Reconfigurable unmanned aerial vehicle distributed attitude control method

Abstract

The application provides a reconfigurable unmanned aerial vehicle distributed attitude control method to realize stable attitude control of the reconfigurable unmanned aerial vehicle in different configurations. An aerial attitude task manager obtains the number of single machines of the combination, the rotor position information and the overall control force and moment required by the combination unmanned aerial vehicle through measurement and top layer calculation, and then transmits the information to each distributed controller. The distributed controller further calculates the required rotating speed of each rotor of the unmanned aerial vehicle unit. The distributed attitude control method can avoid the failure of using the traditional control method to control the attitude of the reconfigurable unmanned aerial vehicle, thereby effectively improving the stability and safety of the reconfigurable unmanned aerial vehicle. Each single machine is controlled by a separate controller, and the calculation pressure is dispersed to multiple processors. Compared with the traditional centralized control mode, the stability is higher, and the control of the whole reconfigurable unmanned aerial vehicle will not be affected by the problem of a single machine.

Description

TECHNICAL FIELD

[0001] The present application relates to a UAV attitude control method, in particular to a reconfigurable UAV distributed attitude control method, and belongs to the technical field of UAVs. BACKGROUND

[0002] UAVs can autonomously perform tasks such as logistics, transportation, distribution, and patrol in civilian fields, and can perform tasks such as coordinated combat, electronic countermeasures, and rescue in military fields, and are core elements in the construction of smart cities and future battlefield environments. With the expansion of the connotation of smart cities and battlefield environments, the development of UAVs faces major challenges such as complex and variable task execution, three-dimensional and multi-dimensional operating environment, continuous expansion of functional requirements, and single and limited carrier configuration. Obviously, traditional fixed-configuration UAVs have been unable to cope with the above challenges, and are subject to the contradiction between size flexibility and functional diversity, and cannot meet the needs of large load, high power, and long endurance.

[0003] A reconfigurable UAV is composed of multiple UAV units, and the combination or disassembly of multiple UAV units is achieved through an autonomous docking device, realizing multi-level reconfiguration of the mechanical, information, and energy layers of the UAV to improve load capacity, endurance, and multi-vehicle cooperation capability. Reconfigurable UAV technology expands the functional task execution boundaries of UAVs, provides more flexible usage of UAVs, and provides new ideas and directions for the development of UAVs, and is expected to become a future disruptive innovation technology. Due to the complex and variable nature of the air environment, the configuration of the reconfigurable UAV is quite different from that of the traditional rotor UAV, and the previous UAV control method cannot meet the attitude control requirements of the reconfigurable UAV. SUMMARY

[0004] Therefore, the present application provides a distributed attitude control method for a reconfigurable UAV to realize stable attitude control of the reconfigurable UAV in different configurations.

[0005] The distributed attitude control method for the reconfigurable UAV: the minimum unit of the reconfigurable UAV is a UAV unit;

[0006] A distributed control system is provided on each UAV unit for control of its own flight power system, which is a rotor driven by a motor;

[0007] The distributed control system includes a flight controller, a distributed sensor, a distributed controller, and a communication system; the distributed controller has pre-stored therein a plurality of structural configurations of combined UAVs and a proportionality coefficient of the required pulling force and moment of each UAV unit in the structural configuration to the overall pulling force and moment;

[0008] When two or more drone units dock to form a combined drone, the distributed controller on one of the drone units is designated as the aerial attitude task manager via remote command.

[0009] The distributed controller receives the single-machine pose data collected by its own distributed sensors and sends it to the air attitude task manager through the communication system.

[0010] The aerial attitude task manager determines the current structural configuration and overall attitude of the combined UAV based on the individual attitudes of each UAV unit received. At the same time, it calculates the overall pull and torque required by the combined UAV based on the flight mission, and then transmits them to each distributed controller through the communication system. The distributed controller then calculates the required rotational speed of the rotor in its own flight power system, and drives the motor through the flight controller to make the corresponding rotor rotate at the required speed.

[0011] In a preferred embodiment of the present invention, the unmanned aerial vehicle (UAV) unit is a quadcopter UAV.

[0012] In a preferred embodiment of the present invention, the combined UAV is composed of a quadcopter UAV units, where a > 1; and the combined UAV has a centrally symmetrical structural configuration.

[0013] The method by which the distributed controller calculates the required rotational speed of each rotor in its own flight propulsion system is as follows:

[0014] The required thrust for the combined drone is evenly distributed to each rotor, and the resulting rotational speed is used as the base rotational speed.

[0015]

[0016] Where: n ijF After the combined pull is evenly distributed to each rotor, the rotational speed of the j-th rotor on the i-th unmanned unit is also the reference rotational speed;

[0017] Let T be the thrust generated by the j-th rotor on the i-th unmanned unit. ij For: T ij =c T n 2 ij The generated torque M ij For: M ij =c M n 2 ij The control efficiency matrix of the combined UAV is:

[0018]

[0019] In equation (1): T is the total tensile force required for the combined UAV, τx , τ y , τ z are the moments required by the combined unmanned aerial vehicle as a whole in three directions of the system body coordinate system, respectively, roll moment, pitch moment and yaw moment; n ij is the rotating speed of the jth rotor on the ith unmanned aerial vehicle unit, j = 1, 2, 3, 4, and the rotors on a single unmanned aerial vehicle unit are numbered in clockwise or counterclockwise order, wherein the rotors numbered 1 and 3 are on a diagonal line, and the rotors thereof adopt positive pitch; the rotors numbered 2 and 4 are on a diagonal line, and the rotors thereof adopt reverse pitch; c T is the lift coefficient of a single rotor, c M is the torque coefficient of a single rotor;

[0020] For the distribution of roll moment and pitch moment, the rotating speed increment of each rotor is the same, then:

[0021]

[0022] In the formula: is the increment of the rotating speed square of each rotor required by the roll moment, is the increment of the rotating speed square of each rotor required by the pitch moment; χ(y ij ) = -y ij / |y ij |, χ(x ij ) = x ij / |x ij |; (x ij , y ij , 0) is the coordinate of the jth rotor on the ith unmanned aerial vehicle unit in the system body coordinate system;

[0023] The yaw moment is evenly distributed to each rotor, then the increment of the rotating speed square of each rotor required by the yaw moment is expressed as:

[0024]

[0025] Then the required rotating speed n ij of each rotor is:

[0026]

[0027] As a preferred mode of the present application, when calculating the required rotating speed of each rotor, the difference between the attitude of each single unmanned aerial vehicle unit and the attitude of the combined unmanned aerial vehicle as a whole is further considered, then:

[0028] n ij ' = n ij + Δn ij

[0029] wherein: n ij is the corrected required rotational speed of each rotor; Δn ij is the incremental rotational speed of the rotor required to correct the deviation of the single machine attitude and the whole machine attitude.

[0030] As a preferred mode of the present application, the unmanned aerial vehicle unit playing the role of the aerial attitude task manager is the main unmanned aerial vehicle unit, and if the aerial attitude task manager on the main unmanned aerial vehicle unit fails, the distributed controller on the remaining unmanned aerial vehicle units in the combined unmanned aerial vehicle is replaced as the aerial attitude task manager through remote instruction.

[0031] As a preferred mode of the present application, a data line interface is arranged on the docking surface of the unmanned aerial vehicle unit, and after two or more unmanned aerial vehicle units are docked to form a combined unmanned aerial vehicle, the data transmission lines are interconnected through the data line interfaces while the unmanned aerial vehicle units are mechanically connected.

[0032] Advantages:

[0033] (1) The distributed attitude control method of the reconfigurable unmanned aerial vehicle can realize stable attitude control of the reconfigurable unmanned aerial vehicle in different configurations, avoid the failure of using traditional control methods to control the attitude of the reconfigurable unmanned aerial vehicle, and effectively improve the stability and safety of the reconfigurable unmanned aerial vehicle. Each single body of the reconfigurable unmanned aerial vehicle is controlled by a separate controller, which disperses the computing pressure to multiple processors. Compared with the traditional centralized control method, i.e. the control method of the unmanned aerial vehicle system with only one processor, the distributed control system has higher stability and will not affect the overall control of the reconfigurable unmanned aerial vehicle due to problems of a single body.

[0034] In addition, the distributed control makes the structure and circuit of the unmanned aerial vehicle system simpler and clearer. Compared with the traditional centralized control, the distributed control is easier to implement and has the characteristics of simplicity.

[0035] The distributed control makes the system more modular, and each reconfigurable unmanned aerial vehicle unit has an independent management mode. With the increase in the number of combined units, the communication overhead increases less, i.e. the change required for the original system is not large, and the system has good scalability.

[0036] (2) In the distributed attitude control system of the present application, the hardware configuration of each unmanned aerial vehicle unit in the combined unmanned aerial vehicle is the same, and each unmanned aerial vehicle unit has the function of the aerial attitude task manager. When the existing aerial attitude manager is damaged, any backup aerial attitude task manager in the combination can be specified to continue to perform the task.

[0037] (3) In the distributed attitude control method of the reconfigurable UAV of the present invention, the distribution of the first type of torque adopts the method that the magnitude of the rotational speed increment of each rotor is the same; this method can make the rotational speed increment of each rotor of the combined UAV consistent, and avoid the situation that the motors corresponding to each or each type of rotor are saturated in power.

[0038] (4) In the distributed attitude control method of the reconfigurable UAV of the present invention, the disturbance offset of the single - machine attitude is generated according to the single - machine attitude data and the overall attitude data, and the attitude control method of the conventional quad - rotor UAV unit is used to form the required rotor rotational speed increment for correction; thus, the attitude offset formed by the inconsistency between the single - machine attitude and the overall attitude caused by various disturbances can be eliminated, the interference forces and interference torques between each reconfigurable UAV unit are reduced, and the aircraft hardware structure is protected. Description of the Drawings

[0039] Figure 1 It is a structural schematic diagram of a single UAV unit;

[0040] Figure 2 It is a schematic diagram of the "square" combination of four UAV units;

[0041] Figure 3 It is a schematic diagram of the "cross" combination of five UAV units;

[0042] Figure 4 It is a schematic diagram of the structure of the distributed control system;

[0043] Figure 5 It is a schematic diagram of the data flow under the "cross" layout of five UAV units;

[0044] Figure 6 It is a schematic diagram of two body coordinate systems under the "field" layout of four UAV units. Detailed Embodiment

[0045] The following combines the drawings and gives examples to describe the present invention in detail.

[0046] In order to solve the problem of the attitude stability of the reconfigurable UAV, this embodiment proposes a distributed attitude control method for the reconfigurable UAV, which is used to achieve stable attitude control of the reconfigurable UAV in different configurations, avoid the situation of failure when using the traditional control method for attitude control of the reconfigurable UAV, and thus effectively improve the stability and safety of the reconfigurable UAV.

[0047] In this example, the minimum unit of the reconfigurable UAV is as Figure 1The unmanned aerial vehicle units shown can be connected through the docking mechanism to achieve reconfiguration; in this example, the unmanned aerial vehicle unit is a quadcopter, having four rotors arranged in a rectangular pattern, each rotor being controlled by a motor; the unmanned aerial vehicle unit has a rectangular frame, and a docking mechanism is arranged at the middle of the periphery of the rectangular frame, so that multiple unmanned aerial vehicle units can be reconfigured into different forms through flexible combination to adapt to task requirements; for example, four unmanned aerial vehicle units are reconfigured in the "mouth" shape as shown in Figure 2 , or five unmanned aerial vehicle units are reconfigured in the "cross" shape as shown in Figure 3 .

[0048] In this example, the "cross" combination unmanned aerial vehicle as shown in Figure 3 is taken as an example to introduce the distributed attitude control method of the reconfigurable unmanned aerial vehicle. For convenience of description, the reconfigured unmanned aerial vehicle of multiple unmanned aerial vehicle units is referred to as a combined unmanned aerial vehicle.

[0049] The distributed attitude control method is based on the distributed control system as shown in Figure 4 , and each unmanned aerial vehicle unit is equipped with a complete distributed control system, and the hardware components are completely the same. The distributed control system includes a flight controller, a distributed sensor, a distributed controller, and a communication system; the distributed control system is used to control the four motors of itself; when multiple unmanned aerial vehicle units are reconfigured and combined to fly, a distributed controller on one unmanned aerial vehicle unit of the combined unmanned aerial vehicle is designated as an air attitude task manager through a remote command, and the air attitude task manager can communicate with all unmanned aerial vehicle units in the combined unmanned aerial vehicle through the communication system to perform information fusion and issue task information. The unmanned aerial vehicle unit that plays the role of the air attitude task manager is referred to as a master unmanned aerial vehicle unit, and if the air attitude task manager on the master unmanned aerial vehicle unit fails, a distributed controller on another unmanned aerial vehicle unit can be replaced as an air attitude task manager (i.e., the master unmanned aerial vehicle unit is replaced) through a remote command, so as to avoid causing flight failure and improve the reliability of the combined unmanned aerial vehicle.

[0050] As shown in Figure 4 and Figure 5 , the role of the air attitude task manager on the unmanned aerial vehicle unit located at the center position in the "cross" combination unmanned aerial vehicle is taken as an example to illustrate the information flow in the control process:

[0051] The distributed controller (including the distributed controller as the air attitude task manager) is an onboard computer, and all distributed controllers have pre-stored therein the structural configurations of multiple common combined unmanned aerial vehicles and the proportion coefficients of the required pulling force and moment of each unmanned aerial vehicle unit in the structural configuration to the overall pulling force and moment (these parameters are obtained through theoretical calculation, for example, for the "cross" combination unmanned aerial vehicle as shown in Figure 6The center-symmetrical structure configuration shown, the required tension and moment of each unmanned aerial vehicle unit occupies the same proportion of the overall tension and moment.

[0052] The distributed controller receives the single-machine pose collected by its own distributed sensor, and sends the single-machine pose to the aerial attitude task manager through the communication system, thereby realizing information fusion; the aerial attitude task manager can determine the combination form of the current combined unmanned aerial vehicle (i.e. the structure configuration and the number of single machines) according to the received single-machine pose of each unmanned aerial vehicle unit; at the same time, the aerial attitude task manager generates reliable current attitude, speed, position and other information of the combined unmanned aerial vehicle as the overall pose data of the combined unmanned aerial vehicle through calculation and fusion filtering. In addition, the aerial information manager forms a control instruction (i.e. the aerial attitude task manager calculates the required tension and moment of the overall combined unmanned aerial vehicle according to the flight task) according to the current state and flight task of the combined unmanned aerial vehicle, and transmits the control instruction and overall pose data to each distributed controller through the communication system. After receiving the control instruction and overall pose data issued by the aerial attitude task manager, the distributed controller performs control allocation calculation on the four motors of itself, obtains the required rotation speed of each rotor in the unmanned aerial vehicle unit, and forms the corresponding motor response to be sent to the flight controller, so that the corresponding rotors rotate at the required rotation speed through the flight controller driving the four motors.

[0053] The combined unmanned aerial vehicles need to transmit instructions to each other, so the communication system scheme between multiple machines needs to be considered. The communication system scheme needs to be determined according to specific tasks and requirements, and can adopt wired and wireless modes.

[0054] The implementation mode of wired communication is to integrate the data line interface and the interface of the unmanned aerial vehicle unit docking mechanism, realize the mechanical connection between the unmanned aerial vehicle units, and interconnect the data transmission lines. The advantage is that it can achieve zero delay and zero packet loss in theory, and has a larger signal transmission speed, but the docking accuracy and stability requirements are increased.

[0055] The implementation mode of wireless communication is that, since the communication distance between each single unit (i.e. unmanned aerial vehicle unit) of the combined unmanned aerial vehicle is usually only a few meters, the data is mainly pose data and instruction data, and zigbee wireless communication protocol is adopted, so that the required bandwidth is lower than the maximum transmission rate of the wireless communication protocol, and the communication delay and packet loss rate also meet the communication requirements. The advantage of wireless communication is that it can effectively reduce the complexity of the system structure and circuit; but it may cause a certain delay, and the anti-interference ability is not ideal.

[0056] Based on this, the communication system in this example adopts the wired communication mode to realize the communication between each unmanned aerial vehicle unit in the combined unmanned aerial vehicle.

[0057] Based on the above distributed control system, the overall idea of the distributed attitude control of the reconfigurable unmanned aerial vehicle is as follows:

[0058] With the distributed control scheme, the aerial attitude task manager senses the structural configuration of the combined unmanned aerial vehicle, calculates the overall attitude data of the combined unmanned aerial vehicle, compares the attitude data required by the task, and generates the control force and torque information required by the whole machine by using the deviation. The aerial attitude task manager does not directly issue control instructions to each rotor, but transmits the structural configuration of the combined unmanned aerial vehicle and the tension and torque information required by the whole combined unmanned aerial vehicle to each distributed controller, and then the distributed controller on each unmanned aerial vehicle unit controls the rotor speed by using the control distribution equation. The hardware configuration of each unmanned aerial vehicle unit in the combined unmanned aerial vehicle is the same, and each unmanned aerial vehicle unit has the function of the aerial attitude task manager. When the existing aerial attitude manager is damaged, any backup aerial attitude task manager in the combination can continue to perform the task.

[0059] The specific control method is as follows:

[0060] First, the ground coordinate system and the machine body coordinate system are established, and the machine body coordinate system includes the system machine body coordinate system S and the single machine body coordinate system M (which is established for each unmanned aerial vehicle unit in the combined unmanned aerial vehicle). Among them, the ground coordinate system and the single machine body coordinate system M are established inside the distributed controller of each unmanned aerial vehicle unit; the system machine body coordinate system S is established after the aerial attitude task manager determines the combined form of the combined unmanned aerial vehicle.

[0061] The ground coordinate system takes the takeoff point position of the combined unmanned aerial vehicle as the coordinate origin O, the X axis is tangent to the great circle of the earth passing through the coordinate origin and the target point (i.e. the target position of the combined unmanned aerial vehicle), the direction of the combined unmanned aerial vehicle flying towards the target point is the positive direction, the Z axis is perpendicular to the horizontal plane, and the downward direction is positive, and the Y axis is perpendicular to the OXZ plane, and its direction is determined according to the right-hand rule; the ground coordinate system is established by the distributed controller of each unmanned aerial vehicle unit after receiving the remote task instruction.

[0062] The single machine body coordinate system is pre-established in the distributed controller of each unmanned aerial vehicle unit, the coordinate origin O1 of the single machine body coordinate system is the center of mass of the unmanned aerial vehicle unit, the X1 axis coincides with the geometric longitudinal axis of the unmanned aerial vehicle unit, and the positive direction is along the flight direction; the Z1 axis is in the longitudinal symmetry plane of the unmanned aerial vehicle unit, perpendicular to the X1 axis, and the positive direction is downward; the Y1 axis is perpendicular to the O1X1Z1 plane, and its direction is determined according to the right-hand rule.

[0063] The system machine body coordinate system S is established in the aerial attitude task manager, the coordinate origin O2 of the system machine body coordinate system S is the center of mass of the combined unmanned aerial vehicle, the X2 axis and the Z2 axis are parallel to the X1 axis and the Z1 axis respectively, and the Y2 axis is perpendicular to the O2X2Z2 plane, and its direction is determined according to the right-hand rule.

[0064] The distributed attitude control method takes the above ground coordinate system and body coordinate system as the reference to establish the dynamics model and control efficiency model of the combined unmanned aerial vehicle.

[0065] First, the air attitude task manager establishes the dynamics model:

[0066] The combined unmanned aerial vehicle is regarded as a rigid body, and the inertia parameters (such as mass, inertia product, and inertia moment) in the traditional rotor unmanned aerial vehicle rigid body dynamics equation set are replaced accordingly, so that the dynamics modeling of the combined unmanned aerial vehicle is established. The dynamics equation set is solved to obtain all the control forces and moments required for the combined unmanned aerial vehicle to perform path planning or attitude control, so as to be decomposed into individual unmanned aerial vehicles subsequently.

[0067] The air attitude task manager sends the determined structure of the current combined unmanned aerial vehicle and the calculated overall control force (tension) and moment of the combined unmanned aerial vehicle required for path planning or attitude control to each distributed controller, and each distributed controller calculates the rotation speed of each rotor. In this example, since the tension and moment required by each unmanned aerial vehicle unit account for the same proportion of the overall tension and moment, the overall tension and moment are directly evenly distributed to each rotor during calculation. Specifically:

[0068] Let the combined unmanned aerial vehicle consist of a (a > 1) quad-rotor unmanned aerial vehicle units, and let the body coordinate system of the i-th unmanned aerial vehicle unit be M i , and the coordinate of the coordinate origin of the unmanned aerial vehicle unit body coordinate system O 1i is (0, 0, 0). The coordinates of the coordinate origin of the system body coordinate system S are (x i , y i , 0). In this example, the four rotors of the quad-rotor unmanned aerial vehicle unit are distributed in a square, and the side length of the square is 2d, as shown in Figure 1 , the coordinates of the corresponding motors in the single machine body coordinate system M i are (d, -d, 0), (d, d, 0), (-d, d, 0), and (-d, -d, 0), which are simply denoted as (x Mi , y Mi , 0).

[0069] The coordinates (x ij , y ij , 0) of the j-th rotor on the i-th unmanned aerial vehicle unit in the system body coordinate system S are:

[0070] (x ij , y ij , 0) = (x Mi , y Mi , 0) + (x i , yi ,0)

[0071] Let the tension T generated by the jth rotor on the ith unmanned unit be: ij T = c ij n T 2 ij Let the torque M generated by the jth rotor on the ith unmanned unit be: ij M = c ij n M 2 ij Then the control efficiency matrix of the combined unmanned aerial vehicle is:

[0072]

[0073] In equation (1): [T τ x τ y τ z ] T is the overall control force (which is the tension) and torque required by the combined unmanned aerial vehicle, specifically: T is the overall tension required by the combined unmanned aerial vehicle (also the combined tension provided by the combined unmanned aerial vehicle), τ x , τ y , τ z are the torques required by the overall combined unmanned aerial vehicle in the three directions of the system body coordinate system, respectively, which are the roll torque, pitch torque, and yaw torque.n ij n is the rotational speed of the jth rotor on the ith unmanned aerial vehicle unit, j = 1, 2, 3, 4, and the rotors on a single unmanned aerial vehicle unit are numbered in clockwise or counterclockwise order, i.e., the rotors corresponding to 1 and 3 are on one diagonal, and the rotors corresponding to 2 and 4 are on the other diagonal, c T is the tension coefficient of a single rotor, c M is the torque coefficient of a single rotor, which can be determined by experiment; since the four rotors on the unmanned aerial vehicle unit are the same, the tension coefficients and torque coefficients of the four rotors are also the same.

[0074] Equation (1) represents the relationship between the overall control force, torque required by the combined unmanned aerial vehicle, and the rotational speed and position of each rotor in the ith unmanned unit. Since the number of controllable motors is 4a, and the controlled quantities are the combined tension and three-axis torque, there are currently only four equations, and the number of variables is greater than the number of equations, which cannot be closed for solution. Therefore, certain constraints need to be added to the distribution of motor speeds to obtain a unique solution for each rotor speed.

[0075] The constraints on the distribution of motor speeds are added below to solve the unique solution of the motor speed represented by the overall control force of the combined unmanned aerial vehicle, the relative position of the rotors, and the number of single machines, i.e., the control distribution equation.

[0076] ​​The resultant tension T of the four motors of the combined unmanned aerial vehicle is perpendicular to the reference plane of the combined unmanned aerial vehicle, that is, in the opposite direction of the Z2 axis, for balancing the gravity and aerodynamic drag and maintaining the flight speed. In the reconstructed combined unmanned aerial vehicle, in order to avoid uneven distribution of tension and generate additional bending moment acting on the fuselage, the resultant tension should be evenly distributed to each rotor, and the rotational speed formed thereby is taken as the reference rotational speed, that is:

[0077]

[0078] In formula (2), n ijF is the rotational speed of the jth rotor on the ith unmanned unit after the resultant tension is evenly distributed to each rotor, and is also the reference rotational speed.

[0079] For the distribution of the moment, it can be divided into the rolling moment and the pitching moment (let them be the first type of moment) which cause the rotation around the X2 axis and the Y2 axis of the system body coordinate system, and the yawing moment (let it be the second type of moment) which causes the rotation around the Z2 axis; the first type of moment is generated when the moments generated by the tension on the mass center of the rotors on both sides of the mass center are unbalanced, and the second type of moment is generated when the counter-torques generated by the rotation of all rotors cannot be offset by each other; the distribution of the moment is to control each unmanned aerial vehicle unit to generate the first type of moment and the second type of moment which meet the requirements of the flight task.

[0080] In the present scheme, only the symmetric reconstruction form shown in Figure 2 and Figure 3 is considered, and other central symmetric types (usually, the reconstructed combined unmanned aerial vehicle is of a central symmetric type); for such a distributed combined unmanned aerial vehicle, the distribution of the moment is carried out in the following manner:

[0081] For the generation of the rolling moment and the pitching moment, the saturation characteristics of the motor and the influence of the moment on the structure need to be considered; such a moment is the additional tension generated by the rotational speed increment of the motor at the reference rotational speed, and the moment generated by the additional tension on the mass center of the combined unmanned aerial vehicle (that is, at the origin O2 of the system body coordinate system); in the present embodiment, the distribution of the first type of moment adopts the manner that the rotational speed increment of each rotor is the same:

[0082] The rotational speed increment of each rotor is the same, so the tension increment ΔT generated by each rotor needs to be the same, and the moment increment generated thereby can be expressed as Δτ ij = ΔTx ij or ΔTy ij , so it can be seen that the farther the rotor is from the mass center of the combined unmanned aerial vehicle, the greater the moment increment generated thereby. This manner can keep the rotational speed increment of each rotor of the combined unmanned aerial vehicle consistent, and avoid the power saturation of the motor corresponding to each or each type of rotor.

[0083] The increment of each rotor speed square required by the first type of moment can be expressed as:

[0084]

[0085] wherein: is the increment of each rotor speed square required by the roll moment, is the increment of each rotor speed square required by the pitch moment; χ(x) is a sign function, i.e. χ(x) = x / |x|, i.e. χ(y ij ) = -y ij / |y ij |, χ(x ij ) = x ij / |x ij |.

[0086] The increment of each rotor speed square required by the yaw moment can be expressed as:

[0087]

[0088] In summary, the square of the speed required by each rotor is the sum of the square of the reference speed and the increment of each rotor speed square corresponding to the three moments (pitch moment, roll moment and yaw moment), and the expression is:

[0089]

[0090] The above formula (5) is the standard form of the control distribution equation.

[0091] In addition, due to the factors such as inaccurate docking position, disturbance of the UAV in the air and elastic deformation of the structure during the docking process, the attitude of a single four-rotor UAV unit may not be completely consistent with the overall attitude of the combined UAV after reconstruction. This situation is easy to cause the control system to be disorderly, and the fuselage structure, docking mechanism and other structures to be subjected to additional bending moment, shear force and other situations. Therefore, the attitude deviation should be corrected as much as possible. For this purpose, the disturbance deviation of the single machine attitude can be generated according to the single machine attitude data and the overall attitude data, i.e. the disturbance deviation represents the difference between the four-rotor UAV unit and the overall combined UAV attitude. According to the disturbance deviation, the rotor speed increment Δn ij required for correction can be directly formed using the attitude control method of the conventional four-rotor UAV unit, and the final speed required by the motor command issued by the distributed controller is:

[0092] n ij ′ = n ij + Δn ij (6)

[0093] wherein n​ij n' is the required rotational speed of each rotor after correction, Δn ij Δn is the required rotational speed increment of each rotor for correcting the deviation between the single machine attitude and the overall attitude.

[0094] From the above, it can be seen that under the condition that the layout of each unmanned aerial vehicle unit of the combined unmanned aerial vehicle is certain, the algorithm can realize one-to-one correspondence between the overall control force and control moment of the combined unmanned aerial vehicle and the rotational speed of each rotor. That is, when the aerial attitude task manager obtains the number of single machines of the combination, the rotor position information and the overall control force and moment required by the combined unmanned aerial vehicle through measurement and top-level calculation, the information is transmitted to each distributed controller, and the distributed controller can obtain the rotational speed of each rotor of the unmanned aerial vehicle unit through formula (5) and (6).

[0095] To sum up, the above is only a preferred embodiment of the present application, and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A reconfigurable unmanned aerial vehicle distributed attitude control method, characterized in that, Each unmanned aerial vehicle unit is provided with a distributed control system for controlling its own flight power system, which is a rotor driven by a motor; The distributed control system comprises a flight controller, a distributed sensor, a distributed controller and a communication system; the distributed controller has pre-stored therein a plurality of structural configurations of the combined unmanned aerial vehicle and a proportionality coefficient of the required tension and moment of each unmanned aerial vehicle unit in the structural configuration to the overall tension and moment; When two or more unmanned aerial vehicle units are docked to form a combined unmanned aerial vehicle, a remote instruction is used to specify that the distributed controller on one of the unmanned aerial vehicle units is an aerial attitude task manager; The distributed controller receives the single-machine attitude of the distributed sensor and sends it to the aerial attitude task manager through the communication system; The aerial attitude task manager determines the structural configuration of the combined unmanned aerial vehicle and the overall attitude according to the received single-machine attitude of each unmanned aerial vehicle unit, and calculates the required tension and moment of the combined unmanned aerial vehicle according to the flight task and transmits it to each distributed controller through the communication system; the distributed controller calculates the required rotation speed of the rotor in its flight power system according to the pre-stored proportionality coefficient and the received overall tension and moment, and drives the motor through the flight controller to make the corresponding rotor rotate at the required rotation speed; Thus, under the condition that the layout of each unmanned aerial vehicle unit of the combined unmanned aerial vehicle is certain, the one-to-one correspondence between the overall control force, control moment and rotation speed of each rotor of the combined unmanned aerial vehicle is realized through the control method; The combined unmanned aerial vehicle is composed of four-rotor unmanned aerial vehicle units, wherein 1; and the combined unmanned aerial vehicle is a central symmetric structural configuration; The method for the distributed controller to calculate the required rotation speed of each rotor in its flight power system is as follows: The required tension of the combined unmanned aerial vehicle is evenly distributed to each rotor, and the rotation speed formed thereby is taken as the reference rotation speed, i.e. wherein: After the tension is evenly distributed to each rotor, the rotating speed of the jthrotor on the ith unmanned unit is also the reference rotating speed. Let the pulling force T generated by the jth rotor on the ith unmanned unit be ij T = c n ij = c n T n 2 ij Let the torque M generated by the jth rotor on the ith unmanned unit be ij M = c n ij = c n M n 2 ij The control efficiency matrix of the combined unmanned aerial vehicle is: Wherein: is the required tension of the combined unmanned aerial vehicle as a whole, , , are the required moments of the combined unmanned aerial vehicle as a whole in three directions of the system body coordinate system, respectively roll moment, pitch moment and yaw moment; n ij is the rotational speed of the jth rotor on the ith unmanned aerial vehicle unit, j = 1, 2, 3, 4, and the rotors on a single unmanned aerial vehicle unit are numbered in clockwise or counterclockwise order, wherein the rotors numbered 1 and 3 are on a diagonal line, and the rotors adopt positive pitch; the rotors numbered 2 and 4 are on a diagonal line, and the rotors adopt negative pitch, is the tension coefficient of a single rotor, is the torque coefficient of a single rotor; For the distribution of roll moment and pitch moment, the rotation speed increment of each rotor is the same, i.e. wherein: is the increment of each rotor rotation speed square required for the roll moment, is the increment of each rotor rotation speed square required for the pitch moment; , ; (x ij ,y ij ,0) is the coordinate of the jthrotor on the ithdrone unit in the system body coordinate system; the yawing moment is evenly distributed to each rotor, then the incremental increase in the square of the rotational speed of each rotor required for the yawing moment is represented as: the required rotational speed of each rotor is: 。 2. The reconfigurable UAV distributed attitude control method of claim 1, wherein, When calculating the required rotation speed of each rotor, the difference between the attitude of each single unmanned aerial vehicle unit and the attitude of the combined unmanned aerial vehicle is further considered, i.e. wherein: is the required rotational speed of each rotor after correction; is the increment of the rotational speed of the rotor required to correct the deviation of the single machine attitude from the overall machine attitude.

3. The reconfigurable UAV distributed attitude control method of claim 1, wherein, The unmanned aerial vehicle unit that plays the role of the aerial attitude task manager is the main unmanned aerial vehicle unit, and if the aerial attitude task manager on the main unmanned aerial vehicle unit fails, the distributed controllers on the remaining unmanned aerial vehicle units of the combined unmanned aerial vehicle are replaced as the aerial attitude task manager through a remote instruction.

4. The reconfigurable UAV distributed attitude control method of claim 1, wherein, The docking surface of the unmanned aerial vehicle unit is provided with a data line interface, and after two or more unmanned aerial vehicle units are docked to form a combined unmanned aerial vehicle, the data transmission lines are interconnected through the data line interface while the unmanned aerial vehicle units are mechanically connected.

Images