A method, device and medium for distributed cooperative motion planning of a UAV swarm based on a flocking framework

Abstract

The application discloses a kind of distributed unmanned aerial vehicle cluster cooperative motion planning method, equipment and medium based on flocking framework, comprising: planning unmanned aerial vehicle cluster center ideal trajectory, generate following speed instruction;Establish unmanned aerial vehicle cluster aggregation model, construct aggregation speed instruction;Design coordination speed instruction, make the action of unmanned aerial vehicle in cluster tend to be consistent;Based on near neighbor unmanned aerial vehicle and local obstacle, construct the separation speed instruction of active steering;According to the speed instruction of unmanned aerial vehicle power limit adjustment, realize unmanned aerial vehicle cluster automatic obstacle avoidance, while along ideal trajectory cooperative flight.The application constructs tangential separation effect, avoids the instantaneous local minimum problem caused by separation effect and movement direction opposite, and by unmanned aerial vehicle autonomous selection passage, avoid the single passage width shrinkage and other easy to appear unmanned aerial vehicle accumulation deadlock condition, with low computational complexity, environment adaptability, distributed deployment scalability and other advantages.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cooperative motion planning of aerial robot swarm, in particular to a distributed unmanned aerial vehicle swarm cooperative motion planning method, device and medium based on a flocking framework. BACKGROUND

[0002] In recent years, rotor unmanned aerial vehicles have been widely used in industries such as express delivery, infrastructure inspection, agricultural monitoring and aerial photography. However, a single unmanned aerial vehicle is usually difficult to adapt to complete large and complex tasks, so cluster systems are increasingly concerned by the academic and industrial communities due to their performance advantages. However, the cooperative flight of unmanned aerial vehicle cluster systems in obstacle environments faces many challenges, such as insufficient coordination ability, limited scalability, poor obstacle environment adaptability, etc. Therefore, it is urgent to study a distributed unmanned aerial vehicle cluster cooperative flight motion planning method with autonomous obstacle avoidance capability to improve the scalability, autonomy and coordination ability of the cluster system, enhance the adaptability to complex environments and improve the task success rate.

[0003] Currently, the mainstream unmanned aerial vehicle cluster cooperative flight planning methods can be roughly divided into centralized and distributed types, each of which has its own characteristics in handling internal conflicts and external environment obstacle avoidance. The centralized planning method relies on a powerful central computing unit to generate globally optimal trajectories for all unmanned aerial vehicles in the cluster. This method models the entire cluster as a whole, which can effectively avoid internal collisions and accurately plan paths around known static obstacles. For example, using tools such as mixed integer linear programming to strictly guarantee the safety interval between trajectories. However, this method has poor scalability, the computing load of the central node increases sharply with the size of the cluster, and the reliability of the communication link is extremely high. Once the center fails or the communication is interrupted, the entire cluster will face the risk of paralysis. The distributed planning method requires unmanned aerial vehicle nodes to have higher autonomy and robustness. The classic artificial potential field-based unmanned aerial vehicle cluster cooperative flight method treats obstacles and other unmanned aerial vehicle nodes as sources of repulsive force, and adjusts the velocity vector in real time to achieve collision-free flight. This method has excellent scalability and is not sensitive to communication and single-point failures, but the traditional potential field model may cause motion jitter or fall into local optimum in complex obstacle terrain due to simple rules, making it difficult to guarantee strict time synchronization and accurate trajectory tracking.

[0004] The core of the above challenges is how to achieve global coordination of planning, large-scale scalability of the system and real-time flexibility of local obstacle avoidance in an unknown obstacle environment. Researchers need to balance the needs of inter-node trajectory coordination, cluster size scalability, flexible obstacle avoidance response and other aspects to improve the safety and flexibility of the cooperative flight of the quadrotor unmanned aerial vehicle cluster. In order to further promote the application of unmanned aerial vehicle cluster in cooperative flight operation, regional exploration and other aspects, it is urgent to invent a safe and flexible unmanned aerial vehicle cluster cooperative flight motion planning method. SUMMARY

[0005] In order to solve the technical problems of large calculation consumption, insufficient environmental adaptability and poor scalability in the current unmanned aerial vehicle cluster cooperative flight, the present application provides a distributed unmanned aerial vehicle cluster cooperative motion planning method based on a flocking framework, equipment and medium, which has the advantages of low calculation complexity, strong environmental adaptability, high scalability of distributed deployment and the like.

[0006] In order to achieve the above technical purpose, the technical scheme of the present application is:

[0007] A distributed unmanned aerial vehicle cluster cooperative motion planning method based on a flocking framework, comprising the following steps:

[0008] S100: According to the initial position and target position of the unmanned aerial vehicle cluster, the center of the unmanned aerial vehicle cluster is planned to move from the initial position to the target position, and the time of each waypoint is obtained by combining the preset flight speed, so as to form an ideal trajectory, and then during the motion of the unmanned aerial vehicle cluster, the following speed instruction of the unmanned aerial vehicle flying along the preset trajectory is generated based on the deviation of the real-time position and the ideal trajectory;

[0009] S200: A sphere with double-layer spherical shells is established as the shape of the unmanned aerial vehicle cluster, and then a unmanned aerial vehicle aggregation speed instruction effective between the two layers of spherical shells is constructed to form a dense unmanned aerial vehicle cluster;

[0010] S300: The parameters including the maximum acceleration and maximum speed of the unmanned aerial vehicle, the minimum relative speed between the unmanned aerial vehicles allowed for maintaining the flexibility of the cluster, the effective range of executing the minimum relative speed, and the distance between the unmanned aerial vehicles for the nearby state are comprehensively considered to calculate the maximum speed allowed between the nearby unmanned aerial vehicles in the unmanned aerial vehicle cluster, and the unmanned aerial vehicle coordination speed instruction is generated to make each unmanned aerial vehicle move consistently in the cluster;

[0011] S400: Based on the near neighbor unmanned aerial vehicle and the local obstacle distribution, the boundary distance for each unmanned aerial vehicle in the cluster to execute avoidance is constructed, and then the separation speed instruction for active turning is generated to realize autonomous avoidance of the near neighbor unmanned aerial vehicle and the environmental obstacle;

[0012] S500: By integrating the above speed commands and adjusting them according to the drone's power limitations, the final drone speed command is generated, enabling the drone swarm to automatically avoid obstacles and fly collaboratively along an ideal trajectory.

[0013] Furthermore, in the method, S100 includes:

[0014] S110: Use the current location of the UAV cluster center as the initial location of the cluster center, and then use the A-Star algorithm to generate the waypoint sequence of the cluster center according to the target location;

[0015] S120: Generates an ideal trajectory for the cluster center based on the waypoint sequence and the preset average flight speed, consisting of the position-time sequence of waypoints;

[0016] S130: During swarm movement, acquire the swarm center in real time, and calculate the coordinates of the closest position to the real-time swarm center on the ideal trajectory. This coordinate is used as the real-time expected position of the swarm center. ;

[0017] S140: Take the ideal trajectory and with The sequence of two adjacent waypoint positions constitutes a motion segment. The desired velocity of the UAV swarm center is calculated by dividing the trajectory length of the motion segment by the duration of the motion segment. The magnitude of this value, normalized to the displacement vector of the motion segment, serves as the desired velocity at the center of the UAV swarm. The direction vector is used to determine the desired velocity of the cluster center. ;

[0018] S150: Will The speed command is issued to the drone swarm to guide the drone swarm to fly along the ideal trajectory.

[0019] Furthermore, the center of the drone swarm is obtained by averaging the positions of each node in the swarm, i.e., the individual drones.

[0020] Furthermore, in the method, S200 includes:

[0021] S210: Real-time desired location of the drone swarm center Centered on a double-layered spherical shell, a target cluster shape is constructed. The inner spherical shell accommodates the free hovering of the UAV cluster, while the outer spherical shell represents the maximum boundary of the aggregation effect. The range of action of the aggregation velocity command of any UAV i in the UAV cluster lies between the two spherical shells, as shown below:

[0022] ;

[0023] wherein, represents the aggregation instruction acting on the UAV i; is the distance between the UAV i and the UAV cluster center, , represents the position of the UAV i; is the negative feedback gain coefficient of the aggregation instruction, ; and are the radii of the inner and outer spherical shells, respectively; represents the direction vector from the cluster center to the UAV i, .

[0024] Further, the inner spherical shell radius of the double-layer spherical shell , the outer spherical shell radius is obtained by the following steps:

[0025] S220: According to the UAV size and the number of cluster nodes, determine the inner spherical shell radius of the double-layer spherical shell :

[0026] ;

[0027] wherein represents the spherical space radius required for a single UAV to hover freely, which should meet the packing radius constraint condition of the cluster under the dense hovering condition; is the number of UAVs in the cluster; represents the space utilization rate of the dense arrangement of UAVs, ; while taking 2-3 times of as ;

[0028] S230: Calculate the packing radius constraint condition of the cluster under the dense hovering condition:

[0029] ;

[0030] wherein, b is the semi-minor axis parameter of the ellipsoid of separation action; is the UAV radius.

[0031] Further, in the method, S300 comprises:

[0032] S310: For a pair of adjacent UAVs i and j, construct the speed alignment instruction form generated by i affected by j:

[0033] ;

[0034] wherein, represents the speed alignment instruction generated by i affected by j; is the magnitude of the relative velocity between UAVs i and j, is the direction vector of the relative velocity between UAVs i and j; is the negative feedback gain coefficient of the alignment instruction; represents the maximum relative velocity allowed between UAVs i and j; represents the maximum flight velocity of a UAV, is the maximum relative velocity between UAVs; wherein as the boundary to trigger the alignment instruction:

[0035] ;

[0036] ;

[0037] wherein, is the square of the maximum velocity allowed between UAVs i and j as an intermediate variable; is the minimum relative velocity allowed to maintain the flexibility of the swarm; is the effective range of ; is the maximum acceleration limit of a UAV; is the distance between UAVs i and j; is the maximum effective distance of the alignment instruction, used to screen nearby UAVs, is the radius of a UAV;

[0038] S320: Synthetically calculate the alignment effect of all nearby UAVs on UAV i, specifically:

[0039] ;

[0040] wherein, represents the velocity alignment instruction generated by UAV i under the influence of all nearby UAVs; is the set of nearby UAVs within the maximum effective distance from UAV i.

[0041] Further, in the method, S400 comprises:

[0042] S410: Construct a separation effect distribution model of UAV i:

[0043] ;

[0044] wherein, k represents a UAV or an obstacle near UAV i; represents the boundary distance of UAV i that needs to avoid k, composed of a rotating ellipsoid term and a spherical term; is the semi-minor axis parameter of the ellipsoid of the separation effect; is the eccentricity of the ellipsoid determined by the speed of UAV i; , is the deflection angle of k relative to i, with the speed direction of UAV i as the reference direction ; is the radius of the sphere term, which is also the collision distance between i and k, and is selected as different values for UAVs or obstacle points; is the radius of the UAV; is calculated by the following equation:

[0045] ;

[0046] ;

[0047] wherein, is the eccentricity of the ellipsoid corresponding to the maximum speed of the UAV i; is the real-time speed of UAV i; represents the obstacle perception radius of the UAV;

[0048] S420: Based on the separation effect distribution model of UAV i, the separation speed instruction of UAV i affected by object k is constructed:

[0049] ;

[0050] wherein, represents the separation speed instruction of UAV i affected by k; is the gain coefficient of the separation instruction; represents the minimum boundary of ; represents the direction vector perpendicular to the speed direction of UAV i and away from k;

[0051] S430: The separation effects of all adjacent objects on UAV i are comprehensively calculated:

[0052] ;

[0053] wherein, represents the separation speed instruction of UAV i affected by all adjacent objects; is the set of adjacent objects within the effective range of the separation effect to UAV i, including the UAV and the environmental obstacles.

[0054] Further, in the method, S500 comprises:

[0055] S510: The comprehensive speed instruction of the UAV is formed by comprehensively calculating the various speed instructions obtained from S100, S200, S300 and S400 :

[0056] ;

[0057] S520: Adjust the comprehensive speed instruction according to the dynamic limit of the maximum acceleration of the unmanned aerial vehicle, and obtain a temporary variable :

[0058] ;

[0059] wherein, is the current speed of the unmanned aerial vehicle i; is the maximum acceleration of the unmanned aerial vehicle; is the time step of the entire cooperative planning method; is the desired speed change amount in , and are the size and direction vector of , respectively;

[0060] S530: Adjust the speed instruction again according to the dynamic limit of the maximum speed of the unmanned aerial vehicle and output to the unmanned aerial vehicle for execution:

[0061] ;

[0062] wherein, is the speed instruction finally output to the unmanned aerial vehicle; is the maximum speed of the unmanned aerial vehicle; and are the size and direction vector of , respectively;

[0063] S540: During the flight of the unmanned aerial vehicle, update the surrounding obstacle map of the unmanned aerial vehicle and the real-time position and speed of the center of the unmanned aerial vehicle cluster every , and re-execute S130, S140, S150, S200, S300, S400, S510, S520 and S530.

[0064] The application also provides an electronic device, comprising:

[0065] one or more processors;

[0066] a storage device for storing one or more programs,

[0067] when the one or more programs are executed by the one or more processors, so that the one or more processors implement the method as described above.

[0068] The application also provides a computer readable medium storing a computer program, which is executed by a processor to implement the method as described above.

[0069] The technical effect of the present application is that the present application proposes a motion planning framework for cooperative flight of a UAV cluster in an obstacle environment, and realizes the combination of four motion instructions of "cluster following - regional aggregation - speed coordination - separation obstacle avoidance" of the UAV cluster in the obstacle environment; proposes a separation speed instruction construction method of active steering, improves the motion efficiency and flexibility of the UAV in avoiding obstacles, so as to meet the task requirements of the cooperative flight of the UAV cluster in the unknown obstacle environment, and realize the autonomous adaptation of the UAV cluster to the environment and the active obstacle avoidance cooperative flight. Compared with other flocking framework-based cluster planning methods, the present application avoids the problem of instantaneous local minimum caused by the fact that most separation actions are opposite to the motion direction by constructing a tangential separation action, and avoids the situation of UAV accumulation deadlock caused by the narrowing of the single passage width by autonomous selection of the UAV, and has the advantages of low computational complexity, strong environmental adaptability, distributed deployment scalability and the like.

[0070] The present application will be further described below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0071] Figure 1 The flowchart of the present application.

[0072] Figure 2 The specific flowchart of the present application on any UAV node.

[0073] Figure 3 The environment map of the simulation experiment of the present application; wherein the cylindrical body represents the position of the obstacle in the simulation environment.

[0074] Figure 4 The simulation experiment trajectory graph of the present application in the simulation scene; wherein (a) and (b) are respectively the three-dimensional view and the top view of the cooperative flight trajectory of the UAV cluster in the obstacle region,

[0075] Figure 5 The change curve of the UAV cluster position in the X, Y and Z three axes during the simulation experiment process of the present application in Figure 4

[0076] Figure 6 The distance change graph of the UAVs with time during the simulation experiment process of the present application in Figure 4 DETAILED DESCRIPTION

[0077] In one embodiment, as shown in Figure 1 , a distributed UAV cluster cooperative motion planning method based on the flocking framework is provided, comprising the following steps:

[0078] ​​S100: According to the initial position and target position of the UAV cluster, a flight point sequence of the UAV cluster center moving from the initial position to the target position is planned, and the time of each flight point is obtained in combination with the preset flight speed, so as to form an ideal trajectory. Then, during the movement of the UAV cluster, a following speed instruction of the UAV flying along the preset trajectory is generated based on the deviation of the real-time position from the ideal trajectory.

[0079] S200: A spherical body with double-layer spherical shells is used as the shape of the UAV cluster, and a UAV aggregation speed instruction is constructed. Then, a UAV aggregation speed instruction effective between the two layers of spherical shells is constructed, so as to establish a UAV cluster aggregation model and form a dense UAV cluster.

[0080] S300: A UAV coordination speed instruction is designed, which comprehensively includes parameters such as the maximum acceleration and maximum speed of the UAV, the minimum relative speed allowed between the UAVs for maintaining the flexibility of the cluster, the effective range of executing the minimum relative speed, and the distance between the UAVs for determining the proximity state. The maximum speed allowed between the proximate UAVs in the UAV cluster is calculated, and the UAV coordination speed instruction is generated, so that the actions of each UAV in the cluster tend to be consistent.

[0081] S400: Based on the proximate UAVs and the local obstacle distribution, a boundary distance for each UAV in the cluster to execute avoidance is constructed, and a separation speed instruction for active turning is generated, so as to realize autonomous avoidance of the UAV from the proximate UAVs and environmental obstacles.

[0082] S500: The above speed instructions are integrated, and adjusted according to the power limitation of the UAV to ensure that the instructions are realizable, so as to finally obtain the UAV speed instruction and send it to the UAV for execution, so as to realize automatic obstacle avoidance of the UAV cluster while flying cooperatively along the ideal trajectory.

[0083] In one embodiment, S100 includes:

[0084] S110: The positions of each node, i.e., a single UAV, in the UAV cluster are averaged to serve as the initial position of the cluster center. Then, according to the target position of the cluster center, an A-Star algorithm is used to generate a flight point sequence of the cluster center.

[0085] S120: According to the preset cluster flight point sequence and average flight speed, a cluster center ideal trajectory composed of a position-time sequence is generated.

[0086] S130: During the movement of the cluster, the average position of each node of the UAV is calculated in real time to obtain the real-time position of the UAV cluster center, and the position coordinate on the ideal trajectory closest to the real-time position of the UAV cluster center is solved to serve as the real-time expected position of the UAV cluster center. The existing distance calculation methods can be used to obtain the result in the specific solution.

[0087] S140: taking the ideal trajectory on and the direction vector of The position sequence of the two adjacent waypoints before and after constitutes a motion segment, and the length of the trajectory of the motion segment is divided by the time length of the motion segment to obtain the magnitude of the expected speed of the UAV cluster center , wherein the time length is obtained from the time sequence in S120. The displacement vector of the motion segment is normalized to obtain the direction vector of the expected speed of the UAV cluster center , thereby determining the expected speed of the cluster center .

[0088] S150: taking as the following speed instruction to issue to the UAV cluster to realize the guidance of the UAV cluster along the ideal trajectory.

[0089] In one embodiment, S200 includes:

[0090] S210: taking the real-time expected position of the UAV cluster center as the center, a double-layer spherical shell is constructed as the target cluster shape, wherein the inner spherical shell of the double-layer spherical shell is used to accommodate the free hovering of the UAV cluster, and the outer spherical shell is the maximum boundary of the aggregation action. The action range of the aggregation speed instruction of any UAV i in the UAV cluster is between the double-layer spherical shells, specifically:

[0091] ;

[0092] wherein represents the aggregation instruction acting on the UAV i. is the distance between the UAV i and the UAV cluster center, represents the position of the UAV i. is the negative feedback gain coefficient of the aggregation instruction, which is taken as in this embodiment. and are the radii of the inner and outer spherical shells, respectively. represents the direction vector from the cluster center to the UAV i.

[0093] S220: according to the UAV size and the number of cluster nodes, the inner spherical shell radius of the double-layer spherical shell is determined, specifically:

[0094] ;

[0095] wherein A spherical space radius required for single UAV free hovering, should satisfy the packing radius constraint condition of the swarm under the dense hovering condition. is the number of UAVs in the swarm. is the space utilization ratio of the dense arrangement of UAVs. Then the embodiment takes 2 times of as , and the size of may also be adjusted according to specific needs, such as taking 3 times .

[0096] S230: Calculate the packing radius constraint condition of the swarm under the dense hovering condition:

[0097] ;

[0098] wherein, is the semi-minor axis parameter of the ellipsoid of the separation effect, and in the embodiment, 3 times of the UAV radius is taken. is the UAV radius. The thus constructed can gather the UAVs in the spherical space with a radius and keep free hovering.

[0099] In one embodiment, S300 includes:

[0100] S310: For a pair of adjacent UAVs i and j, construct the form of the speed alignment instruction generated by the UAV i affected by j, specifically:

[0101] ;

[0102] wherein, represents the speed alignment instruction generated by the UAV i affected by j. is the magnitude of the relative speed between the UAV i and j, is the direction vector of the speed of i relative to j. is the negative feedback gain coefficient of the alignment instruction, and in the embodiment, 2 is taken. . represents the maximum allowed relative speed between i and j. Taking represents the maximum flight speed of the UAV, so is the maximum relative speed between the UAVs.

[0103] S320: Calculate the maximum allowed speed between the UAVs i and j as the boundary of triggering the alignment instruction, specifically:

[0104] ;

[0105] ;

[0106] wherein, is the square of the maximum allowable speed between UAVs i, j as an intermediate variable. is the minimum relative speed allowed to maintain swarm agility, in this embodiment , is the effective range of . is the maximum acceleration limit of the UAV. is the distance between UAVs i, j. is the maximum effective distance of the alignment instruction, used to filter out nearby UAVs.

[0107] S330: Calculate the alignment effect of all nearby UAVs on UAV i, specifically:

[0108] ;

[0109] wherein, represents the speed alignment instruction of UAV i affected by all nearby UAVs. The set of nearby UAVs within the maximum effective distance is reached to UAV i. The can make the speed of nearby UAVs relatively consistent, improving the consistency of swarm action.

[0110] In one embodiment, S400 includes:

[0111] S410: Construct the separation effect distribution model of UAV i, specifically:

[0112] ;

[0113] wherein, k represents a UAV or an obstacle near UAV i. represents the boundary distance of UAV i that needs to avoid k, composed of a rotating ellipsoid term and a spherical term. is the ellipsoid semi-minor axis parameter of the separation effect mentioned in S230, in this embodiment, 3 times the radius of the UAV. is the eccentricity of the ellipsoid determined by the speed of UAV i. is the deflection angle of k relative to i, with the speed direction of UAV i as .

[0114] is the radius of the spherical term, which is also the collision distance between i and k, and different values are selected for UAVs or obstacle points. is calculated by the following equation:

[0115] ;

[0116] ;

[0117] wherein, is the corresponding ellipsoid eccentricity of the UAV at the maximum speed . is the real-time speed of the UAV i. represents the obstacle awareness radius of the UAV.

[0118] S420: based on the separation effect distribution model of the UAV i, a separation speed instruction of the UAV i affected by the object k is constructed, specifically:

[0119] ;

[0120] wherein, represents the separation speed instruction of the UAV i affected by k. is the gain coefficient of the separation instruction, which is taken as in the embodiment. represents the minimum boundary of . represents the direction vector perpendicular to the speed direction of the UAV i and away from k.

[0121] S430: the separation effects of all adjacent objects on the UAV i are comprehensively calculated, specifically:

[0122] ;

[0123] wherein, represents the separation speed instruction of the UAV i affected by all adjacent objects. is achieved to the set of adjacent objects within the effective range of the separation effect, including the UAV and the environmental obstacles. The thus constructed can enable the UAV to achieve autonomous avoidance of surrounding environmental obstacles and other UAVs by active steering.

[0124] In one embodiment, S500 includes:

[0125] S510: the various speed instructions obtained in S100, S200, S300, and S400 are comprehensively combined to form a comprehensive speed instruction of the UAV, specifically:

[0126] ;

[0127] S520: according to the dynamic limit of the maximum acceleration of the UAV, the comprehensive speed instruction is adjusted to obtain a temporary variable , specifically:

[0128] ;

[0129] in, This is the current speed of drone i. That is the maximum acceleration of the drone. It is the time step of the entire collaborative planning method. yes The expected change in velocity within the range, and They are The magnitude and direction vector.

[0130] S530: Based on the dynamic limitations of the drone's maximum speed, the speed command is readjusted and output to the drone for execution, specifically as follows:

[0131] ;

[0132] in, This is the speed command that is ultimately output to the drone in this embodiment. That is the maximum speed of the drone. and They are The magnitude and direction vector.

[0133] S540: During the drone's flight, every... Update the obstacle map around the drone and the real-time location and speed of the drone swarm center, and re-execute S130, S140, S150, S200, S300, S400, S510, S520, and S530.

[0134] This embodiment provides a distributed UAV swarm cooperative motion planning method based on the flopcking framework, in situations such as... Figure 3 Simulation verification was performed in the simulation environment shown, with a cluster of three drones in [the simulated environment]. Figure 3 The simulation environment shown demonstrates physical simulation of cooperative flight along a desired circular trajectory. Figure 3 The three blue circles on the left represent three drone nodes equipped with lidar obstacle perception capabilities, while the cylinder on the right represents the location of obstacles in the simulated environment. The cluster of three drone nodes runs the planning method provided in this embodiment. The ideal trajectory of the cluster center is set as follows: Figure 4 The circular curve shown for "Drone 0" represents the actual trajectories of each drone node in the cluster, obtained from the simulation experiment. These trajectories are curves for drone 1, drone 2, and drone 3, respectively. Figure 4 (a) is a three-dimensional view of the collaborative flight trajectory of the UAV swarm in the obstacle area, and (b) is a top view of the collaborative flight trajectory of the UAV swarm in the obstacle area. Figure 5 This is a graph showing the change in the drone swarm's position along the X, Y, and Z axes. According to... Figure 4 as well asFigure 5 According to the experimental results shown, it can be found that the UAV cluster can autonomously adapt to the obstacle environment, actively avoid obstacles, and select different feasible channels to pass through the obstacle region. Meanwhile, in the non-obstacle environment, the UAVs can quickly gather around the ideal trajectory of the center to form a dense cluster. Meanwhile, according to the UAV distance-time curve shown, it can be found that no collision occurs between the UAVs in the cooperative flight process. Figure 6

[0135] According to the embodiments of the present application, the present application also provides an electronic device and a computer readable medium.

[0136] The electronic device comprises:

[0137] one or more processors.

[0138] a storage device for storing one or more programs,

[0139] When the one or more programs are executed by the one or more processors, the one or more processors implement the aforementioned distributed UAV cluster cooperative motion planning method.

[0140] In specific use, the user can interact with the server which is also an electronic device through the electronic device as a terminal device and based on the network to realize the functions of receiving or sending messages, etc. The terminal device is generally various electronic devices provided with display devices and used based on human-computer interface, including but not limited to smart phones, tablet computers, notebook computers and desktop computers, etc. Various specific application software can be installed on the terminal device as needed, including but not limited to web browser software, instant messaging software, social platform software, shopping software, etc.

[0141] The server is a network server for providing various services, and the distributed UAV cluster cooperative motion planning method provided in the present embodiment is generally executed by the server. In actual application, the terminal device can also directly execute the distributed UAV cluster cooperative motion planning method under the condition of meeting the necessary conditions.

[0142] Similarly, the computer readable medium of the present application has a computer program stored thereon, and the computer program is executed by the processor to implement a distributed UAV cluster cooperative motion planning method of the present application.

[0143] ​The above describes in detail the method for the flocking framework-based distributed UAV cluster cooperative motion planning provided by the application. The principles and implementation manners of the application are described by using specific examples, and the above description of the examples is only used to help understand the core idea of the application. It should be pointed out that, for those skilled in the art, some improvements and modifications can be made to the application without departing from the principles of the application, and these improvements and modifications also fall within the protection scope of the claims of the application.

Claims

1. A method for flocking framework based distributed UAV swarm cooperative motion planning, characterized in that, The method comprises the following steps: S100: according to the initial position and the target position of the UAV cluster, a waypoint sequence of the UAV cluster center moving from the initial position to the target position is planned, and the time of each waypoint is obtained in combination with a preset flight speed to form an ideal trajectory, and then during the movement of the UAV cluster, a following speed instruction of the UAV flying along the preset trajectory is generated based on the deviation of the real-time position and the ideal trajectory; S200: a sphere with double spherical shells is established as the shape of the UAV cluster, and then a UAV gathering speed instruction effective between the two spherical shells is constructed to form a dense UAV cluster; S300: parameters including the maximum acceleration and the maximum speed of the UAV, the minimum relative speed allowed between the UAVs for maintaining the flexibility of the cluster, the effective range of the minimum relative speed, and the distance between the UAVs for the proximity state are comprehensively calculated to obtain the maximum speed allowed between the proximate UAVs in the UAV cluster, and a UAV coordination speed instruction is generated to make the UAVs move in unison in the cluster; S400: based on the proximate UAVs and the local obstacle distribution, a boundary distance for each UAV in the cluster to perform avoidance is constructed, and then a separation speed instruction for active turning is generated to realize the autonomous avoidance of the UAVs from the proximate UAVs and the environmental obstacles; S500: the above speed instructions are comprehensively combined, and the UAV speed instruction is finally generated according to the power limitation of the UAV to realize the automatic obstacle avoidance of the UAV cluster while flying cooperatively along the ideal trajectory; S200 comprises: S210: the real-time expected position of the UAV cluster center As the center, a double-layer spherical shell is constructed as the target cluster shape, wherein the inner spherical shell of the double-layer spherical shell is used to accommodate the UAV cluster free hovering, and the outer spherical shell is the maximum boundary of the aggregation effect. The range of the aggregation speed instruction of any UAV i in the UAV cluster is between the double-layer spherical shells, which is expressed as: ； wherein, represents a gathering instruction acting on the UAV i; is the distance between the UAV i and the cluster center, , represents the position of the UAV i; is a negative feedback gain coefficient of the gathering instruction, ; and are the radii of the inner and outer spherical shells, respectively; represents a direction vector from the cluster center to the UAV i, ; the inner sphere radius of the double-layered sphere shell , the outer sphere radius of the double-layered sphere shell is obtained by the following steps: S220: determining the inner spherical shell radius of the double-layer spherical shell according to the size of the unmanned aerial vehicle and the number of cluster nodes : ； in The spherical spatial radius required for a single drone to hover freely should satisfy the stacking radius constraint condition of the swarm under dense hovering conditions; It refers to the number of drones in the cluster; This indicates the space utilization rate of densely packed drones. Simultaneously take 2-3 times as ; S230: the packing radius constraint condition of the cluster under the condition of dense hovering is calculated: ； wherein, b is the ellipsoid semi-minor axis parameter of the separation; is the drone radius.

2. The method of claim 1, wherein, S100 comprises: S110: the position of the current UAV cluster center is taken as the initial position of the cluster center, and then the A-Star algorithm is used to generate the waypoint sequence of the cluster center according to the target position; S120: the cluster center ideal trajectory composed of the position-time sequence of the waypoints is generated according to the waypoint sequence and the preset average flight speed; S130: During the cluster motion, the real-time UAV cluster center is acquired, the position coordinate closest to the real-time UAV cluster center on the ideal trajectory of the cluster center is solved, and the real-time expected position of the UAV cluster center is taken as the real-time expected position of the UAV cluster center ; S140: taking the ideal trajectory on and with The sequence of two adjacent waypoint positions in front and back constitutes a motion segment, and the length of the trajectory of the motion segment divided by the time length of the motion segment is taken as the expected speed of the UAV cluster center The displacement vector of the motion segment is normalized as the direction vector of the expected speed of the UAV cluster center , thereby determining the expected speed of the cluster center ; S150: send As the follow speed instruction is issued to the UAV cluster, the UAV cluster is guided to fly along the ideal trajectory.

3. The method of claim 2, wherein, The center of the UAV cluster is the average value of the positions of all nodes, i.e., single UAVs in the UAV cluster.

4. The method of claim 1, wherein, S300 comprises: S310: for a pair of proximate UAVs i and j, the speed alignment instruction form of i affected by j is constructed: ； wherein, represents a velocity alignment instruction generated by i being affected by j; is the magnitude of the relative velocity between the drones i and j, is the direction vector of the relative velocity of i with respect to j; is the negative feedback gain coefficient of the alignment instruction; represents the maximum relative velocity allowed between i and j; represents the maximum flight speed of the drones, is the maximum relative velocity between the drones; wherein serves as a limit to trigger the alignment instruction: ； ； wherein, is the square of the maximum allowed speed between drones i, j as an intermediate variable; is the minimum relative speed magnitude allowed to maintain swarm agility; is the effective range of is the maximum acceleration limit of a drone; is the distance between drones i, j; is the maximum effective distance of an alignment instruction used to filter out close drones, is the drone radius;​ S320: the alignment effects of all proximate UAV pairs i are comprehensively calculated, and the specific calculation is as follows: ； wherein, represents a speed alignment instruction for the drone i resulting from the influence of all proximate drones; communicates to the drone i the set of proximate drones whose distance is within the maximum effective distance.

5. The method of claim 4, wherein, S400 comprises: S410: the separation effect distribution model of UAV i is constructed: ； where k represents a UAV or an obstacle near UAV i; represents the boundary distance that UAV i needs to avoid k, which is composed of a rotating ellipsoid term and a sphere term; b is the ellipsoid semi-minor axis parameter of the separation effect; is the ellipsoid eccentricity determined by the speed of UAV i; , is the deflection angle of k relative to i, with the speed direction of UAV i as the ; is the sphere term radius, which is also the collision distance between i and k, and different values are selected for k as a UAV or an obstacle point; is the UAV radius; is calculated by the following equation: ； ； wherein, is the corresponding ellipsoid eccentricity of the UAV at the maximum speed of the UAV i; is the real-time speed magnitude of the UAV i; denotes the obstacle perception radius of the UAV; S420: based on the separation effect distribution model of UAV i, the separation speed instruction of UAV i affected by object k is constructed: ； wherein, represents a separation speed command of the drone i influenced by k; is a gain coefficient of the separation command; represents a minimum boundary of represents a direction vector perpendicular to the drone i speed direction and away from k; S430: the separation effects of all proximate objects on UAV i are comprehensively calculated: ； wherein, represents the separation velocity command of the UAV i resulting from the influence of all the proximate objects; is communicated to the UAV i the set of proximate objects within the effective range of the separation action, including the UAV and the environmental obstacles.

6. The method of claim 5, wherein, S500 comprises: S510: Synthesizing each speed command obtained from S100, S200, S300 and S400 to form a comprehensive speed command of the UAV : ； S520: Adjust the integrated speed command according to the dynamic limit of the maximum acceleration of the UAV, to obtain a temporary variable : ； wherein, is the current speed of the drone i; is the maximum acceleration of the drone; is the time step of the whole cooperative planning method running; is the desired speed change amount within the time step, and are the size and direction vector of respectively; S530: the speed instruction is adjusted again according to the dynamic limitation of the maximum speed of the UAV and is output to the UAV for execution: ； wherein, is the final output to the drone speed command; is the maximum speed of the drone; and are the magnitude and direction vector of respectively; S540: During the flight of the UAV, every The UAV surrounding obstacle map and the real-time position and speed of the UAV cluster center are updated, and S130, S140, S150, S200, S300, S400, S510, S520, and S530 are re-executed.

7. An electronic device, comprising: It comprises: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1-6.

8. A computer readable medium storing a computer program, characterized in that, The computer program is executed by the processor to implement the method of any one of claims 1-6.

Images