Multi-rotor unmanned aerial vehicle cooperative payload transmission formation control method and system

Abstract

The invention relates to the technical field of unmanned aerial vehicle formation cooperative transportation, and provides a multi-rotor unmanned aerial vehicle cooperative payload transmission formation control method and system, and the method comprises the steps: designing a model compensation controller according to the tension relation information of an unmanned aerial vehicle; designing an MCC control law according to a model compensation controller and an admittance model, constructing an objective function and a constraint function of a predictive control law, calculating optimal control input of the unmanned aerial vehicle, obtaining disturbance estimation of the unmanned aerial vehicle through a CFO observer, calculating an expected attitude according to a compensated unmanned aerial vehicle position control quantity, and tracking the expected attitude through a discrete PID controller. According to the method, tension fluctuation induced by load disturbance is mapped into position correction, coupling of tension and displacement control is achieved, the trigger frequency of control action can be effectively reduced, system energy consumption is reduced, system oscillation caused by sudden change of tension is remarkably restrained, and the overall stability and control precision in the collaborative transportation process are improved.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) swarm collaborative transportation technology, and in particular to a method and system for controlling the swarm collaborative payload transmission of multi-rotor UAVs. Background Technology

[0002] Single drones still face numerous limitations in terms of payload capacity, flight endurance, and system robustness, making them ill-suited for high-intensity transportation tasks in complex environments. Therefore, employing multiple drones in coordinated transport operations has become a hot research and application area. In recent years, multi-drone cooperative transport systems have demonstrated significant advantages in improving payload capacity and enhancing system fault tolerance. When one drone malfunctions, the others can continue flying in formation, effectively reducing the risk of mission failure and improving the overall system stability and reliability.

[0003] Currently, in collaborative transportation systems, multi-rotor drones are typically connected to the load via steel cables. While this connection method offers greater load-bearing capacity, it also introduces significant dynamic coupling problems. Changes in cable tension caused by the load can interfere with the attitude control and formation maintenance of each drone, posing a severe challenge to system modeling and control strategy design. Therefore, designing a formation control method that can ensure dynamic tension balance and achieve stable cooperation remains a key challenge in multi-drone collaborative logistics research, requiring further exploration and breakthroughs. Summary of the Invention

[0004] This invention aims to at least solve one of the technical problems existing in related technologies. To this end, this invention provides a method and system for coordinated payload transport formation control of multi-rotor unmanned aerial vehicles (UAVs). By mapping tension fluctuations induced by load disturbances to position corrections, tension and displacement control are coupled. While ensuring force balance, this effectively reduces the trigger frequency of control actions, decreases system energy consumption, and significantly suppresses system oscillations caused by sudden tension changes, thereby improving the overall stability and control accuracy during coordinated transport.

[0005] This invention provides a method for coordinated payload transmission formation control of multi-rotor unmanned aerial vehicles (UAVs), comprising: S1: Generate the tension relationship information of the UAV based on the tension information of the UAV, and design a model compensation controller based on the tension relationship information of the UAV. S2: Design the MCC control law based on the model compensation controller and admittance model, and calculate the position deviation of the UAV based on the UAV's position information and the MCC control law; S3: Construct the objective function and constraint function of the predictive control law based on the formation criteria of UAVs and the discrete augmented dynamics model of UAVs; S4: Calculate the optimal control input for the UAV based on the objective function, constraint function, and UAV position deviation; then calculate the UAV position control quantity based on the optimal control input. S5: Obtain the disturbance estimate of the UAV through the CFO observer, and compensate the UAV's position control quantity through the disturbance estimate of the UAV; S6: Calculate the desired attitude based on the compensated UAV position control quantity, and track the desired attitude through a discrete PID controller to achieve cooperative transportation of multi-rotor UAVs.

[0006] Furthermore, the formula for calculating the pull relationship information of the UAV is as follows: in, For the first The tension of the drone This is the combined inverse matrix of the cable direction. As an auxiliary intermediate vector, For the first The unit vector of the cable connecting the drone and its payload. for The second derivative, for The first derivative, This is the transpose of the matrix.

[0007] Furthermore, the calculation expression for the model compensation controller is: in, For state error, , This is the first state error. This is the second state error. For expected tension, For the first The tension estimate of the drone, For state error feedback, For feedback gain, , For the first feedback gain, For the second feedback gain, This is the error estimate. For the feedback compensation module, for The estimated value, For coupled dynamics, For the system's control gain, To control the output, for The second derivative, for The first derivative.

[0008] Furthermore, the calculation expression for the MCC control law is as follows: in, For the system's state error, For tension deviation, The damping coefficient is... This is the stiffness coefficient. The inertia coefficient, For time.

[0009] Furthermore, the objective function of the predictive control law is calculated as follows: in, for Time of the first The objective function of the predictive control law for an unmanned aerial vehicle (UAV) for Time used for the first The objective function for trajectory tracking of a drone. for Time used for the first The objective function for smoothing the control increments of the unmanned aerial vehicle (UAV) for Time of the first The predicted output of the drone, For the first The desired location of the drone. In the drone Control increment at any given moment; For the first The weighted matrix of the predicted output of the drone. For the first The weighted matrix for the incremental control of unmanned aerial vehicles (UAVs). To find the norm, For control boundaries, This is the prediction bound.

[0010] Furthermore, the expression for calculating the constraint function of the predictive control law is as follows: in, To control the lower bound of the input, In order to be in Time prediction Input at any time, To control the upper limit of the input, In order to be in Time prediction Control increment at any time, To control the lower bound of the rate, To control the upper bound of the rate, To minimize the distance between drones, The maximum distance between drones, For the first A drone in Time prediction The first two dimensions of the output at each moment, For the first A drone in Time prediction The first two dimensions of the output at each moment, To find the norm, For control boundaries, For the prediction boundary, For the first A collection of neighbors of a drone.

[0011] Furthermore, the calculation expression for the CFO observer is as follows: in, For the first drone The tension disturbance estimated at any given time. for The internal state of the CFO system at any given time. For the first drone The system state error at any given time, The observer gain matrix is... For discrete sampling time, This is the transpose of the matrix.

[0012] Furthermore, the expression for calculating the desired pose is: in, For drones Control over time For target drones exist x-axis component of the drone's position control quantity at any given time. For target drones exist The y-axis component of the drone's position control parameters at any given time. For target drones exist z-axis component of the drone's position control quantity at any given time. For drones quality It is the acceleration due to gravity. For target drones exist The expectation of time rolls around. For target drones exist Expected pitch angle at any moment For target drones exist The expected yaw angle at any given moment.

[0013] Furthermore, the calculation expression for the discrete PID controller is: in, For the first A drone in Expected control quantity at any given time It is a saturation function. for Expected control quantity at any given time For proportional gain, for Attitude angle error at time t, for Attitude angle error at time t, For integral gain, For derivative gain, for Attitude angle error at time t, This is the maximum allowable output limit for the control variable.

[0014] The present invention also provides a multi-rotor unmanned aerial vehicle (UAV) cooperative payload transmission formation control system for executing the above-mentioned multi-rotor UAV cooperative payload transmission formation control method, comprising: The first construction module generates the tension relationship information of the UAV based on the tension information of the UAV, and designs a model compensation controller based on the tension relationship information of the UAV. The position deviation calculation module designs an MCC control law based on the model compensation controller and admittance model, and calculates the position deviation of the UAV based on the UAV's position information and the MCC control law. The second construction module constructs the objective function and constraint function of the predictive control law based on the formation criteria of the UAV and the discrete augmented dynamics model of the UAV. The position control quantity calculation module calculates the optimal control input of the UAV based on the objective function, constraint function, and UAV position deviation, and calculates the position control quantity of the UAV based on the optimal control input. The position control quantity compensation module obtains the disturbance estimate of the UAV through the CFO observer and compensates the position control quantity of the UAV based on the disturbance estimate of the UAV. The attitude control module calculates the desired attitude based on the compensated UAV position control quantity and tracks the desired attitude through a discrete PID controller to achieve cooperative transportation of multi-rotor UAVs.

[0015] The above-described one or more technical solutions in the embodiments of the present invention have at least one of the following technical effects: This invention achieves the coupling of tension and displacement control by mapping tension fluctuations induced by load disturbances to position corrections. While ensuring force balance, it effectively reduces the triggering frequency of control actions, decreases system energy consumption, and significantly suppresses system oscillations caused by sudden tension changes, thereby improving the overall stability and control accuracy during coordinated transportation. Furthermore, it eliminates the need for explicit analytical solutions to the tension equilibrium point, avoiding problems of high computational complexity and poor convergence.

[0016] This invention achieves flexible adjustment through trajectory correction, significantly improving the system's real-time response capability and the feasibility of trajectory generation, thus adapting to the needs of complex mission scenarios in multi-aircraft formations. By constructing a high-precision observer and model compensation control structure, it can achieve rapid estimation and effective compensation of tension errors under conditions of rapid tension changes or sudden load changes. The system possesses strong disturbance adaptability, maintaining zero steady-state tension tracking in complex dynamic environments, thereby enhancing the system's robustness and reliability.

[0017] This invention employs a distributed model predictive control method, combining multiple constraints such as minimum spacing, collision avoidance, and tension balance. It can adapt to uncertainties such as unknown cable length and load mass, and has good flexibility, scalability, and safety assurance capabilities to solve the force balance problem in the dynamic load transmission process. It is suitable for large-scale UAV collaborative transportation missions.

[0018] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1This is a flowchart illustrating a multi-rotor UAV collaborative payload transmission formation control method provided by the present invention.

[0021] Figure 2 This is a schematic diagram of the structure of the multi-UAV collaborative transportation system provided in an embodiment of the present invention.

[0022] Figure 3 This is a schematic diagram of the body coordinate system and attitude angles of a quadcopter provided in an embodiment of the present invention.

[0023] Figure 4 This is a block diagram of a cooperative transportation control framework under force balance constraints provided in an embodiment of the present invention.

[0024] Figure 5 This is a control block diagram of tension tracking control provided in an embodiment of the present invention, including a higher-order derivative, disturbance observation and compensation, admittance model, etc.

[0025] Figure 6 This is a schematic diagram comparing the original expected trajectory and the actual running trajectory of each UAV provided in the embodiments of the present invention.

[0026] Figure 7 This is a curve showing the evolution of cable tension over time when using the proposed tension control algorithm, as provided in an embodiment of the present invention.

[0027] Figure 8 This is a curve showing the evolution of cable tension over time without tension control, provided in an embodiment of the present invention.

[0028] Figure 9 This is a graph showing the evolution of the relative distance between three UAVs under tension control, provided by an embodiment of the present invention.

[0029] Figure 10 This is a graph showing the evolution of three-axis trajectory tracking error over time based on DMPC, as provided in an embodiment of the present invention.

[0030] Figure 11 This is a graph comparing the tension tracking performance of different tension controllers in three-axis directions provided in the embodiments of the present invention. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention. The following embodiments are used to illustrate this invention but cannot be used to limit the scope of this invention.

[0032] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0033] The following is combined with Figures 1 to 11 This invention describes a method and system for coordinated payload transmission formation control of multi-rotor unmanned aerial vehicles (UAVs).

[0034] The structure of a multi-drone collaborative transportation system is as follows: Figure 2 As shown, the multi-drone collaborative transportation system includes: multi-rotor drones. The payload and the flexible cables connecting each UAV to the payload. Among them, the first... Multi-rotor drones The load is connected to the load via a corresponding cable, the length of which is denoted as . The horizontal relative distance between any two drones is denoted as Each cable transmits tension to the load along its own direction, and the magnitude of the corresponding tension is denoted as . , The direction is from the load. .

[0035] During collaborative transportation, the desired poses of each UAV are adjusted to make the tension transmitted by each cable tend to be balanced, and the collaborative transport of the load is completed under the conditions of meeting the minimum spacing between UAVs and collision avoidance constraints.

[0036] The principle of a quadcopter drone is as follows: Figure 3 As shown, for the first Considering disturbances from the payload, the dynamic model of a multi-rotor UAV is as follows: in, For the first The x-axis coordinate of the drone in the inertial coordinate system. For the first The y-axis coordinate of the drone in the inertial coordinate system. For the first The z-axis coordinate of the drone in the inertial coordinate system. For the first The pitch angle of the drone, For the first The roll angle of the drone For the first The yaw angle of the drone, For the first The thrust generated by the drone For the first The weight of the drone It is the acceleration due to gravity. To connect the first The x-axis component of the tension in the cable supporting the drone and its payload. To connect the first The tension x-axis component of the coordinate cable of the drone and its payload. The z-axis component of the tension in the cable connecting the drone and the payload. For the first The roll angle control torque of the drone. For the first The pitch control torque of the drone. It is the first Yaw angle control torque for the drone; The x-axis component of the moment of inertia. The y-axis component of the moment of inertia. The z-axis component of the moment of inertia. for The second derivative, for The first derivative.

[0037] The governing equations are: in, For drones State variables, For drones Position output, For drones The control input vector, , for x-axis component, for The y-axis component, for z-axis component, External interference caused by cable tension For the system matrix, To control the input matrix, For the interference input matrix, This is the output matrix.

[0038] Considering the impact of disturbances, the control equations are linearized and discretized, resulting in the following linear discrete-time system model: in, for The state variables of the drone at any given time, for The state variables of the drone at any given time, for The control input vector at time t, for External interference caused by cable tension at any given moment. for Real-time drone location output, For discrete state matrices, , The sampling interval used for position transformation. For discrete control input matrix, , For discrete interference input matrix, , This is the time integration variable.

[0039] like Figure 4 The diagram illustrates the complete control framework for a drone formation. For clarity, only the parts relevant to this application are shown, including: Force tracking control: used to obtain the force tracking control. Tension measurement of the quadcopter and its mooring cable Neighborhood tension mean and cable spatial orientation estimation ; It can be obtained from a tension gauge or an estimation algorithm based on attitude / load swing angle. It can be obtained by fusing IMU, vision, and geometric constraints.

[0040] Position tracking control: Input from tension control... The corrected position of the drone Track references can be received separately. and disturbance estimation As a feedforward, the desired pose is output. ,Will Integrating with trajectory errors, an attitude reference is generated to meet the requirements of traction direction and position tracking.

[0041] Attitude control: Input The internal closed loop also uses the current attitude and angular velocity, which are not explicitly shown in the figure, to output thrust and torque. Based on the attitude error, the system generates resultant force and torque commands that can be executed by the motor layer, which are then used by the dynamics component.

[0042] Quadrotor dynamics model: Input Output signal Under the command, the machine body and load are coupled to move, and the above state is fed back to the Disturbance observer; the conventional attitude / angular velocity feedback is closed within the Attitude control (not separately marked in the figure).

[0043] Disturbance observer: Input signal Output signal Based on position and velocity measurements, the estimated external wind disturbance, parameter uncertainty, and load oscillation coupling, etc., are considered as equivalent disturbances. The compensation is fed into the tension control and position tracking control to improve the robustness and steady-state accuracy of tension tracking and position tracking.

[0044] like Figure 1 As shown, a method for coordinated payload transmission formation control of multi-rotor unmanned aerial vehicles includes: S1: Generate the tension relationship information of the UAV based on the tension information of the UAV, and design a model compensation controller based on the tension relationship information of the UAV. Based on the dynamic model of the UAV, obtain the tension information, load status information and target UAV status information of the UAV; Based on the tension information, load status information, and target drone status information of the drone, tension relationship information is generated. The tension relationship information is used to describe the relationship between the tension information of the cable connecting the load and the target drone and the drone status and load status.

[0045] Adjusting the tension on each drone is necessary, but the external forces cannot be directly modulated by the quadcopter control input. Therefore, an admittance model is used to convert forces and torques into corresponding position and velocity changes within the robot frame.

[0046] To maintain tension balance within the collaborative transport cluster, each drone is modeled as a mass block. Based on force balance analysis, the tension acting on the drone is: in, The weight of the load. For the first The unit vector of the cable connecting the drone and its payload. This is the combined inverse matrix of the cable direction. , It is the acceleration due to gravity. The number of drones; Compared to the payload, the mass of the cable suspending the drone is negligible, and the cable remains taut throughout the entire transport process. Assuming... ,in, The location of the load. For the first The position of the drone, therefore the unit vector can be represented as Assuming So for If the unit vectors of the three drones and the payload are linearly independent, then... , For the combination matrix of rope directions, The system is underdetermined and has no solution. Therefore, this invention requires at least three drones. By differentiating twice with respect to time, the expression for calculating the tension relationship information is as follows: in, For the first The tension of the drone This is the combined inverse matrix of the cable direction. As an auxiliary intermediate vector, For the first The unit vector of the cable connecting the drone and its payload. for The second derivative, for The first derivative, This is the transpose of the matrix.

[0047] The calculation expression for the tension relationship information clarifies the influence of mass, damping, and stiffness parameters on error dynamics.

[0048] Let the first The balance required to mount a drone is The measured tension is Tension error is defined as , Then calculate the position correction value. The corrected setting value is obtained. , For the first Desired location for the drone For the first The corrected position of the drone.

[0049] The tension relationship information is rewritten as follows: in, For the system's control gain, For error signals, For coupled dynamics; The calculation expression for the model compensation controller is: in, For state error, , This is the first state error. This is the second state error. For expected tension, For the first The tension estimate of the drone, For state error feedback, For feedback gain, , For the first feedback gain, For the second feedback gain, This is the error estimate. For the feedback compensation module, for The estimated value, For coupled dynamics, For the system's control gain, To control the output.

[0050] S2: Design the MCC control law based on the model compensation controller and admittance model, and calculate the position deviation of the UAV based on the UAV's position information and the MCC control law; The state error feedback module ensures the stability of the closed-loop system, while the compensation module estimates and cancels complex unmodeled dynamics online, thereby ensuring the convergence of the closed-loop system error; the MCC control law is designed by combining the model compensation controller and the admittance model.

[0051] The admittance model, commonly used in robotic arm control, is introduced into the collaborative transportation scenario of multi-rotor UAVs; the estimated values ​​of cable tension and the unit vector of tension direction formed by the connection between each UAV and the load are obtained, and the expected tension is determined based on the initial equilibrium tension; The tension error induced by load variation is mapped to the desired position offset to generate a corrected reference trajectory for the UAV; Based on the virtual inertia, damping, and stiffness of the admittance parameters, a second-order error dynamic relationship between force and displacement is established, and the position / velocity correction is output. Under tension, the force response of the quadcopter UAV can be modeled as a mass-damper-spring system.

[0052] The admittance model is calculated as follows: in, For the first The corrected position of the drone For the first The desired location of the drone. The inertia coefficient, , for The x-direction component, for The y-direction component of for The z-direction component, The damping coefficient is... , for The x-direction component, for The y-direction component of for The z-direction component, This is the stiffness coefficient. , For expected tension.

[0053] By combining model-compensated control with admittance model, the MCC control law is obtained, and its calculation expression is as follows: in, For the system's state error, For tension deviation, The damping coefficient is... This is the stiffness coefficient. The inertia coefficient, For time.

[0054] The combination of model-compensated control and admittance model provides corrective position dynamics for UAV adjustments.

[0055] S3: Construct the objective function and constraint function of the predictive control law based on the formation criteria of UAVs and the discrete augmented dynamics model of UAVs; The discrete augmented dynamics model of the UAV is as follows: in, for The augmented state matrix at time t. To augment the discrete state matrix, for The augmented state matrix at time t. , To augment the discrete control input matrix, , Let be the dimension of the state vector. For dimension The identity matrix, To control the dimension of the input vector, For discrete output matrices, , For dimension The zero matrix, for The control input increment, , for Constantly control the input. To augment the state matrix, the prediction equation for future output within the optimization range is: in, For the first drone Predict output in real time. For the first The state gain matrix of the drone, For the first drone The augmented state of time, For the first The control gain matrix of the drone, For the first drone The vector that controls the input increment at every moment covers... Time's up Predict the control input increment at the predicted time. For the first drones based Time prediction The output of the time system For the first drones based Time prediction The output of the time system for Time of the first drones based Real-time status information obtained Constantly control the input increment. For the first drones based Time prediction Incremental control input at any given time.

[0056] The formation principles for drones include the tension balance of the connecting cables between drones, collision avoidance between drones, and the physical constraints of drones. Tension balance of connecting cables: The relative positions of the UAV and payload in the cooperative transport formation directly affect the tension of each cable. The energy consumption of the UAV is proportional to the change in cable tension. Generally, as the payload increases, the UAV must generate greater thrust to maintain stable flight and carry the payload. The force balance equation of the system is given by the following formula: in, It is the acceleration vector. The weight of the load.

[0057] For the For a drone, the main factors contributing to energy consumption include overcoming aerodynamic drag, supporting the drone's weight, and mitigating the tension from the load. Energy consumption is expressed by the following formula: in, For the first Energy consumption of drones The drag coefficient, air density, The projected area exposed to the airflow. For the first The velocity vector of the drone When the system operates in equilibrium under non-accelerated conditions, the change in power loss is primarily determined by the change in cable tension. Under tension imbalance, the power loss of one or more UAVs in the formation increases significantly, leading to uneven energy consumption across the entire cooperative transport system. Therefore, tension imbalance degrades underlying performance.

[0058] Collision avoidance between drones and the physical constraints of drones: in, To control the lower bound of the input, In order to be in Time prediction Input at any time, To control the upper limit of the input, In order to be in Time prediction Control increment at any time, To control the lower bound of the rate, To control the upper bound of the rate, To minimize the distance between drones, The maximum distance between drones, For the first A drone in Time prediction The first two dimensions of the output at each moment, For the first A drone in Time prediction The first two dimensions of the output at each moment, To find the norm, For control boundaries, For the prediction boundary, For the first A collection of neighbors of a drone.

[0059] The objective function of the predictive control law is expressed as follows: in, for Time of the first The objective function of the predictive control law for an unmanned aerial vehicle (UAV) for Time used for the first The objective function for trajectory tracking of a drone. for Time used for the first The objective function for smoothing the control increments of the unmanned aerial vehicle (UAV) for Time of the first The predicted output of the drone, For the first The desired location of the drone. In the drone Control increment at any given moment; For the first The weighted matrix of the predicted output of the drone. For the first The weighted matrix for the incremental control of unmanned aerial vehicles (UAVs). For control boundaries, This is the prediction bound.

[0060] S4: Calculate the optimal control input for the UAV based on the objective function, constraint function, and UAV position deviation; then calculate the UAV position control quantity based on the optimal control input. set up The control input is obtained through DMPC optimization. Assuming that the connection points between each UAV and payload are integrated with their respective attitude control subsystems, the UAV can operate without yaw motion, thus eliminating torque disturbances caused by cable tension.

[0061] Each drone uses an adjacent state. and the corrected reference trajectory To optimize its control input, the expression for calculating the position control quantity is: in, for Time of the first Position control parameters for unmanned aerial vehicles (UAVs) for Time of the first drones based The drone's position control parameters are obtained from real-time status information. To compensate for the gain, for Time of the first The drone's position was off.

[0062] S5: Obtain the disturbance estimate of the UAV through the CFO observer, and compensate the UAV's position control quantity through the disturbance estimate of the UAV; The tension disturbance information is as follows: in, for The estimated value of the drone disturbance at the specified time. for The time-matter model compensates for the internal state of the observer. , for The x-direction component, for The y-direction component of for The z-direction component, for Tension error at any moment, The observer gain matrix is ​​used to achieve disturbance compensation through feedforward control. When other external disturbances are much smaller than the tension disturbance, it can be considered that... , This is a disturbance feedforward compensation term. State error feedback ensures the stability of the closed-loop system, while the feedback compensation module estimates and cancels out complex unmodeled dynamics online, thereby ensuring the convergence of the closed-loop error.

[0063] S6: Calculate the desired attitude based on the compensated UAV position control quantity, and track the desired attitude through a discrete PID controller to achieve cooperative transportation of multi-rotor UAVs.

[0064] The expression for calculating the desired attitude is: in, For drones Control over time For target drones exist x-axis component of the drone's position control quantity at any given time. For target drones exist The y-axis component of the drone's position control parameters at any given time. For target drones exist z-axis component of the drone's position control quantity at any given time. For drones quality It is the acceleration due to gravity. For target drones exist The expectation of time rolls around. For target drones exist Expected pitch angle at any moment For target drones exist The expected yaw angle at any given moment.

[0065] For the desired Euler angles , A discrete PID controller was designed to track the error. Defined as: The calculation expression for a discrete PID controller is: in, For the first A drone in Expected control quantity at any given time It is a saturation function. for Expected control quantity at any given time For proportional gain, for Attitude angle error at time t, for Attitude angle error at time t, For integral gain, For derivative gain, for Attitude angle error at time t, This is the maximum allowable output limit for the control variable.

[0066] The entity executing the cooperative transportation control method is a terminal device. Types of terminal devices include, but are not limited to, mobile phones, tablets, laptops, ultra-mobile personal computers (UMPCs), netbooks, personal digital assistants (PDAs), etc. This application embodiment does not impose any restrictions on the specific type of terminal device.

[0067] Figure 5 This is a block diagram of tension tracking control, which is a dual-loop control structure. Tension control and formation holding form the outer loop, while tracking control forms the inner loop.

[0068] For ease of explanation, only the parts relevant to this application are shown, such as... Figure 5 As shown, it includes: HOD: For input Perform higher-order derivative estimation / extraction to obtain and (Used for subsequent error construction and feedforward compensation). Input Output and and will The circuit is bypassed to the left bus as is.

[0069] State error feedback: Generates state feedback control input based on tension state error. Input comes from the left summation node. (Depend on The combined tension state error signal is output. (State error feedback term), fed down to the right summation node.

[0070] Disturbance feedback compensation: Disturbance feedback compensation is performed based on an equivalent unknown disturbance estimate. Input from the CFO. and the left-side bus (shown as the input parallel to the error signal). Output (Disturbance compensation term), and Add the summation nodes on the right.

[0071] b and CFO: b receives from The feedback from b outputs a calibrated signal to the CFO, which is used to excite the disturbance estimation link. The CFO estimates the unknown disturbance based on the input from b and (optionally) the UAV status, and outputs... To Disturbancefeedback compensation.

[0072] UAV controller: Used to incorporate tension estimates and their derivatives into the constraints or feedforwards of attitude / thrust control. Input (Estimated values ​​of tension and its rate of change). Output (The desired or corrected displacement / position amount generated by the control), compared with the reference The error is obtained by subtracting the summation nodes on the right. .

[0073] Impedance model: Achieves the desired dynamic mapping of tension-displacement based on an impedance / compliance model. Input control input. and error (Depend on and (Obtained by subtraction). The output is the internal variable used to drive the tension tracking dynamics (the external ports are not explicitly shown in the figure; their response is influenced by the system loop). Equal amounts).

[0074] Extensive experiments were conducted to verify the effectiveness of the invention. Simulation experiments comparing the expected running trajectory with the actual running trajectory are as follows: Figure 6 As shown, the initial position of the first drone. , The initial position of the second drone , The initial position of the second drone , The initial position of the payload is , The prediction and control levels are set as follows: =20 and The constraint parameters are configured as follows: maximum distance between UAVs. , Minimum distance between drones , Lower bound of control input Upper limit of control input , Control rate lower bound upper limit of control rate , Set the weight matrix as follows Set the parameters and run the simulation. The simulation results are as follows: Figure 6As shown, the expected trajectory of the load is a figure 8.

[0075] Figure 7 and Figure 8 This demonstrates how the tension tracking controller effectively adjusts rope tension and ensures convergence to the desired value.

[0076] like Figure 7 As shown, when It exhibits overshoot and oscillation, with a maximum amplitude of approximately 4.5 N and a minimum of 2.6–2.8 N, after which it decays exponentially. The tension of the three cables was almost all in contact with the ground. Near the 3.6N mark, even with trajectory inflections during the 60–80s period, the tension only showed a small "bump" before immediately converging again. This indicates a clear zero (or near-zero) steady-state error characteristic. The oscillations in the initial stage were quickly suppressed, demonstrating that the outer loop's adherence and disturbance observation feedforward (CFO) effectively handled the "force-to-displacement" coupling. The inner loop's DMPC was not significantly disturbed, and the disturbance to the tension caused by the mid-section trajectory changes was quickly absorbed, reflecting the effectiveness of the feedforward / compensation.

[0077] like Figure 8 As shown, by eliminating the tension control component and relying solely on position / formation control, the result shows three cables encircling each other. It exhibits slow drift with noticeable fluctuations, lacking stable alignment; frequent peaks and troughs, fluctuating between approximately 3.0–3.8 N, with the lowest point approaching 2.8–2.9 N, and an overall rise around 40–60 seconds. Figure 8 This indicates that without a force loop, the acceleration / deceleration of the load and the geometric changes of the cable are directly mapped to tension offset. Equivalently, gusts / maneuvering / model errors will all be directly reflected in the tension, resulting in poor balance and potentially leading to uneven energy consumption and increased attitude load. Figure 7 In comparison, with the addition of tension control, the mean alignment and variance are significantly reduced, resulting in a smoother steady state.

[0078] like Figure 9 It shows the relative distances between the three drones. Figure 9 Figure (a) in the middle is and The relative distance between them Figure 9 Figure (b) is and The relative distance between them Figure 9 Figure (c) in the middle is and The relative distances between these three pairs of distances form three curves. All three curves initially exhibit a slight "drop / peak" before entering a monotonous or small-amplitude oscillating convergent process, eventually stabilizing at a constant value of approximately 3.0–3.1 m. The most dramatic curve shows a noticeable ringing effect within 0–5 seconds, but… The system was adequately damped; thereafter, only millimeter- to centimeter-level perturbations appeared during the 60–80s turning segment. The experiment showed that the DMPC distance constraint was effective; at the same time, the tension control did not disrupt the geometry within the formation. The outer ring changed the reference pose to achieve tension balance, but the inner ring could still lock the inter-machine distance within the safety belt, indicating that the weight and constraint settings were reasonable.

[0079] like Figure 10 The figure shows the DMPC position tracking errors of three UAVs. Figure 10 Figure (a) shows the X-axis position tracking error. Figure 10 Figure (b) shows the Y-axis position tracking error. Figure 10 Figure (c) shows the Z-axis position tracking error. Due to the position change rate reaching the maximum control limit, the Y-axis error fluctuates between 20 s and 80 s.

[0080] The configuration is determined by the rope tension. The desired arrangement can be derived from the force balance equations by setting the angle between the rope and the ground to a specified value. In our simulation, the payload trajectory is given by the following equation: in, The desired location of the load.

[0081] like Figure 11 As shown, Figure 11 Figure (a) shows the X-plane tension tracking task. Figure 11 Figure (b) shows the Y-plane tension tracking task. The mean absolute error (MAE) of the CFO-based MCC-AC in the X and Y directions are 0.0819 N and 0.0303 N, respectively. Compared with the ESO-based MCC-AC (0.0905 N and 0.0347 N), these represent reductions of 9.5% and 12.7%, respectively; compared with the PID controller (0.1207 N and 0.0776 N), these represent reductions of 32.2% and 60.9%, respectively. In the Z direction, as shown... Figure 11 As shown in Figure (c), the MAE of the CFO-based MCC-AC is 0.0333N, which is 34.8% and 33.9% higher than that of the ESO-based MCC-AC (0.0511N) and the PID controller (0.0504N), respectively. These results demonstrate that the compensation function observer can more accurately estimate and compensate for model uncertainties in multidimensional coupled environments.

[0082] The mechanical balance and distributed control principle of the multi-rotor UAV collaborative hoisting system in this application and embodiment are as follows: The terminal control equipment can first acquire the attitude angles, position coordinates, and velocity information of each transport drone, and then measure the amplitude and direction vector of the suspension rope tension in real time. Based on the tension error signal, combined with the relative positions and formation constraints between the drones, the system generates a reference position correction for tension balance, thereby achieving outer-loop tension-position coordinated control.

[0083] Simultaneously, based on the corrected position information and velocity status of each UAV, the system employs a distributed model predictive control (DMPC) algorithm to generate the optimal control input for each UAV, considering control constraints, communication delays, and collision avoidance constraints, thus achieving coordinated trajectory control in the inner loop. The outer loop uses an admittance model to convert the deviation between the desired and actual tension into desired displacement compensation, thereby establishing a dynamic compliant relationship between the load motion and the UAV formation. To suppress external disturbances and nonlinear model errors, the system introduces a Compensation Function Observer (CFO) to achieve rapid estimation and feedforward compensation of unknown disturbances, resulting in a smoother and more stable tension response. Through this multi-level control structure, the system can achieve force balance maintenance and precise path tracking control of the load under the coordinated traction of multiple UAVs, ensuring the stability and safety of the transmission process.

[0084] The control method in this embodiment collects attitude, position, velocity, and tension data of each UAV and the payload system. Based on the tension error signal and the desired tension value, a displacement correction is generated using a compliance model to obtain the desired position reference for each UAV. A distributed model predictive control algorithm is used to perform rolling optimization of the state of each UAV in the prediction time domain to generate the optimal control quantity. External disturbances and model biases are estimated in real time using a Compensation Function Observer (CFO), and the compensation signal is superimposed on the control input to achieve disturbance suppression. The control law and thrust command for each UAV are updated based on constraints (including maximum speed, acceleration, inter-UAV distance, and communication delay). The control commands are executed to drive multiple UAVs to collaboratively transport the payload while maintaining tension balance, achieving smooth tracking of the desired path and stable payload attitude.

[0085] This invention offers numerous advantages over traditional control model systems. Firstly, it features a clear hierarchy and strong real-time performance, with decoupling of outer-loop tension adjustment and inner-loop trajectory control, enabling rapid response to load disturbances. Secondly, it boasts high disturbance rejection and steady-state accuracy, with the CFO observer effectively filtering out high-frequency disturbances and significantly reducing tension steady-state error. Thirdly, it offers superior safety, with DMPC embedding distance and velocity constraints to ensure inter-drone collision avoidance and path feasibility. Fourthly, it exhibits strong scalability, employing a distributed structure that easily extends to more drone formation collaboration scenarios. Finally, it optimizes control accuracy and energy consumption, with tension balancing ensuring even load distribution across each drone, resulting in lower overall energy consumption and smoother transmission.

[0086] A multi-rotor unmanned aerial vehicle (UAV) cooperative payload transfer formation control system, used to execute the aforementioned multi-rotor UAV cooperative payload transfer formation control method, comprising: The first construction module generates the tension relationship information of the UAV based on the tension information of the UAV, and designs a model compensation controller based on the tension relationship information of the UAV. The position deviation calculation module designs the MCC control law based on the model compensation controller and admittance model, and calculates the position deviation of the UAV based on the UAV's position information and the MCC control law. The second construction module constructs the objective function and constraint function of the predictive control law based on the formation criteria of the UAV and the discrete augmented dynamics model of the UAV; The position control quantity calculation module calculates the optimal control input of the UAV based on the objective function, constraint function, and UAV position deviation, and then calculates the UAV's position control quantity based on the optimal control input. The position control quantity compensation module obtains the disturbance estimate of the UAV through the CFO observer and compensates the UAV's position control quantity based on the disturbance estimate. The attitude control module calculates the desired attitude based on the compensated UAV position control quantity, and tracks the desired attitude through a discrete PID controller to achieve cooperative transportation of multi-rotor UAVs.

[0087] Through the coordinated operation of the above modules, tension fluctuations induced by load disturbances are mapped to position corrections, achieving coupling between tension and displacement control. While ensuring force balance, this effectively reduces the triggering frequency of control actions, decreases system energy consumption, and significantly suppresses system oscillations caused by sudden tension changes, thereby improving the overall stability and control accuracy during coordinated transportation.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for coordinated payload transmission formation control of multi-rotor unmanned aerial vehicles, characterized in that, include: S1: Generate the tension relationship information of the UAV based on the tension information of the UAV, and design a model compensation controller based on the tension relationship information of the UAV. S2: Design the MCC control law based on the model compensation controller and admittance model, and calculate the position deviation of the UAV based on the UAV's position information and the MCC control law; S3: Construct the objective function and constraint function of the predictive control law based on the formation criteria of UAVs and the discrete augmented dynamics model of UAVs; S4: Calculate the optimal control input for the UAV based on the objective function, constraint function, and UAV position deviation; then calculate the UAV position control quantity based on the optimal control input. S5: Obtain the disturbance estimate of the UAV through the CFO observer, and compensate the UAV's position control quantity through the disturbance estimate of the UAV; S6: Calculate the desired attitude based on the compensated UAV position control quantity, and track the desired attitude through a discrete PID controller to achieve cooperative transportation of multi-rotor UAVs.

2. The method for coordinated payload transmission formation control of a multi-rotor unmanned aerial vehicle (UAV) according to claim 1, characterized in that, The formula for calculating the thrust relationship information of the drone is: in, For the first The tension of the drone This is the combined inverse matrix of the cable direction. As an auxiliary intermediate vector, For the first The unit vector of the cable connecting the drone and its payload. for The second derivative, for The first derivative, This is the transpose of the matrix.

3. The method for coordinated payload transmission formation control of a multi-rotor unmanned aerial vehicle (UAV) according to claim 1, characterized in that, The calculation expression for the model compensation controller is: in, For state error, , This is the first state error. This is the second state error. For expected tension, For the first The tension estimate of the drone, For state error feedback, For feedback gain, , For the first feedback gain, For the second feedback gain, This is the error estimate. For the feedback compensation module, for The estimated value, For coupled dynamics, For the system's control gain, To control the output, for The second derivative, for The first derivative.

4. The method for coordinated payload transmission formation control of a multi-rotor unmanned aerial vehicle (UAV) according to claim 3, characterized in that, The calculation expression for the MCC control law is: in, For the system's state error, For tension deviation, The damping coefficient is... This is the stiffness coefficient. The inertia coefficient, For time.

5. The method for coordinated payload transmission formation control of a multi-rotor unmanned aerial vehicle (UAV) according to claim 1, characterized in that, The objective function of the predictive control law is expressed as follows: in, for Time of the first The objective function of the predictive control law for an unmanned aerial vehicle (UAV) for Time used for the first The objective function for trajectory tracking of a drone. for Time used for the first The objective function for smoothing the control increments of the unmanned aerial vehicle (UAV) for Time of the first The predicted output of the drone, For the first The desired location of the drone. In the drones Control increment at any given moment; For the first The weighted matrix of the predicted output of the drone. For the first The weighted matrix for the incremental control of unmanned aerial vehicles (UAVs). To find the norm, For control boundaries, This is the prediction bound.

6. The method for coordinated payload transmission formation control of a multi-rotor unmanned aerial vehicle (UAV) according to claim 1, characterized in that, The expression for calculating the constraint function of the predictive control law is: in, To control the lower bound of the input, In order to be in Time prediction Input at any time, To control the upper limit of the input, In order to be in Time prediction Control increment at any time, To control the lower bound of the rate, To control the upper bound of the rate, To minimize the distance between drones, The maximum distance between drones, For the first A drone in Time prediction The first two dimensions of the output at each moment, For the first A drone in Time prediction The first two dimensions of the output at each moment, To find the norm, For control boundaries, For the prediction boundary, For the first A collection of neighbors of a drone.

7. The method for coordinated payload transmission formation control of a multi-rotor unmanned aerial vehicle (UAV) according to claim 1, characterized in that, The calculation expression for the CFO observer is: in, For the first drones The tension disturbance estimated at any given time. for The internal state of the CFO system at any given moment. For the first drones The system state error at any given time, The observer gain matrix is... For discrete sampling time, This is the transpose of the matrix.

8. The method for coordinated payload transmission formation control of a multi-rotor unmanned aerial vehicle (UAV) according to claim 1, characterized in that, The expression for calculating the desired attitude is: in, For drones Control over time For target drones exist x-axis component of the drone's position control quantity at any given time. For target drones exist The y-axis component of the drone's position control parameters at any given time. For target drones exist z-axis component of the drone's position control quantity at any given time. For drones quality It is the acceleration due to gravity. For target drones exist The expectation of time rolls around. For target drones exist Expected pitch angle at any moment For target drones exist The expected yaw angle at any given moment.

9. The method for coordinated payload transmission formation control of a multi-rotor unmanned aerial vehicle (UAV) according to claim 1, characterized in that, The calculation expression for a discrete PID controller is: in, For the first A drone in Expected control quantity at any given time It is a saturation function. for Expected control quantity at any given time For proportional gain, for Attitude angle error at time t, for Attitude angle error at time t, For integral gain, For derivative gain, for Attitude angle error at time t, This is the maximum allowable output limit for the control variable.

10. A multi-rotor unmanned aerial vehicle (UAV) cooperative payload transmission formation control system, characterized in that, To execute the multi-rotor UAV cooperative payload transfer formation control method as described in any one of claims 1 to 9, comprising: The first construction module generates the tension relationship information of the UAV based on the tension information of the UAV, and designs a model compensation controller based on the tension relationship information of the UAV. The position deviation calculation module designs an MCC control law based on the model compensation controller and admittance model, and calculates the position deviation of the UAV based on the UAV's position information and the MCC control law. The second construction module constructs the objective function and constraint function of the predictive control law based on the formation criteria of the UAV and the discrete augmented dynamics model of the UAV. The position control quantity calculation module calculates the optimal control input of the UAV based on the objective function, constraint function, and UAV position deviation, and calculates the position control quantity of the UAV based on the optimal control input. The position control quantity compensation module obtains the disturbance estimate of the UAV through the CFO observer and compensates the position control quantity of the UAV based on the disturbance estimate of the UAV. The attitude control module calculates the desired attitude based on the compensated UAV position control quantity and tracks the desired attitude through a discrete PID controller to achieve cooperative transportation of multi-rotor UAVs.

Images