Unmanned aerial vehicle cooperative hoisting control method and device, computer device and storage medium

Abstract

The application relates to the technical field of unmanned aerial vehicles, in particular to an unmanned aerial vehicle cooperative hoisting control method and device, computer equipment and a storage medium. The unmanned aerial vehicle cooperative hoisting control method comprises the following steps: being applied to an unmanned aerial vehicle cooperative hoisting system, the unmanned aerial vehicle cooperative hoisting system at least comprising multiple unmanned aerial vehicle groups and loads connected with the unmanned aerial vehicle groups, and the method comprising the following steps: constructing a dissipation node system of the unmanned aerial vehicle cooperative hoisting system; constructing a virtual Lagrange model of the dissipation node system according to a Lagrange function, a dissipation function and generalized coordinates of each virtual node in the dissipation node system; determining a control law of each unmanned aerial vehicle and a node expected acceleration of each virtual node according to the virtual Lagrange model; and controlling the unmanned aerial vehicles corresponding to the virtual nodes to operate according to the control law and the node expected acceleration of the virtual nodes. The application provides an unmanned aerial vehicle cooperative hoisting control method which does not need communication, has high robustness and scalability.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) technology, and more specifically, to a UAV collaborative hoisting control method, apparatus, computer equipment, and storage medium. Background Technology

[0002] With the rapid development of drone technology, the potential of drones in logistics, rescue, and industrial applications is becoming increasingly apparent. Especially in scenarios requiring the transport of heavy or large loads, the limited carrying capacity of a single drone makes it difficult to meet the demands. Therefore, multi-drone collaborative transportation has become a research hotspot. In existing technologies, collaborative transportation typically relies on centralized control or real-time communication between drones, which increases system complexity and dependence on communication networks. In complex environments, such as high-altitude or communication-restricted scenarios, existing centralized or communication-dependent methods are prone to failure due to communication interruptions or insufficient computing resources. Furthermore, existing methods lack sufficient robustness and adaptability when facing drone heterogeneity, cable length uncertainties, or drone malfunctions.

[0003] Therefore, there is an urgent need for a drone collaborative hoisting control method that requires no communication, has high robustness and scalability, in order to address the shortcomings of existing technologies. Summary of the Invention

[0004] This application provides a method, apparatus, computer equipment, and storage medium for collaborative hoisting control of unmanned aerial vehicles (UAVs).

[0005] A first aspect of this application provides a drone collaborative lifting control method applied to a drone collaborative lifting system, the drone collaborative lifting system comprising at least multiple drone groups and loads connected to each of the drone groups, the method comprising: Construct a dissipative node system for the UAV collaborative hoisting system; wherein, each virtual node in the dissipative node system corresponds one-to-one with each UAV in the UAV collaborative hoisting system; A virtual Lagrange model of the dissipative node system is constructed based on the Lagrange function, dissipation function, and generalized coordinates of each virtual node in the dissipative node system. The control laws for each UAV and the expected acceleration of each virtual node are determined based on the virtual Lagrange model. The operation of the drone corresponding to each virtual node is controlled according to the control law of each virtual node and the desired acceleration of the node.

[0006] In an optional embodiment of this application, the Lagrangian function of the dissipative node system is determined according to the following method: Determine the total kinetic energy and total potential energy of the UAV collaborative hoisting system; The Lagrangian function of the dissipative node system is determined based on the total kinetic energy and total potential energy of the UAV-coordinated hoisting system.

[0007] In one optional embodiment of this application, the total kinetic energy of the UAV collaborative hoisting system is the sum of the kinetic energies of all UAVs, the center point, and the load in the UAV collaborative hoisting system.

[0008] In one optional embodiment of this application, the total potential energy of the UAV collaborative hoisting system includes total gravitational potential energy and total elastic potential energy.

[0009] In an optional embodiment of this application, the total elastic potential energy of the UAV-coordinated lifting system is determined according to the following method: The adjacency coefficient between virtual nodes in the dissipative node system is determined based on the coordinates of each drone and load in the drone collaborative hoisting system and the preset sensing radius. The total elastic potential energy of the UAV collaborative hoisting system is determined based on the adjacency coefficient between each virtual node, the elastic coefficient of the virtual spring between each virtual node, and the spring length correction value.

[0010] In one optional embodiment of this application, the spring length correction value of the virtual spring is the difference between the current distance and the initial distance of each virtual node.

[0011] In an optional embodiment of this application, the dissipation function of the dissipation node system is determined according to the following method: Determine the relative motion dissipation function and the center rotation dissipation function of the dissipation node system; wherein, the relative motion dissipation function is a function used to characterize the energy dissipation caused by the relative motion between adjacent UAVs; the center rotation dissipation function is a function used to characterize the energy dissipation generated by the rotation of the UAVs around the center point of the UAV group; The dissipation function of the dissipation node system is determined based on the relative motion dissipation function and the central rotation dissipation function.

[0012] In an optional embodiment of this application, the relative motion dissipation function of the dissipation node system is determined according to the following method: The relative motion dissipation function of the dissipation node system is determined based on the spring length correction value of the virtual springs between each virtual node and the damping coefficient of the dissipation force between each UAV.

[0013] In an optional embodiment of this application, the center rotation dissipation function of the dissipation node system is determined according to the following method: The central rotational dissipation function of the dissipation node system is determined based on the generalized coordinates of each virtual node and the friction coefficient between the center point of the UAV group and the UAV.

[0014] In one optional embodiment of this application, determining the desired acceleration of each virtual node based on the virtual Lagrange model includes: For each virtual node, the node-expected acceleration of the UAV is determined based on the system-expected acceleration of the dissipative node system, the input control of the UAV in the virtual Lagrange model, and the mass of the UAV.

[0015] In one optional embodiment of this application, determining the control law for each UAV based on the virtual Lagrange model includes: The external force of the dissipative node system is set to 0, and the input of the dissipative node system is determined as the control input of the UAV. The height of each virtual node in the dissipative node system is configured to be equal. The virtual Lagrange model is then modified to obtain the control law of each UAV.

[0016] In an optional embodiment of this application, the generalized coordinates of each virtual node in the dissipative node system are determined according to the following method: The generalized coordinates of each virtual node in the dissipative node system are calculated based on the coordinates of each drone and load in the drone collaborative hoisting system and the preset height of each drone and load.

[0017] A second aspect of this application provides a drone collaborative lifting control device, applied to a drone collaborative lifting system, the drone collaborative lifting system comprising at least multiple drone groups and loads connected to each drone group, the device comprising: Construct a dissipative node system for the UAV collaborative hoisting system; wherein, each virtual node in the dissipative node system corresponds one-to-one with each UAV in the UAV collaborative hoisting system; A virtual Lagrange model of the dissipative node system is constructed based on the Lagrange function, dissipation function, and generalized coordinates of each virtual node in the dissipative node system. The control laws for each UAV and the expected acceleration of each virtual node are determined based on the virtual Lagrange model. The operation of the drone corresponding to each virtual node is controlled according to the control law of each virtual node and the desired acceleration of the node.

[0018] A third aspect of this application provides a computer device, including: a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the above methods.

[0019] A fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any of the preceding claims.

[0020] The UAV collaborative hoisting control method provided in this application is applied to a UAV collaborative hoisting system, which includes at least multiple UAV groups and loads connected to each UAV group. The method includes: constructing a dissipative node system for the UAV collaborative hoisting system; wherein each virtual node in the dissipative node system corresponds one-to-one with each UAV in the UAV collaborative hoisting system; constructing a virtual Lagrange model of the dissipative node system based on the Lagrange function, dissipation function, and generalized coordinates of each virtual node in the dissipative node system; determining the control law of each UAV and the desired acceleration of each virtual node based on the virtual Lagrange model; and controlling the operation of the UAV corresponding to each virtual node based on the control law and the desired acceleration of each virtual node.

[0021] Firstly, the damping term introduced through the dissipation function ensures that the system energy decreases strictly, allowing the formation to converge quickly and smoothly to the desired formation without oscillation, thus guaranteeing built-in stability and convergence. Secondly, each item (inertial force, elastic force, damping force) in the control law in the embodiments of this application has a clear physical meaning, making the adjustment of control parameters (stiffness, damping) intuitive and easy to debug, thus improving the physical interpretability and modularity of the system. Thirdly, the embodiments of this application employ a distributed control strategy to provide each UAV with a control law and acceleration. The control law of each UAV depends only on the position information of its neighboring nodes, without relying on centralized control or real-time communication between UAVs, thereby reducing system complexity and dependence on communication networks. In complex environments, such as high-altitude or communication-restricted scenarios, the method proposed in this invention can prevent failures caused by communication interruptions or insufficient computing resources. Furthermore, since it does not require a centralized processor, the system is easily scalable to large-scale UAV formations. In addition, it exhibits strong robustness and adaptability when facing situations such as UAV heterogeneity, increased UAV numbers, uncertain cable lengths, or UAV malfunctions.

[0022] In summary, the embodiments of this application provide a collaborative hoisting control method for unmanned aerial vehicle (UAV) swarms that requires no communication, has high robustness and scalability. Attached Figure Description

[0023] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a schematic diagram of the structure of the UAV collaborative hoisting system provided in the embodiments of this application; Figure 2 A flowchart illustrating a UAV collaborative hoisting control method provided in one embodiment of this application; Figure 3 This is a top view illustrating the connection between the corresponding virtual nodes of the drone and the payload; Figure 4 This is a side view illustrating the connection between the corresponding virtual nodes of the drone and the payload; Figure 5 This is a formation diagram of the coordinated hoisting system at key moments during actual flight experiments; Figure 6 It is a side view of the trajectory and formation of the collaborative hoisting system in actual flight experiments; Figure 7 It is a top view of the trajectory and formation of the collaborative hoisting system in actual flight experiments; Figure 8 This is a bird's-eye view comparison of the formations of dissipative node systems with different sensing ranges and formation-based systems; Figure 9 It includes the load distribution, speed, position along the x-axis, and changes in the lifting status of the UAV during the lifting process; Figure 10 It refers to the relative distance between UAVs in a dissipative node system under zero, medium, and high uncertainties. Figure 11 It is the internal distance of the UAV under zero, medium, and high uncertainties in the formation system; Figure 12 It refers to the relative distance of the drone under zero, medium, and high uncertainties in a system where the load is the leader; Figure 13 A comparison of cable tension bands for dissipative structures, formation structures, and load leader structures under uncertain conditions; Figure 14 These are the respective formation configurations for transporting 5, 10, 15, and 20 drones; Figure 15 This is a numerical diagram of the payload capacity of 10 drones; Figure 16 This is a payload distribution diagram of 10 drones with different capabilities; Figure 17 This is a schematic diagram of the structure of a drone collaborative hoisting control device provided in one embodiment of this application; Figure 18 This is a schematic diagram of a computer device structure provided in one embodiment of this application. Detailed Implementation

[0024] In the process of developing this application, the inventors discovered that there is an urgent need for a drone collaborative hoisting control method that requires no communication, has high robustness and scalability, in order to overcome the shortcomings of the prior art.

[0025] To address the aforementioned issues, this application provides a method, apparatus, computer equipment, and storage medium for collaborative hoisting control of unmanned aerial vehicles (UAVs).

[0026] The solutions in this application embodiment can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.

[0027] To make the technical solutions and advantages of the embodiments of this application clearer, the exemplary embodiments of this application will be described in further detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not an exhaustive list of all embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other.

[0028] The following is a brief description of the application environment of the UAV collaborative hoisting control method provided in the embodiments of this application: Please see Figure 1 The UAV collaborative lifting control method provided in this application is applied to a UAV collaborative lifting system. The UAV collaborative lifting system includes at least multiple UAV groups and loads connected to each UAV group. For example, one UAV group includes multiple UAVs. Figure 1 The drones 1, 2, and 3 in the diagram are connected to the same load via cables or similar means, and the load is lifted through coordinated control. The connection point between the drone and the load can be located at the center of mass of the drone group. It should be noted that in this embodiment, the elastic deformation of the cables is ignored, and the cables connected to the load are considered to be under stress.

[0029] Please see Figure 2 The UAV collaborative hoisting control method provided in this application includes the following steps 201-204: Step 201: Construct the dissipative node system of the UAV collaborative hoisting system; For example, each UAV is modeled as a dual-integral system, and the virtual nodes are also constructed using the same modeling method to build corresponding dissipative node systems. In this embodiment, the UAV collaborative lifting system refers to a physical system containing UAVs and a load. The dissipative node system refers to a virtual system formed by modeling the UAV collaborative lifting system based on dissipation theory. Each virtual node in the dissipative node system corresponds one-to-one with each UAV in the UAV collaborative lifting system. For example, UAVs 1, 2, and 3 correspond to virtual nodes 1, 2, and 3 respectively in the dissipative node system. The position, distance, and coupling relationship of each virtual node also correspond to the physical UAVs and the load.

[0030] Step 202: Construct a virtual Lagrangian model of the dissipative node system based on the Lagrangian function, dissipation function, and generalized coordinates of each virtual node in the dissipative node system. Among them, the Lagrange function Dissipation function is used to characterize the dissipation-free dynamics of dissipation node systems. The virtual Lagrangian model, used to characterize the energy dissipation of a dissipative node system, is a standard dissipative complete nonconservative system, where the conservative forces (derived from the Lagrangian function) are represented. ) and dissipative force (from dissipation function) This is unified into a single model, serving as a dynamic equation system including dissipative terms, to characterize the motion / evolution of a real-world UAV collaborative lifting system. Due to energy dissipation in the dissipative node system, the total energy of the virtual node system will gradually decrease.

[0031] In this embodiment, the dynamics of the UAV collaborative lifting system are characterized by a Lagrange model, which can be described, for example, by the following formula (1):

[0032] In formula (1), Indicates the first The coordinates of the drone. For the coordinates of the suspended load, Indicates to Perform differentiation. Describe the Lagrangian function of a dissipative nodal system. Indicates the first The control input for the drone.

[0033] Its corresponding dissipative node system is in equation Based on the above, the description is as follows:

[0034] formula For a dissipative nodal system, the Lagrange model is given, where, Indicates the first The first drone corresponding to the The generalized coordinates of a virtual node, Describe the Lagrangian function of a dissipative nodal system. This represents the dissipation function of a dissipation node system.

[0035] In one optional embodiment of this application, each drone is always connected to its payload via a cable. See also... Figure 3 and Figure 4 Because drones are Since the direction has a consistent flight altitude, the generalized coordinates of the virtual node can be defined as a two-dimensional planar position vector. with vertical height Combining them to form a vector can be expressed, for example, as shown in the following formula (3): (3) In formula (3), These represent the height of the virtual node and the height of the load, respectively. The first A virtual node in Coordinates in the direction, where Indicates the center of formation is Coordinates in direction Indicates the first A virtual node in Coordinates in direction ( ), P n+2 Indicates the load at Coordinates in the direction.

[0036] The calculation process can be simplified and the computational complexity reduced by using the expression in formula (3) above. The position of the UAV in formula (3) above is composed of its two-dimensional coordinates and plane height. The correspondence between the generalized coordinates of the virtual node and the simplified expression of the virtual node in formula (3) above is shown in formula (4) below. In practical applications, the expression of the virtual node can be flexibly converted through formula (4) below to adapt to different computational scenarios: (4) If the virtual coordinate expression of formula (3) above is adopted, the Lagrange model of the dissipative nodal system in formula (2) above is expressed as formula (5): (5) In formula (5), Indicates the center of formation is Coordinates in direction Indicates the first A virtual node in Coordinates in direction ( ), P n+2 Indicates the load at Coordinates in direction Describe the Lagrangian function of a dissipative nodal system. It is the system's dissipation function. Indicates the first The control input for the drone.

[0037] Step 203: Determine the control law of each UAV and the expected acceleration of each virtual node based on the virtual Lagrange model; The virtual Lagrange model in this embodiment includes the Lagrange function and dissipation function of the dissipation node, forming a systematic dynamic model that includes dissipation-free dynamics and system dissipation. It can configure a general control law for each UAV based on a distributed strategy, and then combine the node expected acceleration of each virtual node to achieve independent and cooperative control of each UAV.

[0038] Step 204: Control the operation of the UAV corresponding to each virtual node according to the control law of each virtual node and the desired acceleration of the node.

[0039] Each virtual node corresponds to a control law and a desired node acceleration. This embodiment maps the UAV collaborative lifting system to a virtual node system and employs a control law and desired node acceleration based on a Lagrange-dissipative framework. Firstly, the damping term introduced through the dissipation function ensures that the system energy decreases strictly, allowing the formation to converge quickly and smoothly to the desired formation without oscillation, thus guaranteeing built-in stability and convergence. Secondly, each item (inertial force, elastic force, damping force) in the control law in the embodiments of this application has a clear physical meaning, making the adjustment of control parameters (stiffness, damping) intuitive and easy to debug, thus improving the physical interpretability and modularity of the system. Thirdly, the embodiments of this application employ a distributed control strategy to provide each UAV with a control law and acceleration. The control law of each UAV depends only on the position information of its neighboring nodes, without relying on centralized control or real-time communication between UAVs, thereby reducing system complexity and dependence on communication networks. In complex environments, such as high-altitude or communication-restricted scenarios, the method proposed in this invention can prevent failures caused by communication interruptions or insufficient computing resources. Furthermore, since it does not require a centralized processor, the system is easily scalable to large-scale UAV formations. In addition, it exhibits strong robustness and adaptability when facing situations such as UAV heterogeneity, increased UAV numbers, uncertain cable lengths, or UAV malfunctions.

[0040] In summary, the embodiments of this application provide a collaborative hoisting control method for unmanned aerial vehicle (UAV) swarms that requires no communication, has high robustness and scalability.

[0041] In an optional embodiment of this application, the Lagrangian function of the dissipative node system described in step 202 above is determined according to the following method, which includes at least the following steps 301-302: Step 301: Determine the total kinetic energy and total potential energy of the UAV collaborative hoisting system; Step 302: Determine the Lagrangian function of the dissipative node system based on the total kinetic energy and total potential energy of the UAV collaborative hoisting system.

[0042] In the embodiments of this application, the Lagrangian function of the dissipative nodal system is composed of the difference between kinetic energy and potential energy: (6) In formula (6), Let T represent the Lagrangian function of the dissipative node system, T represent the total kinetic energy of the dissipative node system, and V represent the total potential energy. G V represents gravitational potential energy. E It represents elastic potential energy.

[0043] In one optional embodiment of this application, the total kinetic energy of the dissipative node system The total kinetic energy, including the aforementioned UAV collaborative hoisting system, is the sum of the kinetic energies of all UAVs, the center point, and the load within the UAV collaborative hoisting system. The sum of the total kinetic energy of each UAV node, center point, and load is given by the following formula (7): (7) In formula (7), T represents the total kinetic energy of the dissipative node system. Indicates the first The mass of the object (drone, etc.) corresponding to each virtual node. Indicates the first A virtual node in Coordinates in direction ( ), These represent the height of the virtual node and the height of the load, respectively.

[0044] In one optional embodiment of this application, the total potential energy of the UAV collaborative hoisting system includes total gravitational potential energy and total elastic potential energy.

[0045] The gravitational potential energy term can be calculated using the following formula (8): (8) In formula (8), V G Represents gravitational potential energy. Represents gravitational acceleration. Indicates the first The mass of the object (drone, etc.) corresponding to each virtual node. Indicates the quality of the load. These represent the height of the virtual node and the height of the load, respectively.

[0046] In an optional embodiment of this application, the total elastic potential energy of the UAV collaborative hoisting system described in step 202 above is determined according to the following method, which includes the following steps 401-402: Step 401: Determine the adjacency coefficient between virtual nodes in the dissipative node system based on the coordinates of each UAV and load in the UAV collaborative hoisting system and the preset sensing radius; Step 402: Determine the total elastic potential energy of the UAV collaborative hoisting system based on the adjacency coefficient between each virtual node, the elastic coefficient of the virtual spring between each virtual node, and the spring length correction value.

[0047] The spring length correction value of the virtual spring described above is the difference between the current distance and the initial distance of each virtual node. For example, the total elastic potential energy can be calculated using the following formula (9): (9) In formula (9), This represents the elastic coefficient of the virtual spring between virtual nodes. The adjacency coefficient between each virtual node is represented by the element at the corresponding position in the adjacency matrix. Other elastic potential energy representation methods can also be used, and this application embodiment does not limit this; the spring length correction value of the virtual spring. , is the current distance l between adjacent virtual nodes ij Initial distance l from adjacent virtual nodes ij0 The difference, the current distance between adjacent virtual nodes The definition of (i.e., the virtual spring length) varies depending on the case: (10) In this embodiment, the norm function is defined as the L2 norm by default. Correspondingly, the Lagrangian function of the dissipative node system in the above formula (6) is... The expanded form is as follows: (11) (11) This application embodiment introduces virtual springs between virtual nodes, utilizing the "flexible" characteristics of the virtual springs to give the system natural compliance when facing disturbances or sudden obstacles, avoiding rigid collisions and improving the safety of collaborative transportation.

[0048] In an optional embodiment of this application, the adjacency coefficients between the aforementioned virtual nodes... The adjacency matrix can be determined in the following way: Dissipative force f i The general description is as follows: (12) (12) In formula (12), f i Indicates the first The dissipation force of a drone Indicates the first The first drone corresponding to the The generalized coordinates of a virtual node, Indicates the first The coordinates of the drone. Indicates the first The coordinates of the drone. The coordinates are for the suspended load.

[0049] The dissipative force is generated through the sensing range of the UAV's onboard sensors, and can be specifically expressed by the following formula (13): (13) In formula (13), This indicates the sensing radius of the airborne sensor, which is the preset sensing radius in the embodiments of this application.

[0050] When drones move relative to each other, the total energy of the cooperative system is dissipated, and this dissipation is positively correlated with the intensity of the motion. The dissipation force f in the embodiments of this application... i It is designed as follows (14): (14) In formula (14), the function Used for energy dissipation between drones, For the Energy dissipation between individual drones and the suspended load is used to prevent the drone swarm from rotating.

[0051] One optional embodiment of this application provides a specific dissipative cooperative modeling method. Figure 1 In the coordinate system shown, the positions of the UAV node and the suspended load are represented by a three-dimensional vector. The connection relationships of the virtual node system are established through an adjacency matrix. and Laplace matrix Representation. Adjacency matrix Element w ij and Laplace matrix Define the following formula (15): (15) In formula (15), This represents the adjacency coefficient between virtual nodes. Indicates the first The coordinates of each virtual node.

[0052] In an optional embodiment of this application, step 202 above, the dissipation function of the dissipation node system is determined according to the following method, which includes the following steps 501-502: Step 501: Determine the relative motion dissipation function and the central rotation dissipation function of the dissipation node system; The relative motion dissipation function is a function used to characterize the energy dissipation caused by the relative motion between adjacent UAVs; the center rotation dissipation function is a function used to characterize the energy dissipation caused by the rotation of a UAV around the center point of the UAV group. Step 502: Determine the dissipation function of the dissipation node system based on the relative motion dissipation function and the central rotation dissipation function.

[0053] The dissipation function of the dissipation node system in the embodiments of this application This is the sum of energy losses caused by internal dissipative forces, i.e.:

[0054] In formula (16), This represents the relative motion dissipation function caused by the relative motion between adjacent drones, which helps to stabilize the formation structure of the drone swarm; This represents the central rotation dissipation function generated by rotating around the group's center, the purpose of which is to suppress this rotation in order to save the drone's energy.

[0055] In an optional embodiment of this application, the relative motion dissipation function of the dissipation node system described in step 501 above is determined according to the following method: The relative motion dissipation function of the dissipation node system is determined based on the spring length correction value of the virtual springs between each virtual node and the damping coefficient of the dissipation force between each UAV.

[0056] In an optional embodiment of this application, the center rotation dissipation function of the dissipation node system described in step 501 above is determined according to the following method: The central rotational dissipation function of the dissipation node system is determined based on the generalized coordinates of each virtual node and the friction coefficient between the center point of the UAV group and the UAV.

[0057]

[0058] in, The damping coefficient represents the dissipation force between drones. This represents the coefficient of friction between the group center and the drone; This represents the relative motion dissipation function caused by the relative motion between adjacent UAVs; This represents the central rotation dissipation function generated by rotating around the center of the group.

[0059] Combining with the other formulas mentioned above, The expanded form is as follows (18):

[0060] In an optional embodiment of this application, in conjunction with the above embodiments, after obtaining the specific expressions of the Lagrange function and the dissipation function, they can be substituted into the above formula (5) to derive the Lagrange model (i.e., the dynamic model) of the dissipative node system. It is worth noting that the center of the UAV swarm can be a specific UAV, thereby forming a "leader-follower" structure. For generalized coordinates, the Lagrange model of the dissipative node system is as follows: formula (19):

[0061] In summary, to simplify the expression and facilitate calculation, the Lagrange model of dissipative nodal systems can be unified and integrated into the following formula (20):

[0062] In formula (20), express The second derivative of corresponds to the first derivative in formula (5) above. A virtual node in Coordinates in the direction; The term represents the gravitational function, corresponding to the gravitational potential energy V in the Lagrangian function of the dissipative node system in formula (6) above. G , The elastic force function term corresponds to the elastic potential energy V in the Lagrangian function of the dissipative node system in the above formula (6). E , The dissipation force term includes damping force and friction force, corresponding to the dissipation function in formula (5) above. ; This represents the external control force acting on the dissipative node system, corresponding to the control input of the UAV in formula (5) above. .

[0063] In an optional embodiment of this application, step 204 above, which determines the expected acceleration of each virtual node based on the virtual Lagrange model, includes the following steps: For each virtual node, the node-expected acceleration of the UAV is determined based on the system-expected acceleration of the dissipative node system, the input control of the UAV in the virtual Lagrange model, and the mass of the UAV.

[0064] For example, the The desired acceleration command for the drone is: (twenty one) In formula (21), This represents the expected acceleration of the entire dissipative node system. Indicates the first The mass of the object (drone, etc.) corresponding to each virtual node. Represents gravitational acceleration. Indicates the first The control input for a drone It can be calculated based on the Lagrange model of the dissipative node system in formula (5), or directly configured.

[0065] In an optional embodiment of this application, step 204 above, which determines the control law of each UAV based on the virtual Lagrange model, includes: The dissipative controller is designed based on a dissipative node system. Each UAV employs a control law to simultaneously manage payload transport and formation maintenance. The dissipative-based control law can be expressed, for example, as formula (22): (twenty two) In formula (22), Indicates the first The control input for the drone. Formula (22) contains... The term corresponds to the elastic force term and specifically includes two types of spring forces: one is the spring force of the virtual spring between drones, used to build a cooperative transport formation; the other is the spring force of the virtual spring between the drone and the load, used to achieve the suspension function. and They are respectively with the first The and the first The stiffness and damping coefficients related to each UAV. The spring force and damping force between UAVs are actually the mapping of the spring force and damping force between virtual nodes. Since they are all in the horizontal direction and their magnitudes differ by only a proportional coefficient, the formula (22) here is directly written in the form mapped to the UAVs. and They are respectively with the first The stiffness and damping coefficients related to the drone and its suspended load. These are the elements at the corresponding positions in the adjacency matrix, i.e., the adjacency coefficients, used to describe the adjacency of the first element. The and the first The visibility of a drone. The subscript "hor" indicates the horizontal component of the vector. It is the first The and the first The distance between the drones, and yes The initial value.

[0066] In an optional embodiment of this application, step 203 above, determining the control law for each UAV based on the virtual Lagrange model, includes the following steps: The external force of the dissipative node system is set to 0, and the input term of the dissipative node system is determined as the control input of the UAV. The height of each virtual node in the dissipative node system is configured to be equal. The virtual Lagrange model is then modified to obtain the control law of each UAV. Regarding the above formula (19), when the external force of the dissipative node system is 0, then: At this point, if we want the dissipative node system to satisfy the characteristics of a dissipative system, then: Regarding the above formula, Set as the control input for the i-th drone, i.e. Then we have: Break down the above expression and exclude... Situation:

[0067] because Since the height components of each virtual node are consistent, and considering the actual position and velocity status acquisition and control command execution of the UAV, the above formula (22) can be obtained by improving the above formula.

[0068] In an optional embodiment of this application, the generalized coordinates of each virtual node in the dissipative node system described in step 202 above are determined according to the following method: The generalized coordinates of each virtual node in the dissipative node system are calculated based on the coordinates of each drone and load in the drone collaborative hoisting system and the preset height of each drone and load.

[0069] That is, the above formula The Middle The first drone corresponding to the Generalized coordinates of virtual nodes It can be calculated using the following formula (23): (twenty three) In formula (23), Indicates the first The first drone corresponding to the The generalized coordinates of a virtual node, The coordinates representing the suspended load; , and These represent the preset altitude of the drone, the first... The preset height of each drone and the height of the suspended load; Indicates the first The coordinates of the drone. The coordinates are for the suspended load.

[0070] In one optional embodiment of this application, to more clearly illustrate the UAV collaborative lifting control method provided in this application embodiment, an experiment in a real-world scenario involving a group of five UAVs collaboratively lifting a 5kg load is used as an example to further explain the specific implementation method. The experiment is conducted in an outdoor environment, with each UAV having a load capacity of 2kg. Each UAV can communicate via a local area network to obtain each other's pose and status information.

[0071] The specific operations of this application embodiment are as follows: Please see Figure 1 The drone collaborative lifting system is described as a system consisting of five drones and a suspended load. The dynamics of this drone collaborative lifting system are described by a Lagrange model as follows:

[0072] in, Indicates the first The coordinates of the drone. For the coordinates of the suspended load, Indicates to Perform differentiation; It is the Lagrangian function of the collaborative drone lifting system; It is the first The control input for each drone. Its corresponding virtual node system is in the formula... Based on the above, the description is as follows:

[0073] in, Indicates the first The first drone corresponding to the The generalized coordinates of a virtual node are calculated using the following formula:

[0074] in, These are the coordinates of the suspended load. , and Representing the preset height, the first The height of the drone and the height of the suspended load. The generalized coordinates of the virtual node are defined as two-dimensional planar position vectors. with vertical height Combining to form vectors:

[0075] in, These represent the height of the virtual node and the height of the load, respectively. The first A virtual node in Coordinates in the direction.

[0076] The position of a virtual node is determined by its two-dimensional coordinates and planar height:

[0077] in, It is the Lagrangian function of the dissipative nodal system. It is the system's dissipation function. It is the first The control input for a drone. The Lagrangian function of a dissipative nodal system consists of the difference between kinetic and potential energy:

[0078] Total kinetic energy of dissipative node system include The sum of the kinetic energy of each drone node, its central point, and its payload:

[0079] in, Indicates the first The mass of the object corresponding to each virtual node.

[0080] The gravitational potential energy term is:

[0081] in, Represents gravitational acceleration. This indicates the quality of the load.

[0082] The elastic potential energy is:

[0083] in, It is the difference between the current distance and the initial distance. It is the spring constant of the virtual spring. It is the element at the corresponding position in the adjacency matrix. The definition depends on the situation:

[0084] The expanded form is as follows:

[0085] Dissipation function of a dissipation node system Defined as the sum of energy losses caused by internal dissipative forces. The expanded form is:

[0086] For generalized coordinates, the dynamic model is as follows:

[0087] The control law based on dissipation is designed as follows:

[0088] No. The desired acceleration command for the drone is:

[0089] in, The expected acceleration of the entire dissipative node system.

[0090] The values ​​for the control parameters and physical parameters are as follows:

[0091] Substitute into the formula and In this process, the desired acceleration command for each drone in the actual experiment is obtained:

[0092]

[0093] In one optional embodiment of this application, code is written on a drone and actual flight tests are conducted to verify the effectiveness and reliability of the method. Specific experimental results are as follows: Figure 4 , Figure 5 , Figure 6 As shown.

[0094] like Figure 4 This is a formation diagram of the coordinated hoisting system at key moments during actual flight experiments. Figure 4 As can be seen, five drones can stably lift the load in a coordinated manner and then transport it to the designated building. Throughout the process, the formation of the coordinated lifting system consisting of drones, ropes, and load remains basically unchanged.

[0095] Figure 5 It is a side view of the trajectory and formation of the collaborative hoisting system during actual flight experiments. Figure 6 This is a top-down view of the trajectory and formation of the collaborative hoisting system during an actual flight experiment. The large dot below the collaborative hoisting system represents the load; the smaller dots in the middle of the system represent the UAVs; the smaller dots at the top represent virtual nodes; the solid line between the UAV and the load represents the rope; and the dashed line between the UAV and the virtual nodes represents virtual spring dampers. Figure 5 and Figure 6 As can be seen, the load was lifted to a height of 15m by the collaborative hoisting system, flew about 31m in the y direction, and about 3m in the x direction. After reaching the desired position, the load dropped to a height of about 7m. Throughout the process, the formation remained stable, and the movement trajectories of the drone and the load were relatively smooth without significant shaking.

[0096] In one optional embodiment of this application, the ability of the dissipative node system to cope with some emergencies was verified, including the addition or failure of a single drone, inconsistent cable lengths, inconsistent load capacity, and changes in the number of drones.

[0097] First, consider the scenario of a single drone joining or failing. In actual flight, if one drone moves out of range, the total dissipative forces will not suddenly change and disrupt the formation. Based on this stability characteristic of the dissipative node system, the drone swarm will move towards a new equilibrium state according to the newly formed connections. The interaction forces within the dissipative node system of a particular drone are generated by all its visible drones; therefore, considering the perception limitations in reality is crucial, as changes in drones within the visible range will affect overall transport performance. The dissipative node system exhibits stable performance in terms of payload altitude and formation stability, with a maximum variation of 0.5 meters. Figure 7 As shown. In contrast, the formation-based method is susceptible to perception limitations, resulting in altitude variations of up to 2 meters and horizontal trajectory deviations of 1.5 meters. To test the system's recovery capability before and after a failure, five drones were pre-set to transport a load along the x-axis. At t = 5 seconds, two drones failed and detached from the transport cluster. Subsequently, at t = 10 seconds, one drone was added back in.

[0098] The load distribution in a dissipative transport system and the speed variation of each UAV, such as Figure 8 As shown in the diagram, at t = 5 seconds, a slight change in the vertical speed of the drones was observed, and the system quickly recovered, reforming a stable formation within 1 second. After the failure, the load distribution changed abruptly and automatically shifted to the remaining three drones, with the newly formed formation exhibiting uniformity among the three drones. At t = 10 seconds, the most recently introduced drone began participating in the transport process, and a slight change in the drone's vertical speed was observed, with the load being redistributed to the four drones. The newly formed formation again exhibited uniformity among the four drones. Therefore, it can be concluded that when drones are lost or added to the drone swarm, the remaining drones autonomously adjust their distribution and reorganize into a new formation, ensuring uniform distribution and stable transport of the cable load, achieving real-time plug-and-play functionality, and ensuring the robustness and adaptability of the system.

[0099] Secondly, there is the issue of inconsistent cable lengths. In collaborative hoisting, due to the inherent inaccuracies of sensor measurements, obtaining precise cable length information can be difficult. Therefore, the transportation mechanism must be designed to withstand the potential uncertainty in cable length information. When cable lengths are inconsistent, the UAV in this invention can autonomously adjust its altitude and orientation to maintain cable tension and achieve a balanced tension distribution. In uncertainty analysis, the cable length can be expressed as:

[0100] in, The actual length of the suspension cable is added to the uncertainty of the cable length.

[0101] like Figure 9 , Figure 10 and Figure 11 As shown, when there is uncertainty in the cable length, the adaptability of three solutions was compared through numerical simulation. When the cable length is disturbed, the formation and suspension force of the UAVs will be affected accordingly. The tracking error in collaborative hoisting is defined as...

[0102] in It is the total number of simulation steps. It is the tracking error between the load trajectory and the desired trajectory. This is the time interval between every two simulation steps, here 0.001s. If the cable length measurement is disturbed, the drone's formation configuration and the suspension force it experiences will be affected. Figure 9 , Figure 10 and Figure 11 In this context, the internal distance between drones is described by an average value and the range between the maximum and minimum values. Figure 9 In the simulation, the average internal distance between the drones stabilized at 1.2 meters, 1.5 meters, and 2.2 meters within 2 seconds. This variation in average internal distance determines the tightness or looseness of the drone formation. As uncertainty increases, the drones tend to form looser formations to ensure safe transport without adjusting other parameters. Meanwhile, the distance filling area between the maximum and minimum values ​​changes little, and the overall formation shape remains stable under all three uncertainty conditions, indicating that the system is robust in shape maintenance.

[0103] exist Figure 10 and Figure 11 In the process, the average internal distance between drones oscillated continuously when there was uncertainty, and the formation was significantly affected by the inaccurate cable length. Figure 12 The effect of uncertainty on the cable tension magnitude in three methods is demonstrated. Unexpected oscillations in cable tension directly affect the stability of the load. Figure 12The force band in the system is composed of the maximum and minimum tension values ​​of the UAVs. In all three cases, the cable tension in the dissipative node system exhibits a stable trend throughout the transportation process, while the tension in the formation-based and centralized systems oscillates in the initial stage. Therefore, the method proposed in this invention can ensure a smooth tension transition even when cable length uncertainties exist. Although external wind interference can cause fluctuations and slight differences in the average tension among the four UAVs, the overall distribution remains relatively uniform.

[0104] Regarding the changing size of drone swarms: the drone swarm size in the dissipative node system is scalable. In traditional transportation algorithms, adjusting the size of the drone swarm usually requires redesigning the formation structure to adapt to different load transportation needs, and may even require modifying the algorithm framework, relying on a fixed-size drone swarm and its configuration. However, the collaborative lifting system based on dissipative theory proposed in this invention adopts a decentralized design, which can flexibly adapt to different numbers of drones without adjusting the overall algorithm framework. Figure 13 As shown, the formation of the system depends primarily on the density of the drones. Initially, the drones are randomly distributed within a specified area, and the target locations are known. For example... Figure 14 As shown, when the cluster contains 5, 10, 15, or 20 drones, the system tends to maintain a polygonal frame. As the number of drones participating in flight increases, the initial polygonal formation generates a multi-layered internal structure to achieve reasonable load distribution. In this dissipative node system, drones can spontaneously complete transportation tasks relying solely on neighbor information, and the cluster formation dynamically adjusts to maintain an appropriate density, thus demonstrating scalability for heterogeneous drones. This characteristic significantly improves the stability of the overall transportation process and exhibits high tolerance for new drones, consistent with the phenomenon that "a table can automatically identify four or more reaction forces on the table legs based on the weight of the book," effectively solving the trade-off between flexibility and robustness in traditional methods.

[0105] Finally, there's the situation where drone payload capacities are inconsistent. In this case, when there are inconsistencies in payload capacity among drones in a swarm, the dissipative node system can adaptively adjust according to the payload capacity of each drone. It can also adapt to changes in the payload capacity of the same drone under different operating conditions, such as fluctuations in battery weight or voltage, maintaining stable transport performance and load balance, ensuring the efficient completion of the overall transport mission. For example... Figure 15 As shown, when the UAV's payload capacity exhibits a heterogeneous distribution with μ=16.7 and σ=6.2, the results are as follows: Figure 15 As shown. Therefore, solutions based on dissipation theory can achieve smooth and reasonable load distribution, thereby ensuring the stability of the transportation process. Figure 16As shown, the formation and payload trajectory remained relatively stable throughout the flight, allowing each drone to fully utilize its payload capacity.

[0106] If the drone's flight area is a narrow area, the drone can be controlled to perform coordinated lifting operations through the following steps 201a-205a: Step 201a: Construct a virtual pipeline model for the UAV collaborative hoisting system; Each virtual node in the virtual pipeline model corresponds one-to-one with each drone in the drone collaborative hoisting system; Step 202a: Determine the longitudinal control, lateral control, and desired speed between two adjacent virtual nodes in the virtual pipeline model; The longitudinal control is the control of the forward direction, and the lateral control is the control of the flight boundary.

[0107] Step 203a: Construct a comprehensive speed control term for the virtual pipeline model based on the longitudinal control, the lateral control, and the desired speed; Step 204a: Map the integrated speed control term to the UAV collaborative hoisting system to determine the underlying force control term of the UAV collaborative hoisting system; Step 205a: Control the operation of each UAV in the UAV collaborative hoisting system based on the underlying force control item.

[0108] The integrated speed control term in this embodiment includes only the longitudinal control of each virtual node, the lateral control, and the desired speed. This means that during drone flight, only the flight position of the drone corresponding to each virtual node is needed. The drone position can be obtained in real time through position sensing devices such as GPS and radar. Therefore, there is no need for stable and efficient real-time communication between drones, thus reducing system complexity and dependence on communication networks. Especially in complex environments, there is no need to consider rope length, enabling support for unequal rope lengths. Only the position of each drone needs to be determined; speed and other states are not required, and communication support between drones is unnecessary. Position awareness alone is sufficient, providing anti-interference recovery capabilities, convergence, and robustness, making it suitable for real-world outdoor tasks and obstacle avoidance transportation in narrow passages. In short, this embodiment provides a drone collaborative lifting method for confined spaces.

[0109] In an optional embodiment of this application, step 201a above, the construction of the virtual pipeline model of the UAV collaborative hoisting system, includes the following steps 301a-302a: Step 301a: Construct the UAV dynamics model, load dynamics model, and rope tension model in the UAV collaborative hoisting system; As above Figure 1 The UAV collaborative lifting system, for example, consists of n UAVs and a load connected by ropes. The dynamic model of each UAV is as follows:

[0110] The load dynamics model is as follows:

[0111] in, Indicates the first The position vector of the drone, Indicates the first The velocity vector of the drone For the first The quality of the drone Indicates the first The thrust control input for the drone is, for example, the thrust provided by the drone's propellers. Indicates the first The gravity of the drone Indicates the rope to the first The pulling force of the drone Represents the location vector of the load. Represents the velocity vector of the load. Indicates the load weight. This represents the resultant force of all ropes on the load: .

[0112] The rope tension model is as follows:

[0113] Among them: A i Indicates the first Estimation of tension caused by minute stretching of ropes using a drone ; Indicates the current rope length. Represents the location vector of the load. Indicates the first The position vector of the drone; Indicates the initial length of the rope; It is a unit direction vector; Indicates the first The maximum permissible tension of the rope connecting the drone; , These represent the elastic coefficient and damping coefficient of the rope, respectively.

[0114] Step 302a: Construct the virtual pipeline model based on the UAV dynamics model, the load dynamics model, and the rope tension model.

[0115] In an optional embodiment of this application, the virtual pipeline model is a quadruple comprising: the starting cross-section of the virtual pipeline, the ending cross-section, the differential homeomorphic mapping relationship between the starting and ending cross-sections, and the path generation function. For example, an embodiment of this application defines a virtual pipeline. The quadruple is as follows:

[0116] in, These represent the starting and ending cross sections of the virtual pipeline, respectively, and are convex sets located in two-dimensional or three-dimensional space. This represents the differential homeomorphic mapping relationship between the initial cross section and the final cross section. ; The path generation function is represented as follows: ; It is a set of pairs of points.

[0117] Path curves can be generated based on the path generation function h. Path curve The centerline of the virtual pipeline, The current position of any drone inside the pipeline is The nearest point is: Its tangent vector is: ; It is a differential operator.

[0118] In an optional embodiment of this application, the above-described UAV collaborative hoisting control method based on virtual pipelines further includes the following steps: Determine the mapping relationship between the UAV collaborative hoisting system and the virtual pipeline model.

[0119] That is, the drone, load, and rope are mapped to virtual nodes in the virtual pipeline model, and then the set of relationships between all drones, loads, ropes and their corresponding virtual nodes is determined as the mapping relationship between the drone collaborative hoisting system and the virtual pipeline model, so as to facilitate the subsequent determination and mapping of parameters of the actual drone.

[0120] In an optional embodiment of this application, the step of determining the mapping relationship between the UAV collaborative hoisting system and the virtual pipeline model includes the following steps 401a-402a: Step 401a: Determine the load height of the load in the UAV collaborative hoisting system, as well as the virtual pallet height, the coordinates of the virtual pallet center point, and the load virtual node coordinates of the virtual load node in the virtual pipeline model; Step 402a: Calculate the virtual drone coordinates in the virtual pipeline model based on the load height, the virtual pallet height, the coordinates of the virtual pallet center point, and the coordinates of the load virtual node.

[0121] For example, the virtual node coordinates of the drone The result was obtained from the geometric mapping between the physical drone and the payload:

[0122] in: Indicates the first The virtual node coordinates of the drone. Represents the location vector of the load. Indicates the first The position vector of the drone; Indicates the virtual tray height; z i Indicates the first The height of the drone; z l Indicates the height of the load; and These represent the coordinates of the virtual tray center point and the coordinates of the load virtual node, respectively.

[0123] In an optional embodiment of this application, step 202a above, determining the longitudinal control, lateral control, and desired speed between two adjacent virtual nodes in the virtual pipeline model, includes the following steps 501a-502a: Step 501a: Construct the virtual spring control model, which is used to characterize the spring relationship between any two virtual nodes; Step 502a: Calculate the desired speed between two adjacent virtual nodes based on the virtual spring stiffness coefficient, connection coefficient, initial and current length of the virtual spring, and coordinates of the virtual drone node in the virtual pipeline model.

[0124] The desired velocity between any pair of virtual nodes is given by the spring relationship:

[0125] Where k2 represents the virtual spring constant, w ij Represents the connection coefficient w ij =1 indicates connection, w ij =0 indicates no connection, l ij0 The initial length of the virtual spring, l ij This represents the current length of the virtual spring, which is the distance between the two virtual nodes.

[0126] To further verify the reliability of the virtual spring control, in an optional embodiment of this application, its stability is demonstrated using the Lyapunov function:

[0127] For the i-th virtual node, the desired speed v controlled by its virtual spring is... c,i,d for:

[0128] In one optional embodiment of this application, the virtual pipeline control includes three components: lateral control, longitudinal control, and virtual spring control. Step 202a, determining the longitudinal control, lateral control, and desired speed between adjacent virtual nodes in the virtual pipeline model, includes: Construct a vertical control model; For each virtual drone node, the longitudinal control of each virtual node in the virtual pipeline model is determined based on the path distance along the virtual path in the virtual pipeline from the current position of the virtual drone node, the unit direction control gain, and the maximum saturation limit.

[0129] The path distance of the current position of the virtual drone node along the virtual path in the virtual pipeline refers to the distance between the first position and the second position. The first position refers to the position of the virtual pipeline generation line corresponding to the current position, and the second position refers to the end of the virtual pipeline generation line.

[0130] Longitudinal control is used to control the forward direction and can be determined by the following formula:

[0131] Among them, v lid This indicates longitudinal control, where k1 represents the adjustable parameter for longitudinal control, which can be flexibly configured according to actual conditions. Indicates the current position q of the virtual drone node i The path distance along the virtual path in the virtual pipeline; Indicates the current position q of the virtual drone node i Unit directional control gain, t c (q i ) represents the current position q of the virtual drone node. i The unit tangent vector at that point; This is the maximum speed saturation limit.

[0132] In an optional embodiment of this application, step 202a, determining the longitudinal control, lateral control, and desired speed between two adjacent virtual nodes in the virtual pipeline model, includes: Construct a lateral control model; For each virtual drone node, the lateral control of each virtual node in the virtual pipeline model is determined based on the lateral control model, the tangent vector of the virtual drone node's flight direction, the flight boundary guidance term, and the identity matrix.

[0133] Lateral control is used for boundary preservation. The lateral control in this embodiment can be calculated using the following formula:

[0134] Among them, v t,i,d Indicates longitudinal control. The term is the guiding term for the horizontal boundary, which can be derived from Lyapunov-like functions, where I2 represents the identity matrix, and t c This represents the tangent vector.

[0135] In an optional embodiment of this application, step 203a, the construction of the integrated speed control term of the virtual pipeline model based on the longitudinal control, the lateral control, and the desired speed, includes:

[0136] Among them, V i,d v represents the comprehensive speed control term of the virtual pipeline model. l,i,d Indicates longitudinal control, v c,i,d v represents the desired velocity controlled by the virtual spring. t,i,d This indicates vertical control.

[0137] In an optional embodiment of this application, step 204a above, which involves mapping the integrated speed control term to the UAV collaborative lifting system and determining the underlying force control term of the UAV collaborative lifting system, includes: Based on the virtual speed mapping model, the comprehensive speed control term is mapped to the UAV collaborative hoisting system to obtain the comprehensive control speed of the UAV collaborative hoisting system; Based on the speed error of the integrated control speed, the mass of the UAV, the proportional gain matrix, the integral gain matrix and the differential gain matrix, and the rope tension, the underlying force control terms of the UAV collaborative hoisting system are determined.

[0138] For example, the virtual velocity mapping relationship is as follows:

[0139] The saturation limit is:

[0140] The saturation limit is the maximum speed, and the minimum speed is 0.

[0141] Correspondingly, the underlying force control is as follows:

[0142] in: For speed error; The rope tension is estimated by the compensation function observer; For the proportional gain matrix, integral gain matrix, and differential gain matrix, For the first The gravity of the drone.

[0143] In one optional embodiment of this application, if the drones enter some extremely wide flat areas, the distance between the drones increases significantly, and communication between the drones is prone to instability or even interruption. In this case, the drones can be controlled to perform coordinated lifting operations through the following steps 201b-205b: Step 201b: Construct the velocity tracking controller of the virtual passive system based on the pre-built virtual dynamics model of the virtual passive system; In this system, each virtual node in the virtual passive system corresponds one-to-one with each UAV in the UAV collaborative hoisting system; the speed tracker includes at least: a feedforward term for matching the desired acceleration and a feedback term for suppressing the tracking error; in an optional embodiment of this application, the feedforward term is composed of the product of the UAV mass and the first derivative of the preset desired speed; and / or, the feedback term is composed of the speed tracking error and multiple integral coefficients.

[0144] Step 202b: During the flight of the UAV, the virtual control input of the virtual passive system is determined based on the virtual desired speed tracked by the speed tracking controller. The virtual expected speed refers to a unified expected speed set for the entire virtual passive system or UAV collaborative hoisting system, and is not a separate expected speed set for different UAVs.

[0145] Step 203b: Determine the node acceleration of each virtual node in the virtual passive system under the virtual control input; Step 204b: Map the acceleration of each node to the UAV collaborative hoisting system, and determine the expected thrust of each UAV based on the acceleration of each node and the UAV dynamics model of the UAV collaborative hoisting system; Step 205b: Control the flight of each UAV based on the expected thrust of each UAV.

[0146] The UAV collaborative hoisting control method based on a passive system provided in this application embodiment constructs a speed tracking controller for the virtual passive system according to a pre-built virtual dynamics model of the virtual passive system. During UAV flight, the speed tracking controller tracks a set virtual desired speed to determine the virtual control input of the virtual passive system, and determines the nodal acceleration of each virtual node in the virtual passive system under the virtual control input. Throughout the process, only speed tracking based on the speed tracking controller is needed to determine the desired thrust of each UAV, without relying on centralized control or real-time communication between UAVs, thus reducing system complexity and dependence on communication networks. Especially in complex environments, it does not need to consider rope length, supporting situations with unequal rope lengths. Only the position and speed of each UAV and other states need to be determined, without mutual communication support; it can be achieved solely through position awareness. It has anti-interference recovery capabilities, convergence, and robustness, and is suitable for real outdoor tasks and speed tracking tasks. In other words, this application embodiment provides a UAV collaborative hoisting method for a wide flight surface.

[0147] In an optional embodiment of this application, before constructing the speed tracking controller of the virtual passive system based on the pre-built virtual dynamics model of the virtual passive system in step 201 above, the UAV collaborative lifting system can be modeled first to determine the dynamics model of the UAV collaborative lifting system. This dynamics model includes at least: the dynamics model of the UAVs, the dynamics model of the load, and the dynamics model of the cable. For example, the UAV collaborative lifting system comprises n UAVs and a load connected by a cable. The dynamics modeling of each UAV is as follows:

[0148] The dynamic model of the load is as follows:

[0149] in, For the first The position vector of the drone, For the first The velocity vector of the drone For the first The quality of the drone For the first Thrust control input for unmanned aerial vehicles (UAVs) For the first The gravity of the drone For the cable to the first The pulling force of the drone The location vector of the load. The velocity vector of the load. To bear the weight, The resultant force of all cables on the load: .

[0150] The dynamic model of the cable includes the cable tension model, which can be:

[0151] in, For the cable to the first The pulling force of the drone Estimation of tension caused by minute stretching and contraction of the cable; This is the current cable length; This is the initial length of the cable; It is a unit direction vector; This is the maximum allowable tension of the cable; , These are the cable's elastic coefficient and damping coefficient, respectively.

[0152] In an optional embodiment of this application, after constructing the UAV collaborative hoisting system model and the dynamic model of the UAV collaborative hoisting system, the above-mentioned virtual passive system can be constructed in the following manner: The virtual passive system in this embodiment can be constructed using virtual nodes, virtual trays, virtual springs, and virtual dampers to achieve distributed control. Wherein: The coordinates of the center of the virtual tray are defined as follows: The virtual node corresponding to the load is The virtual passive system and the UAV collaborative hoisting system satisfy the following relationship:

[0153] in, For the first The coordinates of the virtual node corresponding to the drone, where p1 is the position vector of the payload. For the first The position vector of the drone, z u z is the height of the virtual tray, z1 is the height of the load, z i For the first The altitude of the drone.

[0154] In the virtual passive system, the virtual tray is modeled as a sufficiently large but finite disk. All virtual nodes corresponding to the drone are located on this virtual tray; that is, the drone's virtual nodes share the same height as the center of the virtual tray, but are allowed to slide on the tray surface. All virtual nodes are interconnected via virtual springs, and each virtual node is also connected to the center of the virtual tray via a virtual spring.

[0155] In an optional embodiment of this application, the virtual dynamics model includes: an elastic force term characterizing the elastic force induced by the virtual spring, a damping force term characterizing the damping force induced by the virtual damper, and a frictional force term characterizing the frictional force between the virtual node and the moving medium. After constructing the virtual passive system through the above embodiments, the virtual dynamics model of the virtual passive system can be constructed in the following manner: In an optional embodiment of this application, before step 201b above, and before constructing the speed tracking controller of the virtual passive system based on the pre-built virtual dynamics model of the virtual passive system, the above-mentioned UAV cooperative hoisting control method based on the passive system includes the following steps 301b-302b: Step 301b: Construct the virtual passive system based on the UAV collaborative hoisting system; the virtual passive system includes: virtual nodes corresponding to the UAV and the load, a virtual tray corresponding to the UAV flight field, a virtual spring for characterizing the connection relationship between each virtual node, and a virtual damper for characterizing the energy dissipation between each virtual node; Step 302b: Construct the virtual dynamic model of the virtual passive system based on the Lagrange method.

[0156] In an optional embodiment of this application, step 302b above, the construction of the virtual dynamic model of the virtual passive system based on the Lagrange method, includes the following steps 401b-402b: Step 401b: Determine the virtual dissipation function and virtual Lagrange function of the virtual passive system; Step 402b: Construct the virtual dynamic model of the virtual passive system based on the virtual dissipation function, the virtual Lagrange function, and the generalized coordinates of each virtual node in the virtual passive system.

[0157] The dynamics of this virtual passive system are described by the following Lagrange equations:

[0158] in, The Lagrangian function represents the virtual passive system. Represents the dissipation function. This represents the external driving force acting on the intermediate virtual system. This represents the virtual node corresponding to the load. This represents a virtual tray icon. The Lagrange function is expressed as:

[0159] in, Indicates kinetic energy. Represents gravitational potential energy. This represents the elastic potential energy stored in the virtual spring. Kinetic energy. It is the sum of the kinetic energy of all virtual nodes and loads:

[0160] in, This represents the quality associated with each virtual node, for example, the quality of the first virtual node. The quality of the drones, among other things. For the quality of the load. It should be explained that the cumulative limit here is set to... Because it includes virtual tray nodes. and virtual nodes corresponding to the load ,as well as Each virtual node corresponds to a drone.

[0161] gravitational potential energy Determined according to the following formula:

[0162] in, It represents the acceleration due to gravity.

[0163] elastic potential energy The value is determined according to the following formula:

[0164] in, , indicating the first The and the first The elastic coefficient of the virtual spring between virtual nodes, It is a connection matrix, in which, Indicates the first The and the first The virtual nodes are connected by springs. Example as follows:

[0165] variable Represents a node and The current length of the virtual spring between them, and This is the corresponding initial length. The current length is determined according to the following formula:

[0166] At the same time, the initial spring length It can be based on the desired formation of the UAV-cable-load system or on the pre-takeoff position of the UAV. Set in conjunction with other geometric constraints (such as cable length):

[0167] Dissipation function This represents the total energy loss due to internal dissipative forces (which are inherently non-conservative). Any force that causes energy dissipation during the evolution of a system can be classified as a dissipative force. The dissipative function can be obtained by integrating the generalized velocity with the corresponding dissipative force.

[0168] In an optional embodiment of this application, the virtual dissipation function includes: a virtual damping function characterizing energy dissipation between two virtual nodes due to the virtual damper; and a virtual friction function characterizing energy dissipation caused by friction between each virtual node and the moving medium. As shown in the following embodiment of this application, the dissipation function... It consists of two parts:

[0169] in, This is a virtual damping function, used to characterize the energy dissipation caused by the virtual damper between virtual nodes. This is a virtual friction function used to characterize the energy loss caused by friction between the virtual node and its moving medium.

[0170] In an optional embodiment of this application, before step 501b above, and before constructing the initial dynamic model of the virtual passive system based on the virtual dissipation function, the virtual Lagrangian function, and the coordinates of each virtual node and the external driving force of the virtual passive system, the method further includes the following steps 601b-602b: Step 601b: Construct the virtual damping function based on the elastic length variation term of the virtual spring between the two virtual nodes and the damping coefficient of the virtual damper between the two virtual nodes; Step 602b: Construct the virtual friction function based on the node coordinates of each virtual node and the friction coefficient between each virtual node and the moving medium.

[0171] The formulas for the virtual damping function and the virtual friction function are as follows:

[0172] in, The damping coefficient of the virtual damper. The coefficient of friction between the virtual node and the moving medium around it.

[0173] In an optional embodiment of this application, step 402b above, the construction of the virtual dynamic model of the virtual passive system based on the virtual dissipation function, the virtual Lagrangian function, and the generalized coordinates of each virtual node in the virtual passive system, includes the following steps 501b-502b: Step 501b: Construct the initial dynamic model of the virtual passive system based on the virtual dissipation function, the virtual Lagrangian function, the coordinates of each virtual node, and the external driving force of the virtual passive system; Step 502b: Convert the coordinates of the virtual nodes to Lagrange coordinates, and substitute the Lagrange coordinates into the initial dynamic model to obtain the virtual dynamic model of the virtual passive system.

[0174] For example, the initial dynamic model of the virtual passive system obtained above is:

[0175] The generalized Lagrange coordinate system is defined as follows:

[0176] Substituting the generalized Lagrange coordinates into the initial dynamic model above, and after coordinate transformation and simplification, the virtual dynamic model of the virtual passive system can be obtained, as follows:

[0177] in, The mass matrix of the virtual passive system is given by the following formula:

[0178] Gravity is given by the following formula:

[0179] The remaining items This represents the elastic force caused by the virtual spring. This represents the damping force caused by the virtual damper. This represents the frictional force caused by the interaction of the media. The elastic force term is:

[0180] The damping force term is:

[0181] The friction term is:

[0182] in, Indicates the relationship with the first A set of indices for all virtual nodes connected to a given virtual node.

[0183] In one optional embodiment of this application, after constructing the virtual dynamics model of the virtual passive system, a velocity tracker is constructed based on the virtual dynamics model. The specific construction process is as follows: In an optional embodiment of this application, after constructing the dynamic model of the UAV collaborative hoisting system based on the above embodiments, a virtual dynamic model of the virtual passive system can be constructed based on the dynamic model of the UAV collaborative hoisting system. The specific construction process is as follows: The dynamic model of the virtual passive system constructed according to the above embodiments is as follows:

[0184] Correspondingly, in the virtual passive system, the first The virtual dynamics model of a virtual node can be expressed as follows:

[0185] in, Indicates the first The mass of the drone, v represents the speed of the virtual passive system, and the term This represents the elastic force caused by the virtual spring. This represents the damping force caused by the virtual damper. This refers to the frictional force caused by the interaction of media.

[0186] Order No. The tracking error of a virtual node is defined as follows:

[0187] Among them, the desired speed satisfies , where v id v represents the expected speed of each virtual node. d The desired speed of the entire virtual passive system is described, that is, in the embodiments of this application, the entire virtual passive system is set to track a common desired system speed. Instead of requiring each virtual node to track its own desired speed.

[0188] The control input is constructed as follows:

[0189] in, As a feedforward term, it is used to compensate for system dynamics. For feedback items, It is a positive definite matrix.

[0190] By substituting the control input into the virtual dynamics model described above, redundant terms can be eliminated based on the substituted control model, thus obtaining the speed tracking controller provided in this application embodiment, which enables the UAV collaborative hoisting system to stably track the overall desired speed of the system.

[0191] Firstly, in the absence of external input, the dissipative forces within the virtual passive system drive the entire UAV-cable-load collaborative lifting system to evolve into a stable and robust integrated system based on virtual dynamics and physical dynamics. This integrated system possesses strong self-regulation and self-stabilization capabilities when subjected to disturbances, uncertainties, and state transitions. Through the energy dissipation characteristics of the passive system, it ultimately converges to a static and stable equilibrium state.

[0192] Secondly, the embodiments of this application determine the control input of the virtual passive system to track the desired speed of the system, so that within the system, the virtual passive system reaches a stable state, the virtual nodes remain relatively stationary to each other, and the system as a whole can asymptotically follow the target speed.

[0193] In summary, the obtained speed tracking controller decouples the internal coordination and overall trajectory tracking of the multi-agent hoisting system. The separation of internal and external trajectory tracking facilitates the independent design of speed tracking control laws for each UAV to guide the virtual system to achieve the desired motion, providing significant advantages for practical applications.

[0194] In an optional embodiment of this application, step 204b above, which involves mapping the acceleration of each node to the UAV collaborative hoisting system, includes: The corresponding node acceleration is calculated based on the coordinates of each UAV, the load coordinates, the coordinates of each virtual node, and the center point coordinates of the virtual pallet, resulting in the UAV acceleration in the UAV collaborative hoisting system.

[0195] The specific calculation process can be summarized as follows: In practice, the design of a multi-UAV cooperative transportation controller based on a passive system can be carried out in two stages: Phase 1: Construct a system without external input, i.e., without external disturbances u. i =0, a virtual passive system.

[0196] Based on the equations and the actual state of the UAV-cable-load UAV lifting system, a virtual system is established, forming a hybrid system that couples the virtual model with the physical system, which is the aforementioned virtual passive system. As in the above embodiment, external disturbance u... i When = 0, the virtual dynamic model of the virtual passive system is:

[0197] Controllable variables are ,like Able to track expectations :

[0198] The virtual passive system then becomes a strictly output passive system, exhibiting zero-input asymptotic stability and finite-gain stability characteristics, thereby achieving stable tracking and convergence characteristics.

[0199] Phase Two: Constructing a speed tracking controller for a virtual passive system:

[0200] in, Indicates the application of the first External forces on virtual nodes. In practice, items It can be considered a disturbance, and therefore ignored or incorporated into external disturbances. To mitigate steady-state error and improve regulation response, an integral-derivative feedback mechanism is introduced into the control law:

[0201] in, Indicates the first The speed tracking error of each virtual node, K P K represents the proportional control gain. I K represents the integral control gain. d This represents the differential control gain. The control input consists of a feedforward term. Composition, used to match the desired acceleration, and a feedback term. Used to suppress tracking errors. The controller can enable the virtual passive system to use [this feature] by adjusting the control input. Tracking the expected speed Under this control input, the first Acceleration generated by virtual nodes It is given by the following formula:

[0202] By integrating the first and second stages and mapping the desired acceleration of the virtual nodes back to the actual UAV collaborative hoisting system through geometric relationships, the first stage can be obtained. The expected acceleration of the drone a id :

[0203] Then, based on the dynamic model of the actual unmanned aerial vehicle system, the first... The desired thrust of the drone. In an optional embodiment of this application, step 204b above, determining the desired thrust of each drone based on the drone dynamics model of the drone cooperative lifting system and the acceleration of each node, includes: Based on the dynamic model of the actual unmanned aerial vehicle system, the first... Expected thrust of the drone:

[0204] Among them, T id Let m represent the desired thrust of the drone i. i a represents the mass of drone i. id G represents the acceleration of drone i. i f represents the gravity of drone i. i External disturbances to the drone, such as the main external disturbances acting on the drone during transportation, can be measured using force sensors or estimated using disturbance observers (such as extended state observers, compensation function observers, etc.).

[0205] In summary, based on the aforementioned speed tracker, this application's embodiments achieve complete decentralization through control input, eliminating reliance on direct communication between drones. Each drone only needs the relative position and speed information of its neighbors to achieve overall collaborative lifting control. The relative position and speed information of neighbors can be obtained through relative positioning technology, enabling the system to operate effectively even in environments with limited communication.

[0206] It should be understood that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order constraint on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the diagram may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0207] Please see Figure 17 One embodiment of this application provides a drone collaborative lifting control device 1700, applied to a drone collaborative lifting system. The drone collaborative lifting system includes at least multiple drone groups and loads connected to each drone group. The drone collaborative lifting control device 1700 includes: a first construction module 1710, a second construction module 1720, a determination module 1730, and a control module 1740, wherein: The first construction module 1710 is used to construct the dissipative node system of the UAV collaborative hoisting system; wherein, each virtual node in the dissipative node system corresponds one-to-one with each UAV in the UAV collaborative hoisting system; The second construction module 1720 is used to construct a virtual Lagrange model of the dissipative node system based on the Lagrange function, dissipation function, and generalized coordinates of each virtual node in the dissipative node system. The determination module 1730 is used to determine the control law of each UAV and the expected acceleration of each virtual node based on the virtual Lagrange model. The control module 1740 is used to control the operation of the UAV corresponding to each virtual node according to the control law of each virtual node and the desired acceleration of the node.

[0208] In one optional embodiment of this application, the second construction module 1720 is specifically used to determine the total kinetic energy and total potential energy of the UAV collaborative hoisting system; and to determine the Lagrangian function of the dissipative node system based on the total kinetic energy and total potential energy of the UAV collaborative hoisting system.

[0209] In one optional embodiment of this application, the total kinetic energy of the UAV collaborative hoisting system is the sum of the kinetic energies of all UAVs, the center point, and the load in the UAV collaborative hoisting system.

[0210] In one optional embodiment of this application, the total potential energy of the UAV collaborative hoisting system includes total gravitational potential energy and total elastic potential energy.

[0211] In an optional embodiment of this application, the second construction module 1720 is specifically used to: determine the adjacency coefficient between virtual nodes in the dissipative node system based on the coordinates of each drone and load in the drone collaborative hoisting system and a preset sensing radius; and determine the total elastic potential energy of the drone collaborative hoisting system based on the adjacency coefficient between virtual nodes, the elastic coefficient of the virtual spring between virtual nodes, and the spring length correction value.

[0212] In one optional embodiment of this application, the spring length correction value of the virtual spring is the difference between the current distance and the initial distance of each virtual node.

[0213] In an optional embodiment of this application, the second building module 1720 is specifically used to determine the relative motion dissipation function and the center rotation dissipation function of the dissipation node system; wherein, the relative motion dissipation function is a function used to characterize the energy dissipation caused by the relative motion between adjacent UAVs; the center rotation dissipation function is a function used to characterize the energy dissipation generated by the rotation of the UAV around the center point of the UAV group; and the dissipation function of the dissipation node system is determined according to the relative motion dissipation function and the center rotation dissipation function.

[0214] In an optional embodiment of this application, the second building module 1720 is specifically used to determine the relative motion dissipation function of the dissipation node system based on the spring length correction value of the virtual springs between each virtual node and the damping coefficient of the dissipation force between each UAV.

[0215] In an optional embodiment of this application, the second building module 1720 is specifically used to determine the center rotation dissipation function of the dissipation node system based on the generalized coordinates of each virtual node and the friction coefficient between the center point of the UAV group and the UAV.

[0216] In an optional embodiment of this application, the determining module 1730 is specifically used to determine the node expected acceleration of the UAV for each virtual node based on the system expected acceleration of the dissipative node system, the input control of the UAV in the virtual Lagrange model, and the mass of the UAV.

[0217] In an optional embodiment of this application, the determining module 1730 is specifically used to set the external force of the dissipative node system to 0, determine the input of the dissipative node system as the control input of the UAV, configure the height of each virtual node in the dissipative node system to be equal, and correct the virtual Lagrange model to obtain the control law of each UAV.

[0218] In one optional embodiment of this application, the second building module 1720 is specifically used to calculate the generalized coordinates of each virtual node in the dissipative node system based on the coordinates of each drone and load in the drone collaborative hoisting system and the preset height of each drone and load.

[0219] Specific limitations regarding the aforementioned human-machine collaborative hoisting control device 1700 can be found in the limitations of the human-machine collaborative hoisting control method described above, and will not be repeated here. Each module in the aforementioned human-machine collaborative hoisting control device 1700 can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0220] In one embodiment, a computer device is provided, the internal structure of which can be as follows: Figure 18As shown. The computer device includes a processor, memory, system interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and the database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database stores data. The system interface communicates with external terminals via system connection. When the computer program is executed by the processor, it implements the above-described human-machine collaborative hoisting control method. It includes: a memory and a processor; the memory stores the computer program; and the processor executes the computer program to implement any step of the above-described human-machine collaborative hoisting control method.

[0221] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, can perform any step of the above-described human-machine collaborative hoisting control method.

[0222] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0223] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0224] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0225] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0226] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0227] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for collaborative hoisting control using unmanned aerial vehicles (UAVs), characterized in that, An application to a drone collaborative lifting system, the drone collaborative lifting system comprising at least multiple drone groups and loads connected to each drone group, the method comprising: Construct a dissipative node system for the UAV collaborative hoisting system; wherein, each virtual node in the dissipative node system corresponds one-to-one with each UAV in the UAV collaborative hoisting system; A virtual Lagrange model of the dissipative node system is constructed based on the Lagrange function, dissipation function, and generalized coordinates of each virtual node in the dissipative node system. The control laws for each UAV and the expected acceleration of each virtual node are determined based on the virtual Lagrange model. The operation of the drone corresponding to each virtual node is controlled according to the control law of each virtual node and the desired acceleration of the node.

2. The UAV collaborative hoisting control method according to claim 1, characterized in that, The Lagrangian function of the dissipative node system is determined according to the following method: Determine the total kinetic energy and total potential energy of the UAV collaborative hoisting system; The Lagrangian function of the dissipative node system is determined based on the total kinetic energy and total potential energy of the UAV-coordinated hoisting system.

3. The UAV collaborative hoisting control method according to claim 2, characterized in that, The total kinetic energy of the UAV collaborative hoisting system is the sum of the kinetic energies of all UAVs, the center point, and the load in the UAV collaborative hoisting system.

4. The UAV collaborative hoisting control method according to claim 2, characterized in that, The total potential energy of the UAV-coordinated hoisting system includes total gravitational potential energy and total elastic potential energy.

5. The UAV collaborative hoisting control method according to claim 4, characterized in that, The total elastic potential energy of the UAV-coordinated hoisting system is determined according to the following method: The adjacency coefficient between virtual nodes in the dissipative node system is determined based on the coordinates of each drone and load in the drone collaborative hoisting system and the preset sensing radius. The total elastic potential energy of the UAV collaborative hoisting system is determined based on the adjacency coefficient between each virtual node, the elastic coefficient of the virtual spring between each virtual node, and the spring length correction value.

6. The UAV collaborative hoisting control method according to claim 5, characterized in that, The spring length correction value of the virtual spring is the difference between the current distance and the initial distance of each virtual node.

7. The UAV collaborative hoisting control method according to claim 1, characterized in that, The dissipation function of the dissipation node system is determined according to the following method: Determine the relative motion dissipation function and the center rotation dissipation function of the dissipation node system; wherein, the relative motion dissipation function is a function used to characterize the energy dissipation caused by the relative motion between adjacent UAVs; the center rotation dissipation function is a function used to characterize the energy dissipation generated by the rotation of the UAVs around the center point of the UAV group; The dissipation function of the dissipation node system is determined based on the relative motion dissipation function and the central rotation dissipation function.

8. The UAV collaborative hoisting control method according to claim 7, characterized in that, The relative motion dissipation function of the dissipation node system is determined according to the following method: The relative motion dissipation function of the dissipation node system is determined based on the spring length correction value of the virtual springs between each virtual node and the damping coefficient of the dissipation force between each UAV.

9. The UAV collaborative hoisting control method according to claim 7, characterized in that, The central rotational dissipation function of the dissipation node system is determined according to the following method: The central rotational dissipation function of the dissipation node system is determined based on the generalized coordinates of each virtual node and the friction coefficient between the center point of the UAV group and the UAV.

10. The UAV collaborative hoisting control method according to claim 1, characterized in that, The desired acceleration of each virtual node is determined based on the virtual Lagrange model, including: For each virtual node, the node-expected acceleration of the UAV is determined based on the system-expected acceleration of the dissipative node system, the input control of the UAV in the virtual Lagrange model, and the mass of the UAV.

11. The UAV collaborative hoisting control method according to claim 1, characterized in that, The control laws for each UAV are determined based on the virtual Lagrange model, including: The external force of the dissipative node system is set to 0, and the input of the dissipative node system is determined as the control input of the UAV. The height of each virtual node in the dissipative node system is configured to be equal. The virtual Lagrange model is then modified to obtain the control law of each UAV.

12. The UAV collaborative hoisting control method according to claim 1, characterized in that, The generalized coordinates of each virtual node in the dissipative node system are determined according to the following method: The generalized coordinates of each virtual node in the dissipative node system are calculated based on the coordinates of each drone and load in the drone collaborative hoisting system and the preset height of each drone and load.

13. A drone-assisted lifting control device, characterized in that, An apparatus for use in a drone collaborative lifting system, the drone collaborative lifting system comprising at least multiple drone groups and loads connected to each drone group, the apparatus comprising: The first construction module is used to construct the dissipative node system of the UAV collaborative hoisting system; wherein, each virtual node in the dissipative node system corresponds one-to-one with each UAV in the UAV collaborative hoisting system; The second construction module is used to construct a virtual Lagrange model of the dissipative node system based on the Lagrange function, dissipation function, and generalized coordinates of each virtual node in the dissipative node system. The determination module is used to determine the control law of each UAV and the expected acceleration of each virtual node based on the virtual Lagrange model. The control module is used to control the operation of the UAV corresponding to each virtual node according to the control law of each virtual node and the desired acceleration of the node.

14. A computer device, comprising: A memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method according to any one of claims 1 to 12.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 12.

Images