A Fault Recovery Control Method for Multi-Aircraft Hanging Load Formation

Abstract

The present invention relates to the field of aircraft control, and discloses a multi-aircraft suspended load formation fault recovery control method, which includes the following steps: During flight, if a certain aircraft fails, it is immediately removed, and the tether tensions are updated; the remaining effective aircraft are re-formed into a regular polygon, new relative position deviation commands are assigned, and together with the updated tether tensions, they are input into an improved finite-time outer-loop consensus control protocol to generate control components; the position channel control quantity and the desired attitude angle are inversely solved; the desired attitude angle is input into an improved sliding mode attitude controller in the inner loop to generate the attitude channel control quantity; the position and attitude channel control quantities are used to adjust the position and attitude of the aircraft in real time. The method disclosed by the present invention can quickly reconstruct the formation structure, accurately compensate for the change in load tension, maintain the stable flight of the suspended load, and significantly improve the robustness of the system and the task completion rate in the multi-aircraft suspended load formation task after the failure of an individual aircraft.

Description

Technical Field

[0001] The present invention relates to the field of aircraft control, and in particular to a multi-aircraft suspension load formation failure recovery control method. Background Art

[0002] In recent years, multi-rotor aerial vehicles (UAVs) have been widely used in search and rescue, disaster monitoring, logistics and transportation, and other fields due to their simple structure, strong maneuverability, and flexible control. Multiple drones forming a formation to jointly perform load-lifting and transportation tasks has become an efficient and safe operation method, particularly suitable for complex terrain conditions such as outdoor and mountainous areas. However, in actual applications, due to factors such as environmental disturbances, high temperatures, thick smoke, and motor failures, some drones may fail, resulting in formation imbalance, payload trajectory deviation, and even collisions, seriously threatening the safety and stability of the mission. Therefore, how to quickly reconstruct the formation and maintain a stable payload transportation trajectory in the event of a single drone failure has become a key technical problem that needs to be solved in this field.

[0003] Currently, research has proposed fault-tolerant control schemes for multi-vehicle coordinated transport missions. For example, after a failure, the remaining vehicles redistribute tasks through master-slave communication, and the master vehicle adjusts its formation to maintain load balance. Control inputs to the remaining vehicles are optimized online to ensure load trajectory tracking, and MPC (Model Predictive Control) is used to dynamically adjust the hovering point after a failure. However, these approaches are computationally intensive and have limited practical application. Dynamic MPC reconstruction can cause load trajectory deviations due to computational delays. These approaches, when applied to high-dimensional nonlinear systems and nonholonomic constraints, generally suffer from complex modeling, difficult solutions, and low control accuracy, making them difficult to meet real-time and robustness requirements.

[0004] Sliding mode control is widely used in aircraft attitude control due to its robustness and rapid response. However, traditional sliding mode control is prone to chattering during operation, which affects system stability. Although researchers have attempted to improve this problem by introducing nonlinear sliding surfaces or improving reaching laws, the convergence rate remains limited and it is difficult to adapt to the complex coordinated control requirements of multi-aircraft systems.

[0005] Mainstream approaches to multi-vehicle formation control are based on consensus theory, typically using leader-follower structures and virtual leader strategies. While these methods enable coordinated tracking and path planning, traditional consensus protocols often have slow convergence and delayed formation reconfiguration in the face of vehicle failures or changes in communication topology, making it difficult to achieve system stability within a limited timeframe.

[0006] In summary, the existing technology has the following major shortcomings when dealing with individual failures in the multi-rotor aircraft load suspension system:

[0007] (1) The modeling method is complex and difficult to handle non-holonomic constraint systems, resulting in low solution efficiency and poor real-time performance;

[0008] (2) The attitude control system is prone to chattering or slow response, making it difficult to quickly stabilize the load;

[0009] (3) Traditional consensus control protocols have slow convergence speed and low fleet reconstruction efficiency, and cannot meet the rapid response requirements under sudden failures.

[0010] Therefore, there is an urgent need for a new fault recovery mechanism that can achieve high-precision modeling, fast and stable control, and finite-time formation reconstruction to improve the robustness and mission completion efficiency of the multi-aircraft suspension system. Summary of the Invention

[0011] In order to solve the above technical problems, the present invention provides a multi-aircraft suspended load formation fault recovery control method, so as to achieve the purpose of quickly reconstructing the formation structure after the failure of individual aircraft, accurately compensating for the load tension change, maintaining the stable flight of the suspended load, and significantly improving the system robustness and mission completion rate.

[0012] To achieve the above object, the technical solution of the present invention is as follows:

[0013] A multi-aircraft hanging load formation failure recovery control method includes the following steps:

[0014] Step 1: During a mission with multiple drones flying the same payload, if a drone fails, the drone is immediately removed from the control system, its physical connection to the payload is released, and the tether tension between each drone and the payload is updated.

[0015] Step 2: Reorganize the remaining valid aircraft into a regular polygon, establish new azimuths, assign new relative position deviation instructions, and input these instructions, along with the updated tether tensions, into the improved finite-time outer-loop consistency control protocol to generate control components that enable each aircraft to reach the specified position.

[0016] Step three: Using the aircraft model, the position channel control quantity and the desired attitude angle are inversely solved through the control components; the desired attitude angle is input into the inner loop improved sliding mode attitude controller to generate the attitude channel control quantity; the position channel control quantity and the attitude channel control quantity are applied to the aircraft to adjust the position and attitude of the aircraft in real time.

[0017] In the above scheme, in step one, an electronically controlled disconnect device is pre-installed at the hook connecting the aircraft and the load. When the force sensor detects that a certain aircraft has failed, the control system sends a release command to the electronically controlled disconnect device, automatically opens the hook, and disconnects the physical connection.

[0018] In the above solution, in step one, the Udwadia-Kalaba equation is used to calculate the total tether tension:

[0019] ;

[0020] In the formula, represents the direct product of the system mass matrix and the identity matrix, , represents the th vehicle mass, , is the number of vehicles; represents the payload mass; represents the kinematic constraint matrix, where ; represents the tether length error, defined as , is the length of the tether on the th vehicle, represents the position vector difference between the th vehicle and the payload, and are the position vectors of the th vehicle and the payload respectively, is the first derivative of the length error; represents the constraint acceleration term, and , is the first derivative; represents the Moore-Penrose pseudoinverse; and are feedback control gain coefficients, both positive constants; represents the acceleration vector, that is , and are the acceleration vectors of the th vehicle and the payload respectively;

[0021] The total tether tension is calculated by the above formula, where, is a matrix, expressed as follows: , where is the tension on the tether on the th vehicle, is the direction vector along the tether from the payload to the th vehicle. The first rows in this matrix represent the tensions acting on the th vehicle respectively, and the th row represents the vector sum of all tensions; When the number of vehicles After the change, the corresponding also changes, and each component of it also changes.

[0022] In the above solution, in step two, the method of re - forming the remaining effective aircraft into a regular polygon, establishing a new azimuth angle, and distributing new relative position deviation instructions is as follows:

[0023] To ensure the uniform distribution of the load force, the aircraft formation adopts a regular polygon structure, and the load is regarded as the virtual leader of the formation. is the number of aircraft; the coordinates of the th aircraft are:

[0024] ;

[0025] Among them, is the position quantity of the virtual leader, represents the azimuth angle of the th aircraft, is the tether length on the th aircraft, is for the th aircraft's distance from the formation center:

[0026] ;

[0027] When the th aircraft fails, the failed aircraft with the original number is directly removed, and the remaining aircraft are re - numbered as according to their order in the communication topology; the new number of formation sides is , and at the same time, the azimuth angles of the remaining aircraft need to be redistributed:

[0028] ;

[0029] Among them, represents the new azimuth angle of the th aircraft;

[0030] Therefore, the relative coordinate components between the th aircraft and the virtual leader are respectively , , ;

[0031] The horizontal and vertical distances between two aircraft and are:

[0032] ;

[0033] In the load suspension system, it is assumed that all aircraft have the same altitude, so ;

[0034] The newly obtained relative position deviation terms , , , , , are applied, together with the updated tether tension, to the improved finite-time outer-loop consensus control protocol.

[0035] In the above scheme, in step two, the improved finite-time outer-loop consensus control protocol is as follows:

[0036] ;

[0037] where is the control quantity for the -th aircraft obtained from the outer-loop consensus control protocol, is the position control gain between aircraft, is the position control gain between each aircraft and the virtual leader, is the velocity control gain between aircraft, is the velocity control gain between each aircraft and the virtual leader; is the number of aircraft; is the adjacency matrix The element in represents the communication topology between aircraft. If aircraft can receive the state information of aircraft , then , otherwise, ; , ; is the acceleration of the virtual leader, is the velocity of the virtual leader, is the position of the virtual leader, and are the positions of aircraft and respectively, and are the velocities of aircraft and respectively, is the relative position deviation between aircraft and in the formation shape, is the relative position deviation between aircraft and the virtual leader in the formation shape, is the tension exerted by the load on each tether on the th aircraft, , is the gain parameter, is the tension on the tether on the th aircraft, , is a judgment variable for whether each aircraft can sense the position and velocity information of the virtual leader, that is, whether communication is possible, ;

[0038] Calculated from the above formula , , , , are the , , components in the direction of the control quantity.

[0039] In the above solution, in step three, when the aircraft is a quadrotor aircraft, the inverse solution formula is obtained according to the quadrotor aircraft model, and the expected pitch angle and expected roll angle are obtained by inverting the control components , , :

[0040] ;

[0041] Among them, is the control quantity of the quadrotor aircraft position channel and can be directly applied to the individual aircraft; is the expected pitch angle of the th quadrotor aircraft, is the expected roll angle of the th quadrotor aircraft, is the expected yaw angle of the th quadrotor aircraft, and the obtained expected angles are used as inputs to the sliding mode attitude controller.

[0042] In the above solution, in step three, the sliding mode attitude controller adopts an improved sliding mode reaching law, and its form is as follows:

[0043] ;

[0044] Among them, is the sliding mode variable, is the convergence speed term of the sliding mode variable; is the sign function, , , , are the reaching law adjustment parameters, and it is required that , , , ;

[0045] Design a sliding mode controller for the attitude channel of the quadrotor; set the attitude error as:

[0046] ;

[0047] wherein, represents the error of the attitude angle, represents the roll, pitch, and yaw attitude angles of the quadrotor, represents the expected values of the roll, pitch, and yaw attitude angles of the quadrotor;

[0048] Define a sliding mode surface based on PD regulation:

[0049] ;

[0050] wherein, represents the sliding mode surface of each attitude channel, , are the proportional regulation parameter and the differential regulation parameter respectively, represents the first derivative of the attitude angle error. According to the quadrotor model, the attitude controller is designed as:

[0051] ;

[0052] In the formula, , is the first derivative of the attitude angle error, , respectively represent the first derivatives of the roll, pitch, and yaw attitude angles of the th quadrotor, respectively represent the second derivatives of the expected angles of the roll, pitch, and yaw attitude angles of the th quadrotor; , , are the sliding mode surfaces of the three attitude channels; , are the proportional regulation parameter and the differential regulation parameter respectively, , is the sign function, , , , are the reaching law regulation parameters, and it is required that , , , ; Represents the first-order derivative of the attitude angle error; is the disturbance coefficient of the quadrotor model, , and For quadrotor aircraft axis, axis, The moment of inertia of the shaft;

[0053] The attitude channel control quantity obtained , , It is applied to each quadrotor to quickly switch the attitude angle to achieve position tracking.

[0054] In the above scheme, the quadcopter model is as follows:

[0055] ;

[0056] In the formula, , , are the components of the quadrotor position channel, , , are the components of the quadrotor attitude channel, , , is the first-order derivative of the quadrotor position channel component, , , is the second-order derivative of the quadrotor position channel component, , , is the first-order derivative of the quadrotor attitude channel component, , , is the second-order derivative of the quadrotor attitude channel component, , and For quadrotor aircraft axis, axis, The moment of inertia of the shaft, is the acceleration due to gravity, is the mass of the quadrotor, is the disturbance coefficient; , , The load acts on the quadrotor of tether tension in three directions; is the control quantity of the position channel, , , are respectively the of the , , attitude channel control quantities of the quadrotor aircraft; where represents the number of quadrotor aircraft;

[0057] Input the control quantity of the position channel, the control quantity , , of the attitude channel, and the tether tension into the quadrotor aircraft model to adjust the position and attitude of the aircraft.

[0058] Through the above technical solutions, a multi-aircraft suspended load formation fault recovery control method provided by the present invention has the following beneficial effects:

[0059] (1) The invention introduces an improved finite-time consensus protocol, which can complete the formation structure reorganization in a short time. Compared with the traditional consensus protocol, the system robustness and convergence speed are significantly improved;

[0060] (2) The present invention accurately calculates the constraint tension based on the Udwadia-Kalaba equation, ensures the smooth continuity of the load trajectory, and improves the task stability;

[0061] (3) The present invention uses an improved reaching law sliding mode control to reduce the system chattering and effectively improve the control performance under the conditions of interference or parameter uncertainty;

[0062] (4) The control system in the present invention has the ability to maintain the stable operation of the system even when individual aircraft randomly fail in the multi-aircraft cooperative transportation task, meeting the actual application requirements in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art.

[0064] Figure 1 is a schematic diagram of the framework of a multi-aircraft suspended load formation fault recovery control method disclosed in an embodiment of the present invention;

[0065] Figure 2 is a schematic diagram of a single quadrotor aircraft;

[0066] Figure 3A diagram of the multi-aircraft formation operating with a payload, showing how, even with the damage of an individual aircraft, the remaining valid aircraft form a formation and continue to operate with the payload.

[0067] Figure 4 Schematic diagram of angle change during aircraft formation switching. DETAILED DESCRIPTION

[0068] The technical solutions in the embodiments of the present invention will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present invention.

[0069] This invention aims to address technical issues in existing multi-vehicle payload suspension systems, such as slow formation reconfiguration response, large payload trajectory deviations, and severe attitude control chattering, which can occur in the event of individual failures. This paper proposes a multi-vehicle formation failure recovery mechanism based on an improved finite-time consensus protocol and an improved reaching-law sliding mode control. This mechanism enables rapid formation reconfiguration after individual vehicle failures, accurately compensates for changes in payload tension, maintains stable flight of suspended payloads, and significantly improves system robustness and mission completion rate.

[0070] The modeling method and consistency control protocol in the present invention are applicable to multi-aircraft hanging tasks, and have the dual advantages of accurate modeling and rapid coordination. They are particularly suitable for emergency material delivery, coordinated fire fighting and special environment operations.

[0071] During flight, this invention integrates onboard tension sensors with system status information to monitor the aircraft's operating status in real time. If a fault (such as a broken cable or motor stall) is detected, the system immediately implements the following inner and outer loop control strategies within the established finite-time consistency control protocol framework. Figure 1 This is the control block diagram of the entire operating system. The outer loop of each aircraft uses an improved finite time consistency control protocol to control the position and speed of each aircraft, and the control quantity obtained is Perform inverse analysis to obtain the desired attitude angle and position dynamics control quantity of each inner loop attitude control ,The inner loop attitude control of each aircraft adopts an improved sliding mode controller, ,which ensures that the aircraft attitude can follow up quickly and achieve ,rapid response of the aircraft.

[0072] The present invention provides a multi-aircraft formation failure recovery control method. This embodiment uses a quadrotor aircraft as an example to illustrate the method, including the following steps:

[0073] Step 1: During a flight mission with multiple aircraft carrying the same payload, if a failure of an aircraft is detected, the failed aircraft is immediately removed from the control system, its physical connection with the payload is released, and the tether tension between each aircraft and the payload is updated.

[0074] Figure 2 Shown is the schematic diagram of a single quadrotor aircraft. Taking the quadrotor aircraft with an "X" - shaped structure as an example. To clearly describe the basic principle of the quadrotor aircraft, two reference coordinate systems are defined: the inertial coordinate system and the body - fixed coordinate system . The attitude of the quadrotor aircraft is represented by Euler angles , which are respectively represented as the roll angle, pitch angle and yaw angle.

[0075] At the hook where the aircraft is connected to the load, an electrically - controlled disconnection device is pre - installed. When the force sensor detects that a certain aircraft has a fault, the control system sends a release instruction to the electrically - controlled disconnection device to automatically open the hook and disconnect the physical connection. Figure 3 It is a diagram of the process of multiple aircraft in formation suspending and carrying a load. Along with the damage of individual aircraft, it is a schematic diagram of the remaining effective aircraft forming a formation and continuing to operate. During operation, the expected motion trajectory of the load object is always used as the operation trajectory of the entire system, and the load is regarded as the virtual leader of the aircraft formation.

[0076] When multiple quadrotor aircraft are connected to the same load, the constraint force will change dynamically with the multi - body interaction between the aircraft in the system. The Udwadia - Kalaba equation is used to calculate the total tether tension:

[0077] ;

[0078] In the formula, represents the direct product of the system mass matrix and the identity matrix, , represents the mass of the th aircraft, , is the number of aircraft; represents the load mass; represents the kinematic constraint matrix, where ; represents the tether length error, defined as , is the length of the tether on the th aircraft, represents the position vector difference between the th aircraft and the load, and are respectively the position vectors of the th aircraft and the load, is the first - order derivative of the length error; represents the constraint acceleration term, and , is The first derivative of ; express Moore-Penrose pseudoinverse; and is the feedback control gain coefficient, which is a positive constant; represents the acceleration vector, that is , and Respectively The acceleration vectors of the vehicle and payload;

[0079] The total tether tension is calculated using the above formula, where: is a matrix, represented as follows: ,in For the The tension on the tether of the aircraft, For the load to The direction vector of the tether of the spacecraft, The rows represent the effects on The tension on the aircraft, The row represents the vector sum of all the pulling forces; when the number of aircraft After the change, the corresponding It also changes accordingly, and each of its components also changes.

[0080] In step 2, the remaining valid aircraft are rearranged into a regular polygon, new azimuths are established, and new relative position deviation instructions are assigned. These instructions, together with the updated tether tension, are input into the improved finite-time outer-loop consistency control protocol to generate control components that enable each aircraft to reach the specified position.

[0081] The method for re-forming the remaining valid aircraft into a regular polygon, establishing new azimuths, and assigning new relative position instructions is as follows:

[0082] In order to ensure the uniform distribution of load, the aircraft formation adopts the polygonal structure and consider the load as a virtual leader of the formation, is the number of aircraft; The coordinates of the aircraft are:

[0083] ;

[0084] in, is the position of the virtual leader, Indicates The azimuth of the aircraft, It is The length of the tether on the aircraft, For the Distance between an aircraft and the formation center:

[0085] ;

[0086] When the th aircraft fails, the failed aircraft with the original number is directly removed, and the remaining aircraft are re-numbered as according to their order in the communication topology; the new number of formation sides is , and at the same time, it is necessary to re-distribute the azimuth angles of the remaining aircraft:

[0087] ;

[0088] Among them, represents the new azimuth angle of the th aircraft;

[0089] Figure 4 Figure

[0090] is a schematic diagram of the angle transformation during the formation switching of the aircraft. The new angles are used to determine the distances between the aircraft and the virtual leader in the new formation, ensuring a smooth transition of the formation. Therefore, the relative coordinate components between the , , th aircraft and the virtual leader are respectively

[0091] The horizontal and vertical distances between two aircraft and are respectively:

[0092] ;

[0093] In the load suspension system, it is assumed that all aircraft have the same height, so ;

[0094] The newly obtained relative position deviation terms , , , , , are applied to the improved finite-time outer-loop consensus control protocol together with the updated tether tension.

[0095] The improved finite-time outer-loop consensus control protocol is as follows:

[0096] ;

[0097] Among them, The control quantity for the th aircraft obtained by the outer - loop consistency control protocol, is the position control gain between aircraft, is the position control gain between each aircraft and the virtual leader, is the velocity control gain between aircraft, is the velocity control gain between each aircraft and the virtual leader; is the number of aircraft; is the adjacency matrix The element in it represents the communication topology between aircraft. If aircraft can receive the state information of aircraft , then , otherwise, ; , ; is the acceleration of the virtual leader, is the velocity of the virtual leader, is the position of the virtual leader, and are the positions of aircraft and respectively, and are the velocities of aircraft and respectively, is the relative position deviation between aircraft and in the formation shape, is the relative position deviation between aircraft and the leader in the formation shape, is the tension of the load on the th aircraft through each tether, , is the gain parameter, is the tension on the tether of the th aircraft, , is the judgment variable of whether each aircraft can sense the position and velocity information of the virtual leader, that is, whether it can communicate, ;

[0098] is calculated through the above formula , , , , are the components of the control quantity in the , , directions respectively.

[0099] Step 3: Using the aircraft model, the control quantities of the position channel and the desired attitude angles are inversely solved through the control components; the desired attitude angles are input into the improved sliding-mode attitude controller for the inner loop to generate the control quantities of the attitude channel; the control quantities of the position channel and the attitude channel are applied to the aircraft to adjust the position and attitude of the aircraft in real time.

[0100] Based on the control components obtained from the quadrotor aircraft model , , inversely solve the desired pitch angle and the desired roll angle:

[0101] ;

[0102] wherein, is the control quantity of the position channel of the quadrotor aircraft and can be directly applied to the individual aircraft; is the th desired pitch angle of the quadrotor aircraft, is the th desired roll angle of the quadrotor aircraft, is the th desired yaw angle of the quadrotor aircraft, and the obtained desired angles are used for input into the sliding-mode attitude controller.

[0103] The sliding-mode attitude controller adopts an improved sliding-mode reaching law, and its form is as follows:

[0104] ;

[0105] wherein, is the sliding-mode variable, is the convergence speed term of the sliding-mode variable; is the sign function, , , , are the reaching-law adjustment parameters, and it is required that , , , ;

[0106] Design a sliding-mode controller for the attitude channel (roll , pitch , yaw ) of the quadrotor aircraft; set the attitude error as:

[0107] ;

[0108] wherein, represents the error of the attitude angle, represent the roll, pitch, and yaw attitude angles of the quadrotor aircraft represent the expected values of the roll, pitch, and yaw attitude angles of the quadrotor aircraft;

[0109] Define a sliding mode surface based on PD regulation. PD regulation (Proportional-Derivative regulation) is a control method that combines the effects of proportional (P) and derivative (D):

[0110] ;

[0111] where, represents the sliding mode surface of each attitude channel, , are the proportional regulation parameter and the derivative regulation parameter respectively, represents the first derivative of the attitude angle error. According to the quadrotor aircraft model, the attitude controller is designed as:

[0112] ;

[0113] In the formula, , is the first derivative of the attitude angle error, , respectively represent the first derivatives of the roll, pitch, and yaw attitude angles of the th quadrotor aircraft, respectively represent the second derivatives of the expected angles of the roll, pitch, and yaw attitude angles of the th quadrotor aircraft; , , are the sliding mode surfaces of the three attitude channels; , are the proportional regulation parameter and the derivative regulation parameter respectively, , is the sign function, , , , are the reaching law regulation parameters, and it is required that 、 、 、 ; represents the first derivative of the attitude angle error; is the disturbance coefficient of the quadrotor aircraft model, , and are the moments of inertia of the quadrotor aircraft about the axis, axis, axis;

[0114] The obtained attitude channel control quantity , , is applied to each quadrotor aircraft to quickly switch the attitude angle to achieve position tracking.

[0115] Specifically, the quadrotor aircraft model is as follows:

[0116] ;

[0117] In the formula, , , are respectively the components of the position channel of the quadrotor aircraft, , , are respectively the components of the attitude channel of the quadrotor aircraft, , , is the first derivative of the position channel component of the quadrotor aircraft, , , is the second derivative of the position channel component of the quadrotor aircraft, , , is the first derivative of the attitude channel component of the quadrotor aircraft, , , is the second derivative of the attitude channel component of the quadrotor aircraft, , and are the moments of inertia of the quadrotor aircraft about the axis, axis, axis, is the gravitational acceleration, is the mass of the quadrotor aircraft, is the disturbance coefficient; , , are the tether tensions of the load acting on the quadrotor aircraft in the directions of ; is the position channel control quantity, , , are respectively the of the quadrotor aircraft , , attitude channel control quantities; where represents the number of quadrotor aircraft;

[0118] The position channel control quantity , the control quantity of the attitude channel , , and the tether tension are input into the quadrotor aircraft model, so as to act on each quadrotor aircraft and adjust the position and attitude of the aircraft.

[0119] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A fault recovery control method for multi-aircraft suspended load formation, characterized in that The steps include: Step 1: During a mission with multiple drones flying the same payload, if a drone fails, the drone is immediately removed from the control system, its physical connection to the payload is released, and the tether tension between each drone and the payload is updated. Step 2: Reorganize the remaining valid aircraft into a regular polygon, establish new azimuths, assign new relative position deviation instructions, and input these instructions, along with the updated tether tensions, into the improved finite-time outer-loop consistency control protocol to generate control components that enable each aircraft to reach the specified position. Step 3: Using the aircraft model, the position channel control variable and the desired attitude angle are inversely solved by the control components. The desired attitude angle is input into the inner loop improved sliding mode attitude controller to generate the attitude channel control variable. The position channel control variable and the attitude channel control variable are applied to the aircraft to adjust the aircraft's position and attitude in real time. In step 2, the improved finite-time outer-loop consistency control protocol is as follows: ； Among them, is the control quantity of the th aircraft obtained by the outer loop consistency control protocol, is the position control gain between aircraft, is the position control gain between each aircraft and the virtual leader, is the speed control gain between aircraft, is the speed control gain between each aircraft and the virtual leader; is the number of aircraft; is the adjacency matrix The element in represents the communication topology between aircraft. If aircraft can receive the state information of aircraft , then , otherwise, ; , ; is the acceleration of the virtual leader, is the speed of the virtual leader, is the position of the virtual leader, and are the positions of aircraft and respectively, and are the speeds of aircraft and respectively, is the relative position deviation between aircraft and in the formation shape, is the relative position deviation between aircraft and the virtual leader in the formation shape, is the tension of the load on each tether on the th aircraft, , is the gain parameter, is the tension on the tether on the th aircraft, , is a judgment variable on whether each aircraft can sense the position and speed information of the virtual leader, that is, whether it can communicate, ; Obtained by the above formula , , , , are the , , components of the control quantity in the In step 3, the sliding mode attitude controller adopts an improved sliding mode reaching law, which is as follows: ； Among them, is the sliding mode variable, is the convergence speed term of the sliding mode variable; is the sign function, , , , are the reaching law adjustment parameters, and it is required that 、 、 、 ; Design a sliding mode controller for the attitude channel of the quadrotor aircraft; set the attitude error to: ； Among them, represents the error of the attitude angle, represents the roll, pitch, and yaw attitude angles of the quadrotor aircraft, represents the expected values of the roll, pitch, and yaw attitude angles of the quadrotor aircraft; Define the sliding surface based on PD regulation: ； Among them, represents the sliding mode surface of each attitude channel, , are the proportional adjustment parameter and the differential adjustment parameter respectively, represents the first derivative of the attitude angle error. According to the quadrotor model, the attitude controller is designed as: ； wherein , is the first derivative of the attitude angle error, , respectively represent the first derivatives of the roll, pitch, and yaw attitude angles of the th quadrotor aircraft, respectively represent the second derivatives of the desired angles of the roll, pitch, and yaw attitude angles of the th quadrotor aircraft; , , are the sliding mode surfaces of the three attitude channels; , are the proportional adjustment parameter and the differential adjustment parameter respectively, , is the sign function, , , , are the reaching law adjustment parameters, and it is required that , , , ; represents the first derivative of the attitude angle error; is the disturbance coefficient of the quadrotor aircraft model, , and are the moments of inertia of the quadrotor aircraft about the axis, axis, axis; The obtained attitude channel control quantity , , is applied to each quadrotor aircraft to quickly switch the attitude angle to achieve position tracking.

2. The multi-aircraft suspended load formation fault recovery control method according to claim 1, wherein In step one, an electronically controlled disconnect device is pre-installed at the hook connecting the aircraft and the load. When the force sensor detects a failure in a certain aircraft, the control system sends a release command to the electronically controlled disconnect device, automatically opening the hook and disconnecting the physical connection.

3. A multi-aircraft suspended load formation fault recovery control method according to claim 1, characterized in that, In step 1, the total tether tension is calculated using the Udwadia-Kalaba equation: ； In the formula, represents the direct product of the system mass matrix and the identity matrix, , represents the mass of the th aircraft, , and is the number of aircraft; represents the payload mass; ; represents the tether length error, defined as , is the first derivative of the length error, is the length of the tether on the th aircraft, represents the position vector difference between the th aircraft and the payload, and are the position vectors of the th aircraft and the payload respectively; represents the constraint acceleration term, and , is 's first derivative; represents 's Moore-Penrose pseudoinverse; and are feedback control gain coefficients, both being positive constants; represents the acceleration vector, that is, , and are the acceleration vectors of the th aircraft and the payload respectively; Calculate the total tether tension through the above formula, where is a matrix, expressed as follows: , where is the tension on the tether of the th aircraft, is the direction vector of the tether from the load to the th aircraft. The first columns in this matrix respectively represent the tensions acting on the th aircraft, and the th column represents the vector sum of all tensions; when the number of aircraft changes, the corresponding also changes, and each of its components also changes.

4. A fault recovery control method for multi-aircraft suspended load formation according to claim 1, characterized in that In step 2, the remaining valid aircraft are re-formed into a regular polygon, new azimuths are established, and new relative position deviation instructions are assigned as follows: To ensure the uniform distribution of the load force, the aircraft formation adopts a regular -sided structure, and regards the load as the virtual leader of the formation, being the number of aircraft; the coordinates of the -th aircraft are: ； Among them, is the position quantity of the virtual leader, represents the azimuth angle of the th aircraft, is the tether length on the th aircraft, is the distance between the th aircraft and the formation center: ； When the nd aircraft malfunctions, the malfunctioning aircraft with the original number is directly removed, and the remaining aircraft are re-numbered as according to their order in the communication topology; the number of sides of the new formation is , and at the same time, the azimuth angles of the remaining aircraft need to be re-allocated: ； Among them, represents the new azimuth angle of the Therefore, the relative coordinate components between the th aircraft and the virtual navigator are respectively , , ; wherein, and respectively represent the coordinates in the x and y directions of the th aircraft at the new azimuth angle; Two aircraft and The horizontal and vertical distances are respectively: ； In the load suspension system, it is assumed that all aircraft have the same altitude, so ; The obtained new relative position deviation terms ,<000023o>, , , , are applied, together with the updated tether tension, to the improved finite-time outer-loop consensus control protocol. It should be noted that in the original text, there might be a misspelling in "<000023o>", which is likely to be " ". The above translation is based on the corrected understanding.

5. A fault recovery control method for multi-aircraft suspended load formation according to claim 1, characterized in that, In step 3, when the aircraft is a quadrotor aircraft, the inverse solution formula is obtained according to the quadrotor aircraft model, and the desired pitch angle and desired roll angle are inversely solved through the obtained control components , , : ； Among them, is the control quantity of the position channel of the quadrotor aircraft and can directly act on the individual aircraft; is the th desired pitch angle of the quadrotor aircraft, is the th desired roll angle of the quadrotor aircraft, is the th desired yaw angle of the quadrotor aircraft, and the obtained desired angles are used as inputs to the sliding mode attitude controller.

6. A fault recovery control method for multi-aircraft hanging load formation according to claim 1, characterized in that, The quadrotor model is as follows: ； In the formula, , , are the components of the position channel of the quadrotor aircraft respectively, , , are the components of the attitude channel of the quadrotor aircraft respectively, , , is the first derivative of the component of the position channel of the quadrotor aircraft, , , is the second derivative of the component of the position channel of the quadrotor aircraft, , , is the first derivative of the component of the attitude channel of the quadrotor aircraft, , , is the second derivative of the component of the attitude channel of the quadrotor aircraft, , and are the moments of inertia of the quadrotor aircraft about the axis, axis, axis, is the gravitational acceleration, is the mass of the quadrotor aircraft, is the disturbance coefficient; , , are the tether tensions when the load acts on the quadrotor aircraft in the directions of respectively; is the control quantity of the position channel, , , are the control quantities of the attitude channels of the quadrotor aircraft in the , , directions respectively; where represents the number of the quadrotor aircraft; Input the position channel control quantity , the attitude channel control quantity , , and the tether tension into the quadrotor aircraft model to adjust the position and attitude of the aircraft.

Images