Unmanned aerial vehicle cooperative carrying control method and device, storage medium and unmanned aerial vehicle system

Abstract

The invention discloses an unmanned aerial vehicle cooperative carrying control method and device, a storage medium and an unmanned aerial vehicle system, relates to the technical field of unmanned aerial vehicle control, and mainly aims to solve the problem of low unmanned aerial vehicle cooperative carrying control precision. The method mainly comprises the steps that carrying distribution data are determined through a central controller, and control parameters of all unmanned aerial vehicles are determined according to the carrying distribution data; in the process that the unmanned aerial vehicle executes the cooperative carrying task according to the control parameters, carrying distribution data is updated based on a dynamic load adjustment strategy through the central controller, and the control parameters are updated according to the updated carrying distribution data; and in the flight process of the unmanned aerial vehicle according to the control parameters and the updated control parameters, single-vehicle control parameter adjustment is performed on the corresponding unmanned aerial vehicle through each local controller based on at least one of the configured control model, the contact compensation controller and the load compensation controller. The method is mainly used for controlling multi-unmanned-aerial-vehicle cooperative carrying.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle (UAV) control technology, and in particular to a UAV collaborative transportation control method and device, a storage medium, and a UAV system. Background Art

[0002] Tandem-rotor drones, as advanced modern aircraft, hold broad application prospects in numerous fields. However, their system characteristics are extremely complex, with highly coupled and nonlinear lateral and longitudinal channels. This not only increases system complexity but also leads to poor environmental adaptability. In missions involving multiple drones coordinating to carry payloads, uneven load distribution frequently arises, seriously impacting mission execution.

[0003] Most existing control schemes are based on the classic PID control method, which suffers from cumbersome parameter adjustment, poor robustness, and poor control response for nonlinear, strongly coupled systems. While related methods based on modern control theory can improve control effectiveness to a certain extent, they still suffer from high computational complexity, difficult controller design, poor adaptability to system nonlinear characteristics, and sensitivity to changes in system parameters. These methods struggle to meet the high-precision and high-reliability requirements of coordinated transport control for tandem-rotor UAVs. Summary of the Invention

[0004] In view of this, the present invention provides a method and device for controlling cooperative transportation of unmanned aerial vehicles, a storage medium, and an unmanned aerial vehicle system, the main purpose of which is to solve the problem of low accuracy of existing cooperative transportation control of unmanned aerial vehicles.

[0005] According to one aspect of the present invention, a method for controlling cooperative transport of unmanned aerial vehicles (UAVs) is provided. A transport mission is performed collaboratively by multiple UAVs, each equipped with a local controller. Each local controller is in communication with a central controller. The central controller is configured with a dynamic load adjustment strategy. The local controllers are configured with a control model for single-machine control, a contact compensation controller for dealing with contact disturbances, and a load compensation controller for dealing with sudden load changes. The method includes:

[0006] Determining, by the central controller, transport allocation data based on the performance parameters and transport object parameters of each UAV, and determining control parameters of each UAV based on the transport allocation data;

[0007] During the process of the UAVs performing the coordinated transport mission according to the control parameters, the central controller updates the transport allocation data based on a dynamic load adjustment strategy according to the real-time monitoring data collected from each UAV, and updates the control parameters according to the updated transport allocation data;

[0008] During the flight of the UAV according to the control parameters and the updated control parameters, the individual control parameters of the corresponding UAV are adjusted through each local controller based on at least one of the configured control model, contact compensation controller and load compensation controller.

[0009] Furthermore, the control model is constructed based on the feedback gain matrix and the adaptive gain matrix. For each local controller, the control parameters of the corresponding UAV are adjusted through the configured control model, including:

[0010] Generate control laws based on real-time monitoring data, single-machine system parameter estimates, regression vectors, and feedback gain matrices;

[0011] controlling the UAV according to the control law and collecting control output generated by the control law;

[0012] Adaptively adjust the estimated values ​​of the single-machine system parameters according to the control output and the adaptive gain matrix to obtain updated control parameters.

[0013] Furthermore, the control model is constructed based on a plurality of candidate control models, and for each of the local controllers, the corresponding UAV is subjected to single-machine control parameter adjustment through the configured control model, including:

[0014] Calculate the control laws of different candidate control models based on their feedback gain matrices and real-time monitoring data;

[0015] Controlling the UAV according to the control law, and calculating the deviation between the control output and the control expectation under the control of different candidate control models;

[0016] The candidate control model with the smallest deviation is determined as the target control model, and the control parameters are adjusted according to the target control model.

[0017] Furthermore, the contact compensation controller is constructed based on the principle of estimated model difference. When any of the UAVs has a contact disturbance, the contact compensation controller is used to adjust the control parameters of the single UAV, including:

[0018] Obtaining output feedback of the UAV, and generating a basic control input through a nominal controller according to the received control expectation and output feedback;

[0019] The output feedback is modeled in real time using Kalman filtering, and a compensation signal is generated through a contact compensation controller based on the modeling results.

[0020] Performing superimposed compensation on the basic control input based on the compensation signal to obtain a compensated control input;

[0021] The control output generated by the compensated control input is used as output feedback, and the process returns to the step of obtaining the output feedback of the UAV to obtain a control input that approximates the control expectation, thereby achieving control parameter adjustment.

[0022] Furthermore, the load compensation controller includes force control feedback and acceleration feedback. When a sudden load change occurs on any of the UAVs, the load compensation controller is used to adjust the control parameters of the single UAV, including:

[0023] The load compensation controller obtains load change data, force feedback signals and acceleration feedback signals through sensors carried on the drone;

[0024] configuring a force feedback gain coefficient and an acceleration feedback gain coefficient according to the load change data;

[0025] The weight of the force feedback signal in the control loop is adjusted by the force feedback gain coefficient, and the weight of the acceleration feedback signal in the control loop is adjusted by the acceleration feedback gain coefficient, so as to realize the control parameter adjustment of a single UAV in the case of a sudden load change.

[0026] Furthermore, the real-time monitoring data includes real-time load, and the updating of the transport allocation data based on the real-time monitoring data collected from each UAV and a dynamic load adjustment strategy includes:

[0027] For each UAV, respectively adjust the proportional control gain and the differential control gain based on the difference between the real-time load and the target load in the transport allocation data to obtain a load adjustment for each UAV;

[0028] For each UAV, the sum of the load adjustment amount and the real-time load is used as the adjusted load, and the adjusted load of each UAV is used to update the respective target loads in the transport distribution data to obtain updated transport distribution data.

[0029] Furthermore, the real-time monitoring data also includes flight parameters, and the method further includes:

[0030] For each UAV, calculating a flight deviation based on the flight parameters, wherein the flight deviation is used to represent a difference between an observed value based on the flight parameters and a predicted value based on the flight state;

[0031] If the flight deviation is greater than a preset deviation threshold, the UAV is determined to be a faulty UAV;

[0032] When it is determined that the fault level of the UAV is a mild fault based on the comparison result of the key flight parameters of the faulty UAV with the corresponding threshold value, the control parameters of the UAV are adjusted according to the fault compensation gain matrix;

[0033] When it is determined that the fault level is a serious fault, the faulty UAV is guided to return to the base by sending a control instruction, and the transport allocation data is re-determined based on the performance parameters and transport object parameters of the remaining non-faulty UAVs.

[0034] According to another aspect of the present invention, a UAV collaborative transport control device is provided. Multiple UAVs are used to collaboratively perform transport tasks. Each UAV is equipped with a local controller, and each local controller is separately connected to a central controller. The central controller is configured with a dynamic load adjustment strategy. The local controllers are configured with a control model for single-machine control, a contact compensation controller for dealing with contact disturbances, and a load compensation controller for dealing with sudden load changes. The device includes:

[0035] A transport allocation module is configured to determine transport allocation data based on the performance parameters and transport object parameters of each UAV based on the central controller, and determine control parameters of each UAV based on the transport allocation data;

[0036] a load allocation adjustment module, configured to update the load allocation data based on a dynamic load adjustment strategy based on real-time monitoring data collected from each UAV during the process of the UAVs performing the coordinated load mission according to the control parameters, and to update the control parameters based on the updated load allocation data;

[0037] The stand-alone control parameter adjustment module is used to adjust the stand-alone control parameters of the corresponding UAV based on the local controller during the flight of the UAV according to the control parameters and the updated control parameters, by each local controller through at least one of the configured control model, contact compensation controller and load compensation controller.

[0038] According to another aspect of the present invention, a storage medium is provided, wherein the storage medium stores at least one executable instruction, and the executable instruction enables a processor to perform operations corresponding to the above-mentioned UAV collaborative transportation control method.

[0039] According to another aspect of the present invention, there is provided a drone system comprising a plurality of drones and a central controller, wherein each drone is equipped with a local controller, and each local controller is respectively communicatively connected to the central controller;

[0040] The central controller is used to perform operations of carrying allocation of multiple UAVs and updating carrying allocation data;

[0041] Any of the local controllers is used to perform operations for controlling parameters of a single UAV.

[0042] By means of the above technical solution, the technical solution provided by the embodiment of the present invention has at least the following advantages:

[0043] The present invention provides a method and device for controlling cooperative transport of unmanned aerial vehicles (UAVs), a storage medium, and a UAV system. In an embodiment of the present invention, a central controller determines transport allocation data based on performance parameters and transport object parameters of each UAV, and determines control parameters for each UAV based on the transport allocation data. The central controller updates the transport allocation data based on a dynamic load adjustment strategy and updates the control parameters based on the updated transport allocation data while the UAVs perform cooperative transport tasks according to the control parameters, based on real-time monitoring data collected from each UAV. The control parameters are then updated based on the updated transport allocation data. While the UAVs fly according to the control parameters and the updated control parameters, each local controller adjusts the control parameters of its corresponding UAV using at least one of the configured control model, contact compensation controller, and load compensation controller. Local control parameter adjustment ensures the timeliness of the UAV's ability to counteract disturbances and sudden load changes, and ensures accurate control parameter adjustment. Furthermore, transport allocation by the central controller achieves balanced load distribution among multiple UAVs during cooperative transport, avoiding degradation of flight performance and safety hazards caused by uneven loads, thereby significantly improving control accuracy.

[0044] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are specifically listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0046] Figure 1 A flow chart of a UAV cooperative transport control method provided by an embodiment of the present invention is shown;

[0047] Figure 2 A schematic diagram of the geometric relationship of cooperative transportation of UAVs provided by an embodiment of the present invention is shown;

[0048] Figure 3A schematic diagram of a control process for adjusting control parameters of a single machine by a contact compensation controller provided by an embodiment of the present invention is shown;

[0049] Figure 4 A schematic diagram of a control process for adjusting control parameters of a single machine by a load compensation controller provided by an embodiment of the present invention is shown;

[0050] Figure 5 A schematic diagram of the design concept of a single-machine to multi-machine UAV transport cooperative control provided by an embodiment of the present invention is shown;

[0051] Figure 6 The figure shows a block diagram of a UAV cooperative transport control device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0053] Aiming at the problem of low precision in the existing UAV collaborative transportation control. An embodiment of the present invention provides a UAV collaborative transportation control method, which proposes a method based on a distributed control architecture, that is, multiple UAVs collaboratively perform transportation tasks, each UAV is equipped with a local controller, and each local controller is respectively communicated with a central controller. In this process, multiple UAVs are regarded as a whole system, and precise control of the entire system is achieved through the collaboration between the central controller and the local controllers of each UAV. The central controller is equipped with a dynamic load adjustment strategy for global task planning and load distribution strategy formulation. The local controller is equipped with a control model for single-machine control, a contact compensation controller for dealing with contact disturbances, and a load compensation controller for dealing with load mutations. It can achieve precise control of the machine according to the instructions of the central controller and combined with its own state information. As Figure 1 As shown, the method includes:

[0054] 101. Determine transport allocation data based on performance parameters and transport object parameters of each UAV, and determine control parameters of each UAV based on the transport allocation data.

[0055] In an embodiment of the present invention, before multiple drones collaborate to perform a transport mission, the central controller needs to fully understand the performance parameters of each drone, such as maximum load capacity, flight speed range, endurance, etc., and also needs to understand the parameters of the transport object, such as the weight, volume, shape, etc. of the transport object. Based on the above information, the central controller calculates the transport allocation data to clarify the load that each drone should bear in this mission. The transport allocation data can be calculated based on an optimization algorithm. For example, assuming that multiple drones participate in collaborative transport, denoted as U = {u1, u2, ..., un}, and the load capacity of each drone is C i (i=1,2,…,n), the total load (mass of the carried object) is L. The goal of load distribution is to make the load of each UAV as balanced as possible while not exceeding its load capacity. The optimization problem can be expressed as:

[0056]

[0057] l i ≤C i ,i=1,2,…,n

[0058] Among them, l i Indicates the load assigned to the drone ui, is the average load. By solving the above optimization problem, we can get the load distribution data {l1 * ,l2 * ,…,l n *}.

[0059] After determining the transport allocation data, Figure 2 The geometric coordinate relationships between the drones shown determine their control parameters, which guide key actions such as flight attitude, speed, and altitude during the mission. The drones in the figure use the fuselage coordinate system (b1i, b2i, b3i), while the payload uses the Earth coordinate system (e1, e2, e3). l is the rope length, m is the drone's mass, and j is its moment of inertia. Based on the payload assigned to each drone and the coordinate system transformations shown in the figure, it is possible to determine which drone needs to fly at which attitude, speed, and altitude.

[0060] It should be noted that by comprehensively considering the performance parameters of the UAVs and the parameters of the carried objects to determine the transport distribution data and control parameters, the performance advantages of each UAV can be fully utilized, avoiding the situation where some UAVs are overloaded or underloaded due to unreasonable load distribution, thereby improving the efficiency and safety of the entire collaborative transport system.

[0061] In addition, it should be noted that the UAV in this embodiment is a tandem twin-rotor UAV. Of course, the method described in this embodiment can also be applied to any type of UAV participating in collaborative transportation, and the embodiment of the present invention does not make specific limitations.

[0062] 102. When the UAVs perform the collaborative transport mission according to the control parameters, the transport allocation data is updated based on a dynamic load adjustment strategy according to the real-time monitoring data collected from each UAV, and the control parameters are updated according to the updated transport allocation data.

[0063] In this embodiment of the present invention, while UAVs are performing a coordinated transport mission, a central controller continuously collects real-time monitoring data from each UAV. This data includes, but is not limited to, information such as the UAV's load status, flight position, speed, and attitude. Based on this real-time data, the central controller applies a dynamic load adjustment strategy to update the transport allocation data in real time. This dynamic load adjustment strategy flexibly adjusts the load of each UAV based on factors such as the UAV's actual load, flight status, and environmental changes. Subsequently, based on the updated transport allocation data, the control parameters of each UAV are updated again to ensure that the UAVs can continue to perform the mission in accordance with the new load allocation requirements. For example, during a coordinated transport mission, UAV A encounters strong airflow during flight, causing its actual load capacity to decrease and its load status to become abnormal. After detecting this in real time through sensors, the central controller, using the dynamic load adjustment strategy, transfers part of the load originally allocated to UAV A to UAVs B and C. Simultaneously, the control parameters of UAVs A, B, and C are updated to adapt to the new load allocation requirements and continue stable coordinated transport of cargo.

[0064] It should be noted that by collecting UAV monitoring data in real time and updating the transport allocation data and control parameters based on dynamic load adjustment strategies, it is possible to respond promptly to various changes in the mission execution process, such as changes in UAV performance, environmental interference, etc., to ensure that the collaborative transport mission can proceed smoothly and improve the reliability and adaptability of the system.

[0065] 103. While the UAV is flying according to the control parameters and the updated control parameters, each local controller adjusts the single-machine control parameters of the corresponding UAV through at least one of the configured control model, contact compensation controller and load compensation controller.

[0066] In an embodiment of the present invention, while a drone is flying according to the control parameters assigned and updated by a central controller, the local controller onboard each drone will adjust the control parameters of the individual drone based on the actual flight state of the drone using at least one of the configured control model, contact compensation controller, and load compensation controller. The control model is used to implement basic flight control of the drone, ensuring that the drone can fly according to a predetermined trajectory and attitude; the contact compensation controller is used to handle disturbances generated when the drone contacts other drones participating in collaborative transport. By introducing acceleration feedback and force feedback control, it quickly and in real time compensates for disturbances generated during contact, ensuring smooth flight; the load compensation controller is used to respond to sudden load changes that occur during the drone's flight, balancing the impact of load changes by adjusting control parameters to ensure the drone's stable operation.

[0067] It should be noted that by using the control model, contact compensation controller and load compensation controller of each local controller to adjust the control parameters of the single machine, the flight stability of the UAV and its adaptability to various complex situations can be further improved, ensuring that each UAV can maintain the best operating state in the collaborative transportation mission, thereby improving the performance and reliability of the entire collaborative transportation system.

[0068] In one embodiment of the present invention, for further explanation and limitation, adjusting the control parameters of a corresponding UAV using the configured control model includes:

[0069] Generate control laws based on real-time monitoring data, single-machine system parameter estimates, regression vectors, and feedback gain matrices;

[0070] controlling the UAV according to the control law and collecting control output generated by the control law;

[0071] The estimated value of the single-machine system parameter is adaptively adjusted according to the control output and the adaptive gain matrix to achieve single-machine control parameter adjustment.

[0072] The control algorithm based on the control model can be an adaptive control algorithm or a multi-model control algorithm. In the embodiment of the present invention, the control algorithm in the control model is an adaptive control algorithm, that is, the control model is a model constructed based on the feedback gain matrix and the adaptive gain matrix. This algorithm mainly targets the nonlinear characteristics and coupling characteristics of the dual-rotor UAV, so that the adjustment of the control parameters can adapt to the real-time state and environmental changes of the UAV, thereby improving the adaptability and control accuracy of the UAV. The adaptive control algorithm needs to first construct the control law expression, specifically including: assuming that the dynamic model of the UAV is:

[0073]

[0074] Here, x(t) represents the state of the single-machine system, u(t) represents the control input, θ(t) represents the single-machine system parameters, and Δf represents the model uncertainty. The single-machine system state represents the set of key variables of the UAV during operation, such as its position, velocity, and attitude (including pitch, roll, and yaw). These state variables are continuously updated as time t changes, reflecting the real-time motion of the UAV in three-dimensional space and can be determined based on real-time monitoring data. The single-machine system parameters reflect the physical properties of the UAV, such as the mass, length, and inertia of the rotors, as well as parameters related to the flight environment. During actual flight, these parameters may change due to factors such as wear and tear of the UAV and changes in environmental conditions (such as temperature, humidity, and air pressure). The control input is the control signal applied to achieve the desired UAV motion state. For example, the rotational speed of the two rotors is adjusted to control the lift, thrust, and attitude of the UAV.

[0075] According to the dynamic model, the adaptive control law is expressed as:

[0076]

[0077] Where x(t) represents the state of the single-machine system, K represents the feedback gain matrix, Denotes the estimated parameter values ​​of the single-machine system, and φ(x(t)) denotes the regression vector. The estimated parameter values ​​of the single-machine system are estimates of the parameters of the single-machine system. The regression vector contains information related to the state of the single-machine system and is used to link the estimated parameter values ​​with the state of the single-machine system to generate appropriate control inputs.

[0078] In order to make the estimated values ​​of the parameters of the single-machine system gradually approach the actual parameters of the single-machine system, a parameter update law is designed. The parameter update law is expressed as:

[0079]

[0080] Where Γ represents the adaptive gain matrix, and y(t) represents the control output at time t. The adaptive gain matrix determines the adjustment speed and stability of the parameter estimation values ​​of the single-machine system. A larger adaptive gain can make the parameter estimation converge to the true value faster, but it may also cause the parameter estimation to fluctuate greatly; a smaller adaptive gain will make the parameter estimation converge more slowly, but with better stability. Therefore, the adaptive gain matrix can be customized according to specific application requirements, and the embodiments of the present invention do not make specific limitations. The use of adaptive and multi-model control algorithms can automatically adjust parameters according to system status and environmental changes, with high control accuracy; the adaptability to environmental and parameter fluctuations is enhanced, and the robustness is strong.

[0081] In one embodiment of the present invention, for further explanation and limitation, for each local controller, adjusting the control parameters of the corresponding UAV using the configured control model includes:

[0082] Calculate the control laws of different candidate control models based on their feedback gain matrices and real-time monitoring data;

[0083] Controlling the UAV according to the control law, and calculating the deviation between the control output and the control expectation under the control of different candidate control models;

[0084] The candidate control model with the smallest deviation is determined as the target control model, and the control parameters are adjusted according to the target control model.

[0085] In the embodiment of the present invention, there are multiple candidate control models in each local controller, and each candidate control model has its corresponding feedback gain matrix. Assume that the system switches to different control models under different flight phases and mission requirements, which are denoted as M1, M2, ..., M m Each candidate control model has its own corresponding feedback gain matrix. During the flight of the UAV, the local controller will collect the UAV's single-machine system status (such as position, speed, attitude, etc.) in real time and use it as the system input. For each candidate control model, the control input is calculated using the corresponding control law formula based on its feedback gain matrix and real-time monitoring data. Among them, the control law of each control model is expressed as:

[0086] u i (t) = K i x(t),i=1,2,…,m;

[0087] Among them, K i represents the feedback gain matrix of the i-th control model, and x(t) represents the state of the single-machine system.

[0088] After obtaining the control input of each control model, the local controller applies the control input to the UAV respectively, so that the UAV flies according to the corresponding control strategy. During the flight, the actual output of the UAV (such as flight trajectory, attitude, etc.) will be monitored and recorded by the sensor in real time. At the same time, the system will set an expected control output (such as expected flight trajectory, attitude, etc.) according to the mission requirements. For each candidate control model, the deviation between the actual output and the expected output under its control is calculated. For example, let the expected output be y d (t), the actual output is y i (t)(corresponding to M i Model control), the deviation e i (t) = y d (t)-yi (t). Compare the deviation e under the control of different candidate control models i (t)(i=1,2,…,m), select the candidate control model with the smallest deviation as the target control model. The target control model is expressed as:

[0089] u(t)=K σ(t) x(t);

[0090] Wherein, σ(t) represents the switching logic of the control model, which is used to select the current optimal control model. Preferably, the switching logic is expressed as σ(t)=argmin i ‖e i (t)‖. Of course, the switching logic can also be designed using other representations based on the system status, flight phase, or mission requirements, and this is not specifically limited in the present embodiment. The local controller can dynamically select the optimal control model based on the UAV's real-time status and mission requirements, and adjust the UAV's control parameters, thereby improving the UAV's control accuracy and flight stability, ensuring the smooth implementation of the coordinated transport mission.

[0091] In one embodiment of the present invention, for further explanation and limitation, in the event of contact disturbance in any of the UAVs, adjusting the control parameters of the single UAV by the contact compensation controller includes:

[0092] Obtaining output feedback of the UAV, and generating a basic control input through a nominal controller according to the received control expectation and output feedback;

[0093] The output feedback is modeled in real time using Kalman filtering, and a compensation signal is generated through a contact compensation controller based on the modeling results.

[0094] Performing superimposed compensation on the basic control input based on the compensation signal to obtain a compensated control input;

[0095] The control output generated by the compensated control input is used as output feedback, and the process returns to the step of obtaining the output feedback of the UAV to obtain a control input that approximates the control expectation, thereby achieving control parameter adjustment.

[0096] In this embodiment of the present invention, the contact compensation controller is constructed based on the principle of estimated model difference. This principle involves estimating and quantifying the difference between the actual single-vehicle vehicle system model and a pre-set nominal model. The nominal model is a mathematical description of the single-vehicle vehicle system under ideal conditions. It is based on the system's physical properties, dynamics, and other principles, and encompasses the system's expected behavior and parameter relationships under normal operating conditions.

[0097] The specific process of adjusting the control parameters of a single machine through the contact compensation controller includes: Figure 3 As shown in Figure 1, first, the output feedback of the UAV is obtained. This output feedback is used to characterize the actual operating state of the UAV, such as its position, speed, attitude and other key parameters. The nominal controller is the core initial link of the control. It generates the basic control input u based on the received control expectation (i.e. the target state that the UAV is expected to achieve) and the output feedback, using the preset control algorithm. n . This basic control input is the preliminary control instruction given by the nominal controller based on the current known information and conventional control strategy. At the same time, the output feedback is modeled in real time using Kalman filtering. Kalman filtering is an efficient recursive filter that can make the best estimate of the state of the system in the presence of noise. By modeling the output feedback, the changing trends and characteristics of the actual operating state of the drone can be grasped more accurately. The contact compensation controller analyzes the state deviation of the drone under contact disturbance based on the modeling results of the Kalman filter and generates a compensation signal u e , to correct any deviations from the desired state caused by contact disturbances. The compensation signal is then added to the base control input to generate the compensated control input u. Finally, the compensated control input is applied to the drone, and the resulting control output re-enters the system as new output feedback, beginning the next cycle. Through this closed-loop control process, the control input is continuously adjusted and optimized, gradually approaching the desired control state. This allows for the adjustment of control parameters, allowing the drone to maintain stable and accurate operation even in the presence of contact disturbances.

[0098] In one embodiment of the present invention, for further explanation and limitation, when a sudden load change occurs in any of the UAVs, the load compensation controller adjusts the control parameters of the single UAV, including:

[0099] The load compensation controller obtains load change data, force feedback signals and acceleration feedback signals through sensors carried on the drone;

[0100] configuring a force feedback gain coefficient and an acceleration feedback gain coefficient according to the load change data;

[0101] The weight of the force feedback signal in the control loop is adjusted by the force feedback gain coefficient, and the weight of the acceleration feedback signal in the control loop is adjusted by the acceleration feedback gain coefficient, so as to realize the control parameter adjustment of a single UAV in the case of a sudden load change.

[0102] In this embodiment of the present invention, the load compensation controller includes force control feedback and acceleration feedback. In the event of a sudden load change, force feedback and acceleration feedback can quickly respond, effectively suppressing drone oscillation and instability by properly adjusting signal weights. For example, when the load suddenly increases, force feedback can promptly provide sufficient control force to balance the load change, while acceleration feedback can quickly adjust the drone's attitude, preventing the sudden load change from causing the drone to crash or lose control. This significantly enhances the system's stability under complex operating conditions.

[0103] The specific process of adjusting the control parameters of a single machine through the load compensation controller is as follows: Figure 4 As shown in the figure, the process involves: first, using various sensors onboard the drone to accurately acquire load change data, force feedback signals, and acceleration feedback signals. Next, based on the acquired load change data, the force feedback gain coefficient K1 and the acceleration feedback gain coefficient K2 are configured. The force feedback gain coefficients and acceleration feedback gain coefficients can be determined by calculating the load change rate and magnitude through time series analysis. These gain coefficients are then calculated using a gain coefficient algorithm based on the load data. After configuring the gain coefficients, the force feedback gain coefficient is used to adjust the weight of the force feedback signal in the control loop, while the acceleration feedback gain coefficient is used to adjust the weight of the acceleration feedback signal in the control loop. For example, when the load changes suddenly, the force feedback gain coefficient may be appropriately increased to give the force feedback signal a greater role in the control loop and enable faster response to the load change. Conversely, when the drone's attitude changes dramatically, the acceleration feedback gain coefficient may be increased to allow the acceleration feedback signal to better guide the drone's attitude. After these adjustments, the control loop uses the new weightings to comprehensively process the force feedback and acceleration feedback signals, accurately adjusting the control parameters of the individual drone. The adjusted control parameters cause the drone to generate corresponding control actions, thus maintaining stable flight under sudden load changes. This control output then serves as new feedback information, re-entering the control loop, forming a continuously optimized closed-loop control process that continuously approaches the ideal control state.

[0104] In one embodiment of the present invention, for further explanation and limitation, updating the transport allocation data based on the dynamic load adjustment strategy according to the real-time monitoring data collected from each drone includes:

[0105] For each UAV, respectively adjust the proportional control gain and the differential control gain based on the difference between the real-time load and the target load in the transport allocation data to obtain a load adjustment for each UAV;

[0106] For each UAV, the sum of the load adjustment amount and the real-time load is used as the adjusted load, and the adjusted load of each UAV is used to update the respective target loads in the transport distribution data to obtain updated transport distribution data.

[0107] In the embodiment of the present invention, the real-time monitoring data includes real-time load. During the collaborative transport process, the load status and flight parameters of each UAV are monitored in real time by sensors. i The real-time load of the aircraft is yi(t), and its flight parameters include position, speed, and attitude. The central controller dynamically adjusts the load distribution plan based on the monitoring data. The adjustment strategy can be expressed as:

[0108]

[0109] Where, Δl i (t) represents the load adjustment of the i-th UAV at time t, k p represents the proportional control gain, k d Denotes the differential control gain, k p and k d Used to adjust the speed and stability of load distribution; i * represents the target load of the i-th UAV, which can also be the initially assigned load. The adjusted load can be calculated according to the following formula:

[0110] y i (t+Δt)=y i (t)+Δl i (t);

[0111] Among them, y i (t+Δt) represents the load of the i-th UAV at time t+Δt. The above formula can be used to calculate the adjusted load of each UAV involved in the transport, and then the updated transport allocation data can be obtained based on this adjusted load and flight parameters.

[0112] In one embodiment of the present invention, for further illustration and limitation, the method further includes:

[0113] For each UAV, calculating a flight deviation based on the flight parameters;

[0114] If the flight deviation is greater than a preset deviation threshold, the UAV is determined to be a faulty UAV;

[0115] When it is determined that the fault level of the UAV is a mild fault based on the comparison result of the key flight parameters of the faulty UAV with the corresponding threshold value, the control parameters of the UAV are adjusted according to the fault compensation gain matrix;

[0116] When it is determined that the fault level is a serious fault, the faulty UAV is guided to return to the base by sending a control instruction, and the transport allocation data is re-determined based on the performance parameters and transport object parameters of the remaining non-faulty UAVs.

[0117] In the embodiment of the present invention, the flight deviation is used to represent the difference between the observed value based on the flight parameter and the predicted value based on the flight state. Specifically, the flight deviation can be calculated based on the fault detection model, which is expressed as:

[0118] r i (t)=Hz i (t)-Cx i (t);

[0119] Among them, H represents the observation matrix, C represents the state matrix, r i (t) represents the flight deviation of the i-th UAV, z i (t) represents the flight parameters of the i-th UAV, x i (t) represents the single-machine system status of the i-th UAV. When the flight deviation is greater than the preset deviation threshold, it indicates that the current UAV's flight status is unstable and can be determined to be a faulty UAV. Faults can be divided into mild faults and severe faults. In the case of mild faults, the control parameters of the UAV can be adjusted based on the fault compensation gain matrix. The formula for adjusting the control parameters using the fault compensation gain matrix can be expressed as:

[0120]

[0121] Among them, u i 0 (t) represents the control input under normal conditions, K f is the fault compensation gain matrix.

[0122] In the event of a serious failure, the current drone must continue its mission. Therefore, it is necessary to promptly guide the faulty drone back to base and transfer its payload to other functioning drones. Load transfer redistributes the payload, treating the remaining undamaged drones as all participating drones. The specific process is identical to step 101 and will not be further described in detail in this embodiment of the present invention.

[0123] In an application example, the overall design concept of the above-mentioned UAV cooperative transport control method is as follows: Figure 5As shown in the figure. To enable any single UAV involved in the transport to possess cooperative transport capabilities, the first step is to design adaptive control parameters for the single UAV and parameter control adjustments to cope with sudden load changes, corresponding to the first UAV schematic in the figure. Next, the single UAV's ability to resist contact disturbances is designed to achieve a smooth transition from single flight to multi-UAV coordinated flight, corresponding to the second UAV schematic in the figure. Finally, the load distribution design for each UAV during multi-UAV coordinated transport is considered, corresponding to the third multi-UAV transport schematic in the figure, thus completing the overall design of UAV coordinated transport control.

[0124] The present invention provides a method for controlling cooperative transport of unmanned aerial vehicles (UAVs). In an embodiment of the present invention, a central controller determines transport allocation data based on performance parameters and transport object parameters of each UAV, and determines control parameters for each UAV based on the transport allocation data. The central controller updates the transport allocation data based on a dynamic load adjustment strategy and updates the control parameters based on the updated transport allocation data while the UAVs perform the cooperative transport mission according to the control parameters, based on real-time monitoring data collected from each UAV. The control parameters are then updated according to the updated transport allocation data. While the UAVs fly according to the control parameters and the updated control parameters, each local controller adjusts the control parameters of its corresponding UAV using at least one of the configured control model, contact compensation controller, and load compensation controller. The local control parameter adjustment ensures the timeliness of the UAV's ability to counteract disturbances and sudden load changes, and the accuracy of control parameter adjustment. Furthermore, the central controller performs transport allocation to achieve load balancing among multiple UAVs during the cooperative transport process, avoiding degradation of flight performance and safety hazards caused by uneven loads, thereby greatly improving control accuracy.

[0125] Furthermore, as a response to the above Figure 1 The embodiment of the present invention provides a UAV cooperative transport control device, such as Figure 6 As shown, the device includes:

[0126] The transport allocation module 31 is configured to determine transport allocation data based on the performance parameters and transport object parameters of each UAV based on the central controller, and determine control parameters of each UAV based on the transport allocation data;

[0127] a load allocation adjustment module 32 configured to update the load allocation data based on a dynamic load adjustment strategy based on real-time monitoring data collected from each UAV during the process of the UAVs performing the coordinated load mission according to the control parameters, and to update the control parameters based on the updated load allocation data;

[0128] The stand-alone control parameter adjustment module 33 is used to adjust the stand-alone control parameters of the corresponding UAV based on the local controller during the flight of the UAV according to the control parameters and the updated control parameters, through at least one of the configured control model, contact compensation controller and load compensation controller.

[0129] Furthermore, the single machine control parameter adjustment module 33 includes:

[0130] A first generating unit is used to generate a control law based on real-time monitoring data, estimated values ​​of single-machine system parameters, regression vectors and feedback gain matrices;

[0131] a control unit, configured to control the UAV according to the control law and collect control outputs generated by the control law;

[0132] The first adjustment unit is used to adaptively adjust the estimated value of the single-machine system parameter according to the control output and the adaptive gain matrix to obtain an updated control parameter.

[0133] Furthermore, the single machine control parameter adjustment module 33 further includes:

[0134] A first calculation unit is used to calculate the control laws of different candidate control models according to the feedback gain matrix and real-time monitoring data of each candidate control model;

[0135] a second calculation unit, configured to control the UAV according to the control law and calculate a deviation between a control output and a control expectation under control of different candidate control models;

[0136] The second adjustment unit is used to determine the candidate control model with the smallest deviation as the target control model, and adjust the control parameters according to the target control model.

[0137] Furthermore, the single machine control parameter adjustment module 33 further includes:

[0138] A second generating unit is configured to obtain output feedback of the UAV and generate a basic control input according to the received control expectation and output feedback through a nominal controller;

[0139] A modeling unit, configured to perform real-time modeling of the output feedback using a Kalman filter, and generate a compensation signal through a contact compensation controller based on the modeling result;

[0140] a compensation unit, configured to perform superimposed compensation on the basic control input based on the compensation signal to obtain a compensated control input;

[0141] The third adjustment unit is used to use the control output generated by the compensated control input as output feedback, return to the step of obtaining the output feedback of the drone, so as to obtain a control input that approximates the control expectation and realize control parameter adjustment.

[0142] Furthermore, the single machine control parameter adjustment module 33 further includes:

[0143] An acquisition unit, configured for the load compensation controller to acquire load change data, force feedback signals, and acceleration feedback signals through sensors carried on the drone;

[0144] a configuration unit, configured to configure a force feedback gain coefficient and an acceleration feedback gain coefficient according to the load change data;

[0145] The fourth adjustment unit is used to adjust the weight of the force feedback signal in the control loop through the force feedback gain coefficient, and to adjust the weight of the acceleration feedback signal in the control loop through the acceleration feedback gain coefficient, so as to realize the control parameter adjustment of the single UAV in the case of sudden load change.

[0146] Furthermore, the transport distribution adjustment module 32 includes:

[0147] For each UAV, respectively adjust the proportional control gain and the differential control gain based on the difference between the real-time load and the target load in the transport allocation data to obtain a load adjustment for each UAV;

[0148] For each UAV, the sum of the load adjustment amount and the real-time load is used as the adjusted load, and the adjusted load of each UAV is used to update the respective target loads in the transport distribution data to obtain updated transport distribution data.

[0149] Furthermore, the device further comprises:

[0150] A flight deviation calculation module, configured to calculate a flight deviation for each UAV based on the flight parameters, wherein the flight deviation is used to represent a difference between an observed value based on the flight parameters and a predicted value based on the flight state;

[0151] a fault detection module, configured to determine that the UAV is a faulty UAV if the flight deviation is greater than a preset deviation threshold;

[0152] a first fault response module, configured to adjust control parameters of the UAV according to a fault compensation gain matrix when the fault level of the UAV is determined to be a minor fault based on a comparison result of the key flight parameters of the faulty UAV with corresponding thresholds;

[0153] The second fault response module is used to guide the faulty UAV to return to the base by sending a control instruction when the fault level is determined to be a serious fault, and to redetermine the carrying allocation data based on the performance parameters and carrying object parameters of the remaining non-faulty UAVs.

[0154] The present invention provides a cooperative transport control device for unmanned aerial vehicles. In an embodiment of the present invention, a central controller determines transport allocation data based on performance parameters and transport object parameters of each unmanned aerial vehicle, and determines control parameters for each unmanned aerial vehicle based on the transport allocation data. The central controller updates the transport allocation data based on a dynamic load adjustment strategy and updates the control parameters based on the updated transport allocation data while the unmanned aerial vehicles perform cooperative transport tasks according to the control parameters, based on real-time monitoring data collected from each unmanned aerial vehicle. During flight according to the control parameters and the updated control parameters, each local controller adjusts the control parameters of its corresponding unmanned aerial vehicle using at least one of the configured control model, contact compensation controller, and load compensation controller. The local control parameter adjustment ensures the timeliness of the unmanned aerial vehicle's ability to resist disturbances and sudden load changes, and the accuracy of control parameter adjustment. At the same time, transport allocation by the central controller achieves balanced load distribution among multiple unmanned aerial vehicles during cooperative transport, avoiding degradation of flight performance and safety hazards caused by uneven loads, thereby greatly improving control accuracy.

[0155] According to one embodiment of the present invention, a storage medium is provided, which stores at least one executable instruction. The computer-executable instruction can execute the UAV cooperative transportation control method in any of the above method embodiments.

[0156] According to one embodiment of the present invention, a drone system is provided, the system comprising a plurality of drones and a central controller, each drone being equipped with a local controller, each local controller being respectively communicatively connected to the central controller;

[0157] The central controller is used to perform operations of carrying allocation of multiple UAVs and updating carrying allocation data;

[0158] Any of the local controllers is used to perform operations for controlling parameters of a single UAV.

[0159] Obviously, those skilled in the art will appreciate that the various modules or steps of the present invention described above can be implemented using a general-purpose computing device, centralized on a single computing device, or distributed across a network of multiple computing devices. Alternatively, they can be implemented using program code executable by a computing device, which can then be stored in a storage device and executed by the computing device. In some cases, the steps shown or described can be performed in a different order than that shown, or can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the present invention is not limited to any particular combination of hardware and software.

[0160] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A UAV cooperative transport control method, characterized in that: A transport mission is performed collaboratively by multiple drones, each equipped with a local controller, each of which is in communication with a central controller. The central controller is configured with a dynamic load adjustment strategy; the local controllers are configured with a control model for single-machine control, a contact compensation controller for dealing with contact disturbances, and a load compensation controller for dealing with sudden load changes. The method includes: Determining, by the central controller, transport allocation data based on the performance parameters and transport object parameters of each UAV, and determining control parameters of each UAV based on the transport allocation data; During the process of the UAVs performing the coordinated transport mission according to the control parameters, the central controller updates the transport allocation data based on a dynamic load adjustment strategy according to the real-time monitoring data collected from each UAV, and updates the control parameters according to the updated transport allocation data; During the flight of the UAV according to the control parameters and the updated control parameters, each local controller adjusts the single-machine control parameters of the corresponding UAV based on at least one of the configured control model, contact compensation controller and load compensation controller.

2. The method according to claim 1, characterized in that The control model is constructed based on the feedback gain matrix and the adaptive gain matrix. For each local controller, the control parameters of the corresponding UAV are adjusted through the configured control model, including: Generate control laws based on real-time monitoring data, single-machine system parameter estimates, regression vectors, and feedback gain matrices; controlling the UAV according to the control law and collecting control output generated by the control law; Adaptively adjust the estimated values ​​of the single-machine system parameters according to the control output and the adaptive gain matrix to obtain updated control parameters.

3. The method according to claim 1, characterized in that The control model is constructed based on multiple candidate control models. For each local controller, the control parameters of the corresponding UAV are adjusted by the configured control model, including: Calculate the control laws of different candidate control models based on their feedback gain matrices and real-time monitoring data; Controlling the UAV according to the control law, and calculating the deviation between the control output and the control expectation under the control of different candidate control models; The candidate control model with the smallest deviation is determined as the target control model, and the control parameters are adjusted according to the target control model.

4. The method according to claim 1, wherein The contact compensation controller is constructed based on the principle of estimated model difference. When any of the UAVs has a contact disturbance, the contact compensation controller is used to adjust the control parameters of the single UAV, including: Obtaining output feedback of the UAV, and generating a basic control input through a nominal controller according to the received control expectation and output feedback; The output feedback is modeled in real time using Kalman filtering, and a compensation signal is generated through a contact compensation controller based on the modeling results. Performing superimposed compensation on the basic control input based on the compensation signal to obtain a compensated control input; The control output generated by the compensated control input is used as output feedback, and the process returns to the step of obtaining the output feedback of the UAV to obtain a control input that approximates the control expectation, thereby achieving control parameter adjustment.

5. The method according to claim 1, characterized in that The load compensation controller includes force control feedback and acceleration feedback. When a sudden load change occurs on any of the UAVs, the load compensation controller is used to adjust the control parameters of the single UAV, including: The load compensation controller obtains load change data, force feedback signals and acceleration feedback signals through sensors carried on the drone; configuring a force feedback gain coefficient and an acceleration feedback gain coefficient according to the load change data; The weight of the force feedback signal in the control loop is adjusted by the force feedback gain coefficient, and the weight of the acceleration feedback signal in the control loop is adjusted by the acceleration feedback gain coefficient, so as to realize the control parameter adjustment of a single UAV in the case of a sudden load change.

6. The method according to claim 1, characterized in that The real-time monitoring data includes real-time load, and the updating of the transport allocation data based on the dynamic load adjustment strategy according to the real-time monitoring data collected from each UAV includes: For each UAV, respectively adjust the proportional control gain and the differential control gain based on the difference between the real-time load and the target load in the transport allocation data to obtain a load adjustment for each UAV; For each UAV, the sum of the load adjustment amount and the real-time load is used as the adjusted load, and the adjusted load of each UAV is used to update the respective target loads in the transport distribution data to obtain updated transport distribution data.

7. The method according to claim 6, characterized in that The real-time monitoring data also includes flight parameters, and the method further includes: For each UAV, calculating a flight deviation based on the flight parameters, wherein the flight deviation is used to represent a difference between an observed value based on the flight parameters and a predicted value based on the flight state; If the flight deviation is greater than a preset deviation threshold, the UAV is determined to be a faulty UAV; When it is determined that the fault level of the UAV is a mild fault based on the comparison result of the key flight parameters of the faulty UAV with the corresponding threshold value, the control parameters of the UAV are adjusted according to the fault compensation gain matrix; When it is determined that the fault level is a serious fault, the faulty UAV is guided to return to the base by sending a control instruction, and the transport allocation data is re-determined based on the performance parameters and transport object parameters of the remaining non-faulty UAVs.

8. A UAV cooperative transport control device, characterized in that: The transport mission is performed collaboratively by multiple drones, each equipped with a local controller, each of which is in communication with a central controller. The central controller is configured with a dynamic load adjustment strategy; the local controllers are configured with a control model for single-machine control, a contact compensation controller for dealing with contact disturbances, and a load compensation controller for dealing with sudden load changes. The device includes: A transport allocation module, configured to determine transport allocation data based on the performance parameters and transport object parameters of each UAV through a central controller, and determine control parameters of each UAV based on the transport allocation data; a load allocation adjustment module, configured to update the load allocation data based on a dynamic load adjustment strategy according to real-time monitoring data collected from each drone, and to update the control parameters according to the updated load allocation data, while the drones are performing the coordinated load mission according to the control parameters; The stand-alone control parameter adjustment module is used to adjust the stand-alone control parameters of the corresponding UAVs through each local controller based on at least one of the configured control model, contact compensation controller and load compensation controller during the flight of the UAV according to the control parameters and the updated control parameters.

9. A storage medium, characterized in that: The storage medium stores at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the UAV cooperative transportation control method according to any one of claims 1 to 7.

10. A drone system, characterized in that: It includes multiple drones and a central controller. Each drone is equipped with a local controller, and each local controller is connected to the central controller in communication. The central controller is used to perform operations corresponding to the UAV cooperative transportation control method according to any one of claims 1, 6, and 7; Any of the local controllers is used to perform operations corresponding to the UAV collaborative transportation control method according to any one of claims 1 to 5.

Images