the communication network, and the communication volume is extremely large. If the planning node fails or other drones cannot receive the planning results, the cluster cannot fly in formation.

[0007] (2) Before the UAVs perform formation flight missions, the flight formation has been described in advance in the onboard planning and control system through track points or geometric expressions, including the spacing and orientation between UAVs. This type of formation flight can only fly according to the preset formation, or change between a limited number of formations by sending formation change commands. It lacks flexibility and cannot be applied in practice. At the same time, this method can only describe simple formations such as herringbone and straight line. Users who do not have relevant knowledge find it difficult to describe heterogeneous UAV cluster formations with complex geometric configurations for real mission objectives.

[0008] (3) All drones are considered as homogeneous intelligent entities, each of which is an autonomous decision-making node. Simple rules are defined for the drone cluster, and each drone self-organizes to form a formation based on the rules.

[0009] For example, in the Chinese invention patent with publication number CN108983818A, a method for UAV formation transformation based on virtual structure is disclosed, which translates the formation transformation instruction into the formation transformation control instruction represented by the virtual particle coordinate parameters, and transforms the formation control problem into the problem of UAV tracking virtual particles, which effectively reduces the dependence of cluster UAVs on the central planning node and communication network; however, the motion parameters of the virtual node are still planned on the ground control base station. As the number of UAVs increases and the scale of the cluster system increases, the cluster management problem will become extremely complex, and the computing resources of the central planning unit will be difficult to meet the real-time requirements of UAV cluster applications. Such methods cannot be applied to heterogeneous UAV clusters with different flight capabilities and mission payloads, and formations cannot be designed according to specific cluster mission objectives.

[0010] Therefore, how to provide UAV swarm commanders with an efficient and convenient way to set up formation tasks, so that users without relevant knowledge can also efficiently design formation configurations according to the specific needs of UAV swarm usage, and enable UAV swarms to rely on a small amount of inter-machine communication and highly autonomous collaborative control to achieve the user's expected formation configuration, is a key issue that needs to be urgently solved in UAV swarm formation flight. Summary of the invention

[0011] In order to solve at least one of the above technical problems, the present application provides a distributed UAV swarm formation and collaborative control method, including:

[0012] Step S1, discretize the horizontal space of the drone cluster formation into multiple grids corresponding to the formation configuration matrix. If a certain grid requires a drone to fly, the element value of the formation configuration matrix corresponding to the grid is set to a non-zero value according to the requirement for the drone to fly, otherwise the corresponding element value of the formation configuration matrix is ​​zero;

[0013] Step S2: sending a formation flight instruction carrying a formation configuration number to each UAV, wherein the formation configuration number is associated with a corresponding formation configuration matrix;

[0014] Step S3, each UAV determines its position in the formation according to the formation configuration matrix;

[0015] Step S4, each UAV makes an autonomous decision on flight according to the target position and set speed of the UAV cluster formation, wherein the autonomous decision-making flight includes, when the UAV cluster formation flies toward the target position, each UAV adjusts the heading angle according to its own position and the target position so as to keep the formation formation unchanged; and each UAV autonomously calculates the flight speed control instruction according to the flight set speed of the UAV cluster formation and the flight speed of other UAVs in the formation so as to make the expected arrival time of each UAV at the target position tend to be consistent.

[0016] Preferably, in step S1, the formation horizontal space is discretized into a plurality of grids based on the greatest common divisor of the expected spacings between the UAVs in the formation.

[0017] Preferably, step S2 further comprises:

[0018] Multiple formation flight instructions are sent to each drone in sequence. When the drone cluster formation reaches the specified target task according to the formation configuration matrix given by the previous formation flight instruction, each drone transitions to the spatial node position specified by the formation configuration matrix given by the next formation flight instruction.

[0019] Preferably, before step S3, the method further includes uploading a plurality of preset formation configuration matrices to the drones for pre-storage, and each drone retrieves the corresponding formation configuration matrix according to the formation configuration number.

[0020] Preferably, step S3 further comprises:

[0021] Step S31, extracting a first set consisting of non-zero value elements in the formation configuration matrix;

[0022] Step S32, traverse the elements in the first set that specify the aircraft type, load type and ID in turn, if there is an element whose aircraft type, load type and ID are consistent with the aircraft itself, the aircraft will fly the formation node corresponding to the element, and determine the position of the aircraft in the formation according to the position of the element in the formation configuration matrix, otherwise execute step S33;

[0023] Step S33, determine a second set of drones of the same aircraft type and load type that can communicate with the local aircraft, remove the drones specified to fly each node in step S32, form a third set, and traverse the elements of the first set with the specified aircraft type and load type in turn. If there is an element whose aircraft type and load type are consistent with the local aircraft, select a drone from the third set according to the agreed rules to fly the formation node corresponding to the element. If the local aircraft is selected to fly the node, determine the position of the local aircraft in the formation according to the position of the element in the formation configuration matrix. Otherwise, execute step S34;

[0024] Step S34, remove the drones designated to fly each node in step S33 from the third set to form a fourth set, traverse the remaining elements in the first set in turn, and select a drone from the fourth set to fly the formation node corresponding to the element according to the agreed rules. If the aircraft is selected to fly the node, the position of the aircraft is determined according to the position of the element in the formation configuration matrix, otherwise enter a waiting state.

[0025] Preferably, the agreed rules include drone IDs from small to large, or drone flight fees from low to high.

[0026] Preferably, in step S4, adjusting the heading angle includes:

[0027] Step S41, determining a flight baseline based on the starting position and target position carried by the formation flight instruction;

[0028] Step S42: When the UAV deviates from the flight baseline, a virtual gravitational force is formed; when the distance between the UAV and an obstacle or other UAVs in the formation is less than a threshold, a virtual repulsive force is formed; and the heading angle of the UAV is corrected based on the resultant force of the virtual gravitational force and the virtual repulsive force.

[0029] Preferably, in step S42, the correction value Δψ of the heading angle includes:

[0030] Δψ=sgn(F n )·arctan(||F||);

[0031] Among them, F n is the normal component of the virtual repulsive force perpendicular to the motion of the drone, and F is the resultant force of the virtual attraction and the virtual repulsive force.

[0032] Preferably, step S42 further comprises:

[0033] When the distance between the UAV and other UAVs in the formation is less than a threshold, the flight altitude of the UAV is adjusted to further displace the two UAVs approaching each other in altitude.

[0034] Preferably, in step S4, autonomously calculating the flight speed control strategy includes:

[0035] Step S43, each UAV calculates the first time expectation of reaching the target position of the UAV;

[0036] Step S44, determining a second expected time for other drones communicating with the drone to reach the target location;

[0037] Step S45: using a first-order model to describe the speed of the UAV so that the first time expectation approaches the second time expectation, and solving the corresponding flight speed control instruction.

[0038] Through distributed task allocation and collaborative control, this application can autonomously fly in a formation configuration without the participation of ground measurement and control stations or any other central planning nodes; it solves the problem of drone clusters' dependence on the computing and communication capabilities of central planning nodes, and enables drone clusters to obtain good robustness, flexibility and scalability. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a flow chart of an implementation method of the distributed UAV cluster formation and collaborative control method of the present application.

[0040] Figure 2 This is a schematic diagram of a simple irregular UAV formation configuration according to a preferred embodiment of the present application.

[0041] Figure 3 This application Figure 2 Schematic diagram of the formation configuration matrix generated by the drone formation configuration of the illustrated embodiment.

[0042] Figure 4 This is a schematic diagram of a heterogeneous distributed UAV swarm formation mission configuration according to a preferred embodiment of the present application.

[0043] Figure 5 This application Figure 4 Schematic diagram of the formation configuration matrix generated by the drone formation configuration of the illustrated embodiment.

[0044] Figure 6 This is a cluster collaborative formation flight control flow chart of a preferred embodiment of the present application.

[0045] Figure 7 It is a schematic diagram of the formation flying baseline of a preferred embodiment of the present application.

[0046] Figure 8 This is a schematic diagram of the virtual force analysis on the UAV.

[0047] Fig. 9 This is a schematic diagram of generating UAV heading instructions. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical scheme and advantages of the implementation of this application clearer, the technical scheme in the implementation of this application will be described in more detail in combination with the drawings in the implementation of this application. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described implementation is a part of the implementation of this application, not all of the implementations. The implementation described below with reference to the drawings is exemplary and is intended to be used to explain this application, and cannot be understood as a limitation on this application. Based on the implementation in this application, all other implementations obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. The implementation of this application is described in detail below in combination with the drawings.

[0049] The distributed drone clusters targeted by this application include:

[0050] (1) Several heterogeneous drones

[0051] The UAV system is mainly composed of an aircraft platform, a cluster management system, a flight control system, a communication system that supports data exchange between aircraft, mission payloads, ground control equipment, support equipment, and optional equipment. Its heterogeneity is reflected in the flight capabilities of the aircraft platform and the different mission payloads it carries. There is no master-slave distinction between UAVs, and they are on an equal footing in the cluster system. There is no central node responsible for cluster planning; aircraft can exchange flight missions and status.

[0052] (2) Cluster ground tracking and control station

[0053] Because the coordinated planning of formation flight missions for the distributed UAV clusters targeted by this application is completed on the aircraft through distributed collaboration, the cluster ground measurement and control station focuses on quickly designing the formation flight configuration, flight speed, recovery point and other mission parameters according to the mission before the cluster mission is executed or during the flight, monitoring the cluster flight status in real time, processing the returned payload information in real time, and intervening in the control of the cluster or single aircraft when necessary.

[0054] like Figure 1 As shown, the present application provides a distributed UAV cluster formation and collaborative control method, which mainly includes:

[0055] Step S1, discretize the horizontal space of the drone cluster formation into multiple grids corresponding to the formation configuration matrix. If a certain grid requires a drone to fly, the element value of the formation configuration matrix corresponding to the grid is set to a non-zero value according to the requirement for the flying drone, otherwise the corresponding element value of the formation configuration matrix is ​​zero.

[0056] In this step, the UAV cluster formation configuration space is discretized into several grids, and each node in the grid corresponds to an element of the discrete matrix to form a discrete matrix; specifically, Figure 2 As shown, in an optional embodiment, the formation horizontal space is discretized into M×N grids based on the greatest common divisor d of the expected spacing between the UAVs in the formation, corresponding to the formation configuration space discrete matrix: C = [C ij ] M×N , where if the grid requires a drone to fly, C ij Set to a non-zero value; otherwise C ij = 0. Assuming that all aircraft are of the same type of UAVs, and there are no special requirements for aircraft at each position; if the grid is a formation node, C ij =99; otherwise C ij = 0. The formation discrete matrix can be written as Figure 3 As shown, usually the element value refers to the ID of the drone. If it is 99, it means that any drone can be selected to fill in.

[0057] The user can configure the elements C of this matrix ij , set whether there is a formation drone in the grid space, and specify which type or which specific number of drones will fly the formation node.

[0058] Usually, according to the mission requirements, the aircraft type, payload capacity, aircraft number, etc. are encoded as matrix elements to identify more complex formation scenarios. In a heterogeneous distributed UAV cluster, certain positions in the formation usually need to be operated by UAVs of specific types and payloads. In order to identify more information in the formation discrete matrix, this application uniformly encodes the aircraft type, payload capacity, and aircraft number as matrix elements to identify the formation mission information. A typical heterogeneous distributed UAV cluster formation mission is as follows: Figure 4 shown.

[0059] In a specific embodiment, the drone type FlightType={FWB, FWS, RTH, RTM}, the payload type PayloadType={ECM, CMR, ATA, ATG}, the aircraft number UAVID={1,2,…,99} are defined, and the meanings and values ​​of the set elements are shown in Table 1.

[0060] Table 1 Definition and value of heterogeneous UAV capability coding elements

[0061]

[0062]

[0063] The encoding method used in this embodiment is: ij=FlighType×100+PayloadType×10+UAVID, the drone type, payload type, and aircraft number are uniformly identified as a 4-digit integer. In particular, when a grid only has requirements for the aircraft type and payload type, but does not specify a certain ID aircraft to fly to the location, UAVID=0. For example, 2101 represents a drone with a small fixed-wing aircraft type, an electronic countermeasure payload type, and a number of 01. Another example is 3200, which represents a drone with an aircraft type of unmanned helicopter and a payload type of communication relay, and any ID. Another example is 21, which represents a drone with an ID of 21 regardless of the aircraft type and payload type. Another example is 99, which represents any drone.

[0064] In a specific embodiment, the formed Figure 4 The formation configuration matrix related to the UAV cluster formation represented by Figure 5 It is understandable that the present application can also encode the drones by other encoding methods, as long as it can accurately refer to each drone in the formation, for example, the drone type, load type, aircraft number decimal weighted method is used to represent the flight capability of the drone, and as an element of the formation configuration space discrete matrix, in this embodiment, the encoded drone capability dimensions may also include remaining flight time, fault status, partial load quantity (such as the number of attack weapons), onboard computing capacity, etc. In addition, the encoding method can also be bit-coded, etc.

[0065] The above encoding method is only an embodiment of the present application, and the embodiment is only used to explain the present application, rather than to limit the present application. Other encoding methods such as bit-by-bit encoding are within the protection scope of the present application.

[0066] Step S2: Sending a formation flight instruction carrying a formation configuration number to each UAV, wherein the formation configuration number is associated with a corresponding formation configuration matrix.

[0067] In this step, the formation flight instructions are shown in Table 2. The formation flight instructions not only carry the formation configuration (number), but also include the latitude and longitude, altitude of the target mission, and the spacing between each UAV (element spacing).

[0068] Table 2 Formation flight instruction format

[0069]

[0070]

[0071] In some optional embodiments, step S2 further includes: sending multiple formation flight instructions to each drone in sequence, and when the drone cluster formation reaches the specified target task according to the formation configuration matrix given by the previous formation flight instruction, each drone transitions to the spatial node position specified by the formation configuration matrix given by the next formation flight instruction.

[0072] After the ground commander designs the UAV cluster formation and generates the formation configuration space discrete matrix, he sets the formation flight mission to the UAV cluster through the following steps: 1) Upload the designed K formation configuration matrices: C1, C2, ..., C K To all drones, and the drones are saved to the onboard memory after entering the matrix; 2) The complete flight mission of the drone cluster is decomposed into a series of formation flight mission sequences flying to a certain target location in a certain formation configuration, and the setting of all formation flight missions is completed by sending a series of formation flight instructions.

[0073] Step S3: Each UAV determines its own position according to the formation configuration matrix.

[0074] Each aircraft in the distributed UAV cluster has received exactly the same formation configuration space discrete matrix and formation flight mission setting instructions uploaded by the cluster ground measurement and control station. Multiple UAVs need to find their own roles and collaborate to complete the overall cluster formation flight mission through distributed formation task analysis and collaborative control.

[0075] In some optional implementations, step S3 further includes:

[0076] Step S31, extracting a first set consisting of non-zero value elements in the formation configuration matrix.

[0077] After receiving the formation configuration matrix C and the formation task binding instructions uploaded by the cluster ground control station, UAV p first performs task analysis and allocation to obtain the formation flight subtask to be performed by the UAV. Extract the first set of non-zero elements in the matrix C, C′ = [C′ ij ] P×Q .

[0078] Step S32, traverse the elements in the first set that specify the aircraft type, load type and ID in turn. If there is an element whose aircraft type, load type and ID are consistent with those of the aircraft itself, the aircraft will fly the formation node corresponding to the element, and determine the position of the aircraft according to the position of the element in the formation configuration matrix. Otherwise, execute step S33.

[0079] First, for the element C′ in C′ ij (i=1,2,…,P,j=1,2,…,Q) is analyzed. According to the previous description, it mainly includes the following situations:

[0080]

[0081] The drone type, payload type and ID required by the formation node are extracted and stored to complete the discrete matrix analysis of the formation configuration.

[0082] In step S32, the elements that meet the drone type, payload type and ID in the third case are mainly targeted. If the drone meets this rule, the drone is solidified. Specifically, at each terminal of the distributed drone, if node C′ ij (i=1,2,…,P,j=1,2,…,Q) The ID of the designated drone is consistent with the ID of the aircraft, and the type of drone and payload of the aircraft are compared to see if they are the same as C′ ij If the requirements are consistent, record the flight node C′ ij and its index in C, the task parsing and collaborative control process ends; if all elements in C′ are traversed without specifying the local ID for execution, it means that no element specifically specifies the local machine, and execution continues.

[0083] Step S33, determine a second set consisting of drones of the same aircraft type and payload type that can communicate with the local machine, eliminate the drones specified to fly each node in step S32, form a third set, and traverse each element in the first set with the specified aircraft type and payload type in turn. If there is an element whose aircraft type and payload type are consistent with those of the local machine, select a drone from the third set according to the agreed rules to fly the formation node corresponding to the element. If the local machine is selected to fly the node, determine the position of the local machine according to the position of the element in the formation configuration matrix. Otherwise, execute step S34.

[0084] Specifically, this step is mainly aimed at the case where the aircraft type and payload type are met, but the ID is not limited, and the second set T = {ID i}, traverse the elements in the collection, if a node in C' has an assigned ID i Execute and remove the element to get the third set T';

[0085] The formation nodes in the first set C′ that specify the aircraft / load type to be flown are traversed in sequence. If the aircraft meets the requirements of the node for the aircraft / load type, one of the aircraft in the first set and the third set T' is selected to perform the task according to the agreed rules; if the aircraft is selected to fly the formation node, the task parsing and collaborative control process ends; if the aircraft is not selected by all formation nodes that specify the aircraft / load type to be flown, continue to execute step S34.

[0086] Step S34, remove the drones designated to fly each node in step S33 from the third set to form a fourth set, traverse the remaining elements in the first set in turn, and select a drone from the fourth set to fly the formation node corresponding to the element according to the agreed rules. If the aircraft is selected to fly the node, the position of the aircraft is determined according to the position of the element in the formation configuration matrix, otherwise enter a waiting state.

[0087] In this step, the elements in the third set T' are traversed. If the drone has been assigned a task in step 33, the element is removed to obtain a fourth set T".

[0088] The formation nodes that can be executed by any UAV in the first set C′ are traversed in turn, and one of the aircraft in the fourth set T″ and the aircraft is selected to perform the task according to the agreed rules; if the aircraft is selected to fly the formation node, the task parsing and collaborative control process ends; if the aircraft is not selected by all formation nodes, the aircraft does not accept the formation flight mission and enters the waiting state.

[0089] In some optional implementations, the agreed rules include the drone ID from small to large, or the drone flight fee from low to high. In alternative implementations, the agreed rules may also include the health or fault status of the drone, the drone's onboard computing power, the payload carried by the drone, and the results obtained by linear or nonlinear calculation methods such as weighted sum based on the above parameters.

[0090] Step S4, each UAV makes an autonomous decision on flight according to the target position and set speed of the UAV cluster formation, wherein the autonomous decision-making flight includes, when the UAV cluster formation flies toward the target position, each UAV adjusts the heading angle according to its own position and the target position so as to keep the formation formation unchanged; and each UAV autonomously calculates the flight speed control instruction according to the flight set speed of the UAV cluster formation and the flight speed of other UAVs in the formation so as to make the expected arrival time of each UAV at the target position tend to be consistent.

[0091] After completing the task assignment, the drones need to perform distributed collaborative control to achieve high-precision formation flight. Through cluster collaborative formation flight control, drones at different speeds and in different locations form a stable formation and keep flying to the formation target position. The process is as follows: Figure 6 As shown:

[0092] First, the interactive collection of status information between UAVs is carried out. Secondly, the distributed formation collaborative control based on the mimicry physics algorithm and the UAV formation speed synchronization control based on the arrival time are respectively executed.

[0093] In terms of interactive collection of status information between drones, since there is no centralized planning node in the cluster, each drone needs to exchange flight status and mission information with other drones in the cluster. The information exchanged includes but is not limited to what is described in Table 3.

[0094] Table 3 Interaction status information between drones

[0095]

[0096]

[0097] From the status information synchronized with other UAVs, the system clock, flight status (position, speed, heading), remaining arrival time to the formation target position and other information of the UAV that maintains the communication connection can be extracted as the input of the formation collaborative control algorithm.

[0098] The distributed formation cooperative control based on the mimetic physics algorithm mainly involves adjusting the heading angle of each UAV. In some optional implementations, in step S4, adjusting the heading angle includes:

[0099] Step S41: determining a flight baseline based on the starting position and target position carried in the formation flight instruction.

[0100] The virtual flight starting point and target point of the aircraft are generated according to the starting position A and target position B of the formation mission. The directions of the virtual formation starting position and target position are the same. The line between the two is the baseline for this formation flight. Figure 7 In the embodiment, the respective baselines of the five UAVs in a V-shaped formation are identified.

[0101] Step S42: When the UAV deviates from the flight baseline, a virtual gravitational force is formed; when the distance between the UAV and an obstacle or other UAVs in the formation is less than a threshold, a virtual repulsive force is formed; and the heading angle of the UAV is corrected based on the resultant force of the virtual gravitational force and the virtual repulsive force.

[0102] The mimic physics algorithm assumes that there is a virtual force field in the flight space of the drone, in which the formation flight baseline generates a gravitational force F on the drone. att , guide the drones to form the correct formation configuration; when the drone is too close to obstacles or other drones (less than the safe flight distance R safe ) generates a repulsive force F on the machine rep , avoid the risk of collision; calculate the resultant force F of gravity and repulsion acting on the aircraft.

[0103] If the UAV is on the formation baseline at this time and the resultant force is 0, the flight heading should be F repDecomposed into radial force F along the direction of motion of the drone r and the normal force F n , where ||F|| determines the size of the heading angle correction Δψ (Δψ∈(-π / 2,π / 2)), and the direction of Fn determines the direction of Δψ. The expression of the correction Δψ is:

[0104] Δψ=sgn(F n )·arctan(||F||);

[0105] Among them, F n is the normal component of the virtual repulsive force perpendicular to the motion of the drone, and F is the resultant force of the virtual gravitational force and the virtual repulsive force. c =ψ0+△ψ. Under the action of the obstacle potential field, the virtual force and heading command generation strategy of the UAV are as follows: Figure 8 and Fig. 9 shown.

[0106] In an alternative implementation, the expression of the correction amount Δψ can also be set as:

[0107]

[0108] In some optional implementations, step S42 further includes:

[0109] When the distance between the UAV and other UAVs in the formation is less than a threshold, the flight altitude of the UAV is adjusted to further displace the two UAVs approaching each other in altitude.

[0110] In this step, when there are a large number of drones, the horizontal repulsion force is often difficult to ensure flight safety without collision, so the height adjustment is added; if the distance D between drones i and j is less than R safe , the height command is:

[0111]

[0112] In an alternative embodiment, the corrected height parameter H cmdi It can also be generated using the following function:

[0113] H cmdi =H swarm +a sgn(H i -H j )f(D,H i ,H j )

[0114] in,

[0115] In terms of the speed synchronization control of the drone formation based on the arrival time, the present application mainly solves the flight speed of each drone so that each drone arrives at the target point at the same time. In some optional embodiments, in step S4, the autonomous calculation of the flight speed control strategy includes:

[0116] Step S43, each UAV calculates the first time expectation of reaching the target position of the UAV;

[0117] Step S44, determining a second expected time for other drones communicating with the drone to reach the target location;

[0118] Step S45: using a first-order model to describe the speed of the UAV so that the first time expectation approaches the second time expectation, and solving the corresponding flight speed control instruction.

[0119] Specifically, suppose that UAV i can communicate with drones numbered C1, C2, …, C n The n drones have stable communication, assuming L i It is the estimated remaining path length of the i-th UAV along the planned path to reach the target location at time t. i Derivative with respect to t v i is the flight speed of the ith UAV at time t. Then the expected arrival time of the ith UAV is T i =t+L i / v i , define τ i =T i -t is the remaining expected time for the aircraft to reach the formation target position. The speed of the UAV formation flight is set to V swarm , then for other UAVs, the expected time to reach the designated position of the formation can be expressed as Among them, η ij is the weighted coefficient of each neighboring drone’s arrival time, The UAV swarm formation control requires that the expected arrival time of multiple UAVs at the target location should be consistent, that is,

[0120] Assume that the speed control of UAV i can be described by a first-order model: V cmdi is the speed command of UAV i, and the following formation speed control strategy is adopted:

[0121]

[0122] but:

[0123]

[0124] When t→∞, If the cluster communication network contains a directed spanning tree, all drones in the cluster can reach the formation target point at the same time.

[0125] Based on the above formula, each drone can calculate the speed command in a distributed manner.

[0126] Finally, through autonomous decision-making, the heading, speed, and altitude instructions of the UAV platform in this decision cycle are obtained and sent to the navigation and flight control system to control the movement of the UAV to complete the formation flight mission.

[0127] In the process of UAV clusters collaboratively performing formation flight missions, it is sometimes necessary to update the formation configuration, adjust the flight speed, scale the formation size, etc., which can also be efficiently performed through formation flight mission characterization and setting methods.

[0128] (1) Update the formation configuration: If the formation configuration discrete matrix C i If the flight has been uploaded to the drone cluster, a formation flight mission setting instruction is sent, the formation configuration number is i, and the latitude and longitude of the mission target location, flight speed and grid spacing are set; if the configuration C i If not uploaded, the formation configuration discrete matrix is ​​first sent to the UAV cluster.

[0129] (2) Adjust flight speed: Send a formation flight mission setting command to update the mission speed parameters.

[0130] (3) Scaling the formation size: If you need to enlarge the formation size, increase the element spacing parameter. Otherwise, reduce the element spacing parameter. Send a formation flight mission setting command to update the element spacing parameter.

[0131] When a UAV in flight receives the above formation change command, it needs to reorganize the formation according to the command type:

[0132] (1) Adjust flight speed: modify the formation flight speed parameters and execute the flight mission according to the new flight speed;

[0133] (2) Modify the formation target position: regenerate the formation target position and baseline according to the updated formation position, and re-form the formation under the guidance of the new baseline;

[0134] (3) Modify the formation spacing: regenerate the formation target position and baseline according to the new formation spacing, and re-form the formation under the guidance of the new baseline;

[0135] (4) Update the formation configuration: The ground tracking and control station sends instructions to the formation configuration C i Switch to C j , and the formation spacing remains unchanged; if the matrix C i With C jhave the same dimensions, and C j If there is a task node with the specified local ID in the , then compare the matrix element corresponding to the current formation position of the local to the local i With C j Whether they have the same rows and columns; if so, the position of the aircraft remains unchanged and the flight can continue; otherwise, the distributed task parsing and allocation process is executed to obtain the new formation task of the aircraft, and the formation target position and baseline are regenerated according to the updated formation position, and the formation can be re-formed under the guidance of the new baseline.

[0136] This application has the following beneficial effects:

[0137] (1) This application realizes efficient representation of formation configuration information based on the formation configuration space discrete matrix and the comprehensive coding of heterogeneous UAV flight capabilities. A simple formation configuration space discrete matrix and a flight mission setting instruction can characterize the formation's geometric configuration (relative orientation, aircraft spacing, etc.), formation flight speed, and formation topology node requirements for heterogeneous UAV flight capabilities; it greatly simplifies the description method of heterogeneous UAV cluster formation configuration for specific tasks and reduces the occupation of communication bandwidth when setting and changing formation configuration.

[0138] (2) This application simplifies the formation configuration design into the design of a formation configuration space discrete matrix. Users only need to fill in the corresponding formation space grid whether the formation topology node needs to be operated by a drone and what kind of drone capabilities are required to operate the drone. This allows people who have no knowledge or experience in drone clusters to design formation configurations based on specific mission scenarios and use drone clusters. This greatly improves the ease of use of drone clusters and expands application scenarios.

[0139] (3) After the user uploads the formation configuration space discrete matrix and formation setting tasks, the drone cluster described in this application can fly autonomously according to the formation configuration through distributed task allocation and collaborative control without the participation of ground measurement and control stations or any other central planning nodes; thus solving the problem of drone clusters' dependence on the computing and communication capabilities of the central planning nodes, and enabling drone clusters to obtain good robustness, flexibility and scalability.

[0140] (4) During formation flight, users can upload new formation configuration space discrete matrices or send formation task setting instructions according to mission requirements to easily adjust formation tasks, including formation configuration, formation target position, flight speed adjustment, and formation scaling. The drone cluster can autonomously adjust formation parameters or re-plan the formation according to the instructions, effectively improving the flexibility and ease of use of the drone cluster formation.

[0141] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A distributed UAV swarm formation and collaborative control method, It is characterized in that The method comprises: Step S1, discretize the horizontal space of the drone cluster formation into multiple grids corresponding to the formation configuration matrix. If a certain grid requires a drone to fly, the element value of the formation configuration matrix corresponding to the grid is set to a non-zero value according to the requirement for the drone to fly, otherwise the corresponding element value of the formation configuration matrix is ​​zero; Step S2: sending a formation flight instruction carrying a formation configuration number to each UAV, wherein the formation configuration number is associated with a corresponding formation configuration matrix; Step S3, each UAV determines its own position according to the formation configuration matrix; Step S4, each UAV makes an autonomous decision on flight according to the target position and set speed of the UAV cluster formation, wherein the autonomous decision-making flight includes, when the UAV cluster formation flies toward the target position, each UAV adjusts the heading angle according to its own position and the target position so as to keep the formation formation unchanged; and each UAV autonomously calculates the flight speed control instruction according to the flight set speed of the UAV cluster formation and the flight speed of other UAVs in the formation so as to make the expected arrival time of each UAV at the target position tend to be consistent.

2. The distributed UAV swarm formation and collaborative control method according to claim 1, It is characterized in that In step S1, the horizontal space of the formation is discretized into multiple grids based on the greatest common divisor of the expected spacing between the UAVs in the formation.

3. The distributed UAV swarm formation and coordinated control method as claimed in claim 1, It is characterized in that Step S2 further comprises: Multiple formation flight instructions are sent to each drone in sequence. When the drone cluster formation reaches the specified target task according to the formation configuration matrix given by the previous formation flight instruction, each drone transitions to the spatial node position specified by the formation configuration matrix given by the next formation flight instruction.

4. The distributed UAV swarm formation and collaborative control method according to claim 1, It is characterized in that Before step S3, it further includes uploading a plurality of preset formation configuration matrices to the drones for pre-storage, and each drone retrieves the corresponding formation configuration matrix according to the formation configuration number.

5. The distributed UAV swarm formation and collaborative control method according to claim 1, It is characterized in that Step S3 further comprises: Step S31, extracting a first set consisting of non-zero value elements in the formation configuration matrix; Step S32, traverse the elements in the first set that specify the aircraft type, load type and ID in turn, if there is an element whose aircraft type, load type and ID are consistent with the aircraft itself, the aircraft will fly the formation node corresponding to the element, and determine the position of the aircraft according to the position of the element in the formation configuration matrix, otherwise execute step S33; Step S33, determine a second set of drones of the same aircraft type and load type that can communicate with the local machine, remove the drones specified to fly each node in step S32, form a third set, traverse the elements of the first set that specify the aircraft type and load type in turn, if there is an element whose aircraft type and load type are consistent with the local machine, select a drone from the third set according to the agreed rules to fly the formation node corresponding to the element, if the local machine is selected to fly the node, determine the position of the local machine according to the position of the element in the formation configuration matrix, otherwise execute step S34; Step S34, remove the drones designated to fly each node in step S33 from the third set to form a fourth set, traverse the remaining elements in the first set in turn, and select a drone from the fourth set to fly the formation node corresponding to the element according to the agreed rules. If the aircraft is selected to fly the node, the position of the aircraft is determined according to the position of the element in the formation configuration matrix, otherwise enter a waiting state.

6. The distributed UAV swarm formation and coordinated control method as claimed in claim 5, It is characterized in that The agreed rules include drone IDs from small to large, or drone flight fees from low to high.

7. The distributed UAV swarm formation and coordinated control method according to claim 1, It is characterized in that In step S4, the steps of adjusting the heading angle of each drone include: Step S41, determining a flight baseline based on the starting position and target position carried by the formation flight instruction; Step S42: When the UAV deviates from the flight baseline, a virtual gravitational force is formed; when the distance between the UAV and an obstacle or other UAVs in the formation is less than a threshold, a virtual repulsive force is formed; and the heading angle of the UAV is corrected based on the resultant force of the virtual gravitational force and the virtual repulsive force.

8. The distributed UAV swarm formation and coordinated control method according to claim 7, It is characterized in that In step S42, the correction value Δψ of the heading angle includes: △ψ=sgn(F n )·arctan(||F||); Among them, F n is the normal component of the virtual repulsive force perpendicular to the motion of the drone, and F is the resultant force of the virtual attraction and the virtual repulsive force.

9. The distributed UAV swarm formation and coordinated control method according to claim 7, It is characterized in that Step S42 further includes: When the distance between the UAV and other UAVs in the formation is less than a threshold, the flight altitude of the UAV is adjusted to further displace the two UAVs approaching each other in altitude.

10. The distributed UAV swarm formation and coordinated control method according to claim 1, It is characterized in that In step S4, autonomously calculating the flight speed control strategy includes: Step S43, each UAV calculates the first time expectation of reaching the target position of the UAV; Step S44, determining a second expected time for other drones communicating with the drone to reach the target location; Step S45: using a first-order model to describe the speed of the UAV so that the first time expectation approaches the second time expectation, and solving the corresponding flight speed control instruction.

Images