A distributed tilt-rotor UAV control distribution method and device

Abstract

The present application relates to a distributed tilt-rotor UAV control allocation method and device. The method includes: finding the optimal projection point on the boundary of the closed three-dimensional reachable torque set as a reachable virtual control torque based on a priority projection algorithm, designing a cascade decoupling control allocation algorithm to calculate the control quantities of the tilt angle and rotor speed; simplifying the control allocation matrix using the reference tilt angle characteristics under the rotor mode, analyzing the control allocation matrix to obtain a first tension constant and a second tension constant; collecting a virtual control torque in a hovering flight state, correcting the first tension constant and the second tension constant according to the virtual control torque, and solving the control quantity of the rotor speed using the accurate first tension constant and the second tension constant to obtain the rotor speed. This method can improve the accuracy of UAV control allocation.

Description

Technical Field

[0001] The present application relates to the field of UAV flight control technology, and in particular to a distributed tilt-rotor UAV control distribution method and device. Background Art

[0002] Tilt-rotor UAVs (UAVs) are a new type of aircraft that combine the features of rotary-wing and fixed-wing UAVs, capable of vertical takeoff and landing, fixed-point hovering, and high-speed flight. While this unique configuration offers unique advantages, it also presents significant challenges for flight control. Control allocation is crucial for achieving stable flight for tilt-rotor UAVs. This process maps the virtual control torque output by the controller to each actuator, enabling the actuator to deliver the desired torque to achieve attitude and altitude control. The input to control allocation is the virtual control torque, and the output is the actuator control command. Research on tilt-rotor UAVs typically addresses the control allocation problem through simple direct allocation or pseudo-inverse allocation methods. However, in-depth research on the robustness, accuracy, and achievable torque set characteristics of control allocation algorithms is lacking. In practical applications, for power-constrained tilt-rotor UAVs, efficient and reliable control allocation algorithms can not only effectively improve actuator efficiency but also enhance the vehicle's controllability and flight control performance. Summary of the Invention

[0003] Based on this, it is necessary to provide a distributed tilt-rotor UAV control distribution method, device, computer equipment and storage medium that can improve the accuracy of UAV control distribution to address the above technical problems.

[0004] A distributed tilt-rotor UAV control allocation method, the method comprising:

[0005] Construct a control allocation model based on the rotor torque generation mechanism;

[0006] The control allocation matrix is ​​expressed as a linear control allocation model, which includes the control allocation matrix and the actuator control quantity;

[0007] A three-dimensional achievable moment set is obtained by constructing a achievable control set based on multiple actuator control quantities and performing affine transformation on the achievable control set using the control allocation matrix.

[0008] The control allocation matrix is ​​used to determine the vertices and multiple reference points of the three-dimensional reachable moment set, and a closed three-dimensional reachable moment set is constructed based on the vertices and reference points;

[0009] A priority-based projection algorithm finds the optimal projection point on the boundary of a closed three-dimensional achievable torque set as the achievable virtual control torque. Based on the achievable virtual control torque, a cascade decoupling control allocation algorithm is designed to calculate the control variables of the tilt angle and rotor speed.

[0010] The control allocation matrix is ​​simplified by using the reference tilt angle characteristics under the rotor mode, and the control allocation matrix is ​​analyzed to obtain the perturbation parameters; the perturbation parameters include the first tension constant and the second tension constant;

[0011] Collecting a virtual control torque in a hovering flight state, and correcting the first tension constant and the second tension constant according to the virtual control torque to obtain accurate first tension constant and second tension constant;

[0012] The control variable of the rotor speed is solved using the accurate first tension constant and the second tension constant to obtain the rotor speed.

[0013] In one embodiment, a control allocation model is constructed based on a rotor torque generation mechanism, including:

[0014] According to the rotor torque generation mechanism, the control allocation model is constructed as follows:

[0015]

[0016] Among them, R r represents the rolling moment, P r represents the pitching moment, Y r represents the yaw moment, T r represents the tension, η i and ω i Represent the tilt angle and speed of the i-th rotor, k f 、k d represents the thrust constant and torque constant of the forward rotor, represents the position of the rotor in the body coordinate system, k′ f , k′ d represents the tail rotor force constant and torque constant, c*=cos(*), s*=sin(*), η i represents the tilt angle, ω i Indicates the rotor speed.

[0017] In one embodiment, determining the vertices and multiple reference points of a three-dimensional reachable moment set using a control allocation matrix includes:

[0018] The vertices of the three-dimensional reachable moment set are determined using the control allocation matrix:

[0019]

[0020] Among them, B i represents the i-th column vector of the control allocation matrix B, and represents the upper and lower bounds of actuator i;

[0021] The reference point of the three-dimensional reachable moment set is determined by using the control allocation matrix:

[0022]

[0023] Where m represents the total number of executors.

[0024] In one embodiment, before designing a control allocation algorithm based on the achievable virtual control torque as input to calculate the actuator control quantity, the method further includes:

[0025] According to the symmetry of the distributed tilt-rotor UAV, the two rotors on the right are divided into one group, the two rotors on the left are divided into the second group, and the tail rotor is regarded as the third group. The rotor speed and tilt angle in the group are kept consistent to simplify the control distribution model. The simplified control distribution model is:

[0026]

[0027] In one embodiment, the achievable virtual control torque includes a virtual roll control torque, a virtual pitch control torque, and a virtual heading control torque; the actuator control quantity includes a tilt angle and a rotor speed; and a control allocation algorithm is designed to calculate the actuator control quantity based on the achievable virtual control torque as input, including:

[0028] In the control allocation implementation process, a cascade decoupling control allocation algorithm is designed based on the idea of ​​first determining the tilt angle and then solving the rotor speed;

[0029] Calculate the tilt angles of the three sets of rotors according to the virtual heading control torque;

[0030] According to the tilt angle, the problem of solving the underdetermined equations of the rotor speed in the simplified control allocation model is simplified to the problem of solving the well-determined equations, and the control variable of the rotor speed is obtained.

[0031] In one embodiment, calculating the tilt angles of the three sets of rotors based on the virtual heading control torque includes:

[0032] The tilt angles of the three sets of rotors are calculated based on the virtual heading control torque:

[0033]

[0034] Among them, η represents the reference tilt angle, ∈ is a positive proportional coefficient, and Y r Indicates the virtual heading control torque.

[0035] In one embodiment, the problem of solving the underdetermined set of equations for the rotor speed in the simplified control allocation model is simplified to the problem of solving the well-determined set of equations based on the tilt angle, and the control variable of the rotor speed is obtained, including:

[0036] According to the tilt angle, the problem of solving the underdetermined equations of the rotor speed in the simplified control distribution model is simplified to the problem of solving the well-determined equations, and the rotor speed is obtained as

[0037]

[0038] In one embodiment, the process of correcting the first tension constant and the second tension constant according to the virtual control torque includes:

[0039] The method for correcting the first tension constant and the second tension constant according to the virtual control torque is:

[0040]

[0041] Among them, {P r , T r} represents the virtual control torque, and G represents gravity.

[0042] A distributed tilt-rotor UAV control distribution device, comprising:

[0043] A model building module is used to build a control allocation model based on the rotor torque generation mechanism; the control allocation matrix is ​​expressed as a linear control allocation model, and the linear control allocation model includes the control allocation matrix and the actuator control quantity;

[0044] A three-dimensional reachable moment set construction module is used to construct an admissible control set based on multiple actuator control quantities, perform an affine transformation on the admissible control set using a control allocation matrix, and obtain a three-dimensional reachable moment set. The control allocation matrix is ​​used to determine the vertices and multiple reference points of the three-dimensional reachable moment set, and a closed three-dimensional reachable moment set is constructed based on the vertices and reference points.

[0045] The control allocation module is used to find the optimal projection point on the boundary of the closed three-dimensional achievable torque set based on the priority projection algorithm as the achievable virtual control torque. The cascade decoupling control allocation algorithm is designed based on the achievable virtual control torque as input to calculate the control variables of the tilt angle and rotor speed.

[0046] A force constant correction module is used to simplify the control allocation matrix using the reference tilt angle characteristics under the rotor mode, analyze the control allocation matrix, and obtain perturbation parameters; the perturbation parameters include the first force constant and the second force constant; collect virtual control torque in the hovering flight state, and correct the first and second force constants based on the virtual control torque to obtain accurate first and second force constants;

[0047] The relationship matrix solving module is used to solve the control variable of the rotor speed by using the accurate first tension constant and the second tension constant to obtain the rotor speed.

[0048] The above-mentioned distributed tilt-rotor UAV control allocation method and device, this application establishes a system reachable moment set based on spatial geometry, designs a priority-based projection algorithm to alleviate the actuator saturation problem; on this basis, designs a cascade decoupling control allocation algorithm, and designs an online correction method for the uncertainty factors in the allocation model to reduce the control allocation error. Compared with existing control allocation methods, the present invention can improve the control allocation accuracy of tilt-rotor UAVs, reduce the adverse effects of actuator saturation, and enhance flight stability. When applied to a series of tilt-rotor multi-rotor platforms such as distributed tilt-rotor UAVs, it can effectively alleviate the actuator saturation problem, improve control allocation accuracy, and improve system control performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 1 is a flow chart of a distributed tilt-rotor UAV control allocation method according to an embodiment;

[0050] Figure 2 is a schematic diagram of a tilt-rotor UAV model according to one embodiment;

[0051] Figure 3 is a schematic diagram of a reachable moment set in one embodiment;

[0052] Figure 4 Schematic diagram of the implementation process based on the projection algorithm in another embodiment; (a) is the virtual control torque {R r , P r} is the projection implementation process when it is reachable; (b) is the virtual control torque {R r , P r}Projection implementation process when it is unreachable.

[0053] Figure 5 A schematic diagram of a control allocation simulation result based on online correction in one embodiment;

[0054] Figure 6 The figure is a structural block diagram of a distributed tilt-rotor UAV control distribution device in one embodiment. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0056] In one embodiment, Figure 1 As shown, a distributed tilt-rotor UAV control allocation method is provided, comprising the following steps:

[0057] Step 102 : construct a control allocation model based on the rotor torque generation mechanism; express the control allocation matrix as a linear control allocation model, which includes the control allocation matrix and actuator control variables.

[0058] A control allocation model is established for a tilt-rotor UAV. Figure 2 As shown, the UAV has five rotors connected to the servo, and the rotors have the ability to tilt longitudinally. Under the rotor control mechanism, the five rotors and five tilt angles of the tilt-rotor UAV are used for flight control to achieve attitude and altitude tracking. The corresponding virtual control torque includes the rolling torque R r , pitching moment P r , yaw moment Y r and Rally T r The heading control is achieved through the differential adjustment of the tilt angle, and the roll, pitch and altitude control are achieved based on the rotor speed adjustment. i and ω i The tilt angle and speed of the i-th rotor are respectively expressed in rad and rad / s, and the position of the rotor in the body coordinate system is expressed in The four forward rotors have the same configuration, and the thrust constant and torque constant are k f 、k d The tail rotor's pull constant and torque constant are k' f , k′ d According to the rotor torque generation mechanism, the control allocation model is expressed as:

[0059]

[0060] Where c*=cos(*), s*=sin(*). The goal of control allocation is to control the virtual control torque {R r , P r , Y r , T r}, calculate the control amount of the actuator, that is, the tilt angle η i and rotor speed ω i , so that the aircraft provides the corresponding torque to achieve attitude and altitude control. As can be seen from the above equation, the control allocation of tilt-rotor UAV is an underdetermined equation problem with infinite solutions. Because the allocation model is strongly coupled and nonlinear, conventional allocation methods are difficult to directly apply.

[0061] The linear control allocation model is represented as v = Bu, where B is the control allocation matrix and u is the actuator control variable. The resulting set is called the admissible control set. v represents the virtual control torque, and the resulting set is called the achievable moment set. The control allocation matrix and the admissible control set together determine the shape of the achievable moment set. The geometric characteristics of the achievable moment set contain rich information and are important for analyzing the controllability of the system.

[0062] In step 104, an admissible control set is constructed based on multiple actuator control quantities, and an affine transformation is performed on the admissible control set using a control allocation matrix to obtain a three-dimensional reachable moment set. The vertices and multiple reference points of the three-dimensional reachable moment set are determined using the control allocation matrix, and a closed three-dimensional reachable moment set is constructed based on the vertices and reference points.

[0063] This application aims at the most common three-dimensional control allocation problem and proposes a method for constructing a reachable moment set based on spatial geometry. This method has the advantages of being simple and intuitive, with low computational complexity, and a more comprehensive analysis of the geometric characteristics of the reachable moment set, which helps to deepen the understanding of the problem in specific control allocation problems.

[0064] In three-dimensional control allocation, the control set is affine transformed by the control allocation matrix to obtain a three-dimensional reachable moment set. This process can be understood as the column vector of the control allocation matrix is ​​linearly transformed to obtain the reachable moment set. The key to establishing the reachable moment set based on spatial geometry is to determine the vertex of the reachable moment set according to the control allocation matrix. In the control allocation model, B i Denotes the i-th column vector of the allocation matrix B, and the upper and lower bounds of executor i are respectively and Indicates that there is a special vertex p0 in the reachable moment set:

[0065]

[0066] In three-dimensional space, p0 is usually the lowest vertex in the reachable moment set. When a certain actuator takes the upper bound and other actuators take the lower bound, m reference points can be determined. Point p i Expressed as:

[0067]

[0068] Taking point p0 as a new origin, there are m reference vectors Parallel to the boundary of the three-dimensional reachable moment set, all boundary vertices of the reachable moment set can be obtained based on m vectors, and then a closed three-dimensional reachable moment set can be constructed and its volume can be calculated.

[0069] Step 106 , using a priority-based projection algorithm to find the optimal projection point on the boundary of the closed three-dimensional achievable torque set as the achievable virtual control torque, and designing a cascade decoupling control allocation algorithm based on the achievable virtual control torque as input to calculate the control variables of the tilt angle and the rotor speed.

[0070] In conventional three-dimensional control allocation, virtual control torques typically include virtual roll torque, pitch torque, and thrust, which achieve roll, pitch, and altitude control, respectively. In actual flight, attitude control is even more crucial: as long as the attitude is within the effective control range, errors in altitude tracking will not cause the aircraft to roll or crash. When the virtual control torque is outside the reachable torque set, priority is given to minimizing the allocation errors of the virtual roll and pitch torques. Based on this, the optimal projection point on the boundary of the reachable torque set is found as the new virtual control torque.

[0071] For tilt-rotor UAVs, the virtual control unreachable problem usually occurs in large attitude flight or large disturbance conditions. In engineering practice, these two states last for a short time, and the altitude change is small in a short time, so attitude control is more important, that is, virtual control R r With P r The allocation priority is higher than T r Combined with the control characteristics of tilt-rotor UAV, a priority-based projection algorithm is designed. The core idea is: when the virtual control torque given by the controller is unreachable (outside the reachable torque set), find a corresponding point on the reachable torque set so that the two points are in O r R r P r The projection distance of the plane is the smallest.

[0072] When the virtual control torque is on the reachable torque set, control allocation can be performed directly; when the virtual control torque is unreachable, a priority-based projection algorithm is introduced to find a new virtual control torque on the reachable torque set to reach the virtual control torque. According to the above idea, the projection process when the torque is unreachable is as follows: Figure 4 (b) shows that, in addition, Figure 4 (a) is the virtual control torque {R r , P r}Projection implementation process when reachable.

[0073] For a three-point distributed tilt-rotor UAV, combined with the symmetry of the aircraft, the two right rotors are divided into one group, the two left rotors into the second group, and the tail rotor into the third group. The rotor speed and tilt angle within the group are kept consistent, that is, η2 = η1, η4 = η3 and ω2 = ω1, ω4 = ω3. The control allocation model is simplified to:

[0074]

[0075] The rotor mode controls heading through the tilt angle differential. During the control allocation implementation process, the tilt angle is first determined, and then the rotor speed is solved. Based on this idea, a cascade decoupling control allocation algorithm is designed.

[0076] First, consider the heading differential control mechanism, through the virtual heading control torque Y r Calculate the tilt angles of the three sets of rotors:

[0077]

[0078] Where η represents the reference tilt angle and ∈ is a positive proportional coefficient.

[0079] Secondly, after the first step, the rotor tilt angles {η1, η3, η5} can be determined. On this basis, the three sets of rotor speeds {ω1, ω3, ω5} are solved. Then, the problem of solving the underdetermined system of equations is simplified to the problem of solving the exactly determined system of equations. The distribution model can be written as:

[0080]

[0081] Then there is

[0082]

[0083] The control variables for the tilt angle and rotor speed can be obtained through the above two steps. First, the actuators are divided into three groups based on the symmetry of the aircraft, reducing the number of variables to be solved. Then, a decoupling method is designed based on the rotor control characteristics to sequentially calculate the tilt angle and rotor speed. The cascade decoupling control allocation method transforms the complex coupled nonlinear control allocation into a problem requiring the solution of a well-defined set of equations, resulting in analytical expressions for the actuator control variables. This method is computationally efficient and easy to apply in engineering applications.

[0084] Step 108 , using the reference tilt angle characteristics under the rotor mode to simplify the control allocation matrix, analyze the control allocation matrix, and obtain perturbation parameters; the perturbation parameters include a first tension constant and a second tension constant; collect a virtual control torque in the hovering flight state, and correct the first tension constant and the second tension constant according to the virtual control torque to obtain accurate first tension constant and second tension constant.

[0085] Step 110 : Using the accurate first tension constant and the second tension constant, the control variable of the rotor speed is solved to obtain the rotor speed.

[0086] Aiming at the problem of large control allocation errors caused by parameter perturbation, a control allocation parameter correction method is proposed. Under the cascade decoupling control allocation framework, the control allocation matrix is:

[0087]

[0088] The parameters in the control distribution matrix are obtained through identification tests and actual measurements. The torque constant k d , k′ d Much smaller than the tensile constant k f , k′ f ; In the rotor mode, the reference tilt angle η = 0 rad, η1 and η3 are small, then the control allocation matrix in the rotor mode can be simplified to:

[0089]

[0090] The difference between the statically measured tension coefficient and the actual tension coefficient in flight will lead to distribution errors. To address this problem, an online correction method is designed to obtain the tension coefficient under dynamic flight conditions, thereby improving the control distribution accuracy. When the aircraft is hovering stably, the rotor speed ω i and tilt angle η i This fundamental fact remains constant and is unrelated to the accuracy of the control allocation model. In actual systems, the control allocation parameters are fixed values. In a stable hovering state, the three-axis torque generated by the aircraft is zero, and the vertical pull is equal to gravity. Therefore, the control allocation model shows that the rotor speed and differential angle have a unique solution. Based on this finding, an online correction method can be designed to obtain accurate pull constants and reduce control allocation errors.

[0091] When the control distribution parameters are accurate, the virtual control torque is equal to the actual torque, and the actual three-axis torque of the aircraft in the stable hovering state is zero, and the vertical pull is equal to the gravity, that is, when the aircraft is in stable hovering, R r 、P r The theoretical value is zero, T r The theoretical value is gravity G. In a stable hovering state, the perturbation of the tension constant will cause the virtual control torque to deviate from the theoretical value. The tension constant can be continuously corrected according to the virtual control torque. When the virtual control torque is equal to the theoretical value, the corresponding tension constant is the real value.

[0092] The core idea of ​​the online correction method is: in the hovering flight state, the virtual control torque {P r , T r}do

[0093] As a feedback term, the design correction mechanism continuously adjusts the tension constant k f With k f , so that the virtual control torque during hovering approaches the theoretical value. The sign of P r With k f Positive correlation, and k′ f Negative correlation, T r With k f , k′ fBased on the above analysis, the following online correction method can be designed by introducing integrals:

[0094]

[0095] Obtaining accurate control allocation parameters based on flight data in a stable hovering state can reduce the impact of modeling errors, effectively improve control allocation accuracy, and improve flight control performance.

[0096] The corrected first and second tension constants are input into the control variable of the rotor speed to solve for the rotor speed. At this point, the tilt angle and rotor speed are calculated, and the control allocation target is achieved.

[0097] In the aforementioned distributed tilt-rotor UAV control allocation method, this application establishes a system reachable moment set based on spatial geometry and designs a priority-based projection algorithm to alleviate actuator saturation. On this basis, a cascade decoupling control allocation algorithm is designed, and an online correction method is designed for the uncertainty factors in the allocation model to reduce control allocation errors. Compared to existing control allocation methods, the present invention can improve the control allocation accuracy of tilt-rotor UAVs, reduce the adverse effects of actuator saturation, and enhance flight stability. When applied to a series of tilt-rotor multi-rotor platforms, such as distributed tilt-rotor UAVs, it can effectively alleviate actuator saturation, enhance control allocation accuracy, and improve system control performance.

[0098] In one embodiment, a control allocation model is constructed based on a rotor torque generation mechanism, including:

[0099] According to the rotor torque generation mechanism, the control allocation model is constructed as follows:

[0100]

[0101] Among them, R r represents the rolling moment, P r represents the pitching moment, Y r represents the yaw moment, T r represents the tension, η i and ω i Represent the tilt angle and speed of the i-th rotor, k f 、k d represents the thrust constant and torque constant of the forward rotor, represents the position of the rotor in the body coordinate system, k′ f , k′ d represents the tail rotor force constant and torque constant, c*=cos(*), s*=sin(*), η i represents the tilt angle, ω i Indicates the rotor speed.

[0102] In one embodiment, determining the vertices and multiple reference points of a three-dimensional reachable moment set using a control allocation matrix includes:

[0103] The vertices of the three-dimensional reachable moment set are determined using the control allocation matrix:

[0104]

[0105] Among them, B i represents the i-th column vector of the control allocation matrix B, and represents the upper and lower bounds of actuator i;

[0106] The reference point of the three-dimensional reachable moment set is determined by using the control allocation matrix:

[0107]

[0108] Where m represents the total number of executors.

[0109] In one embodiment, before designing a control allocation algorithm based on the achievable virtual control torque as input to calculate the actuator control quantity, the method further includes:

[0110] According to the symmetry of the distributed tilt-rotor UAV, the two rotors on the right are divided into one group, the two rotors on the left are divided into the second group, and the tail rotor is regarded as the third group. The rotor speed and tilt angle in the group are kept consistent to simplify the control distribution model. The simplified control distribution model is:

[0111]

[0112] In a specific embodiment, the rotor speed constraint ω is considered i ≤1070rad / s, ω5≤740rad / s (i=1, 2, 3, 4), simplify the control distribution model, and combine it with the system test parameters to obtain the following specific distribution model:

[0113]

[0114] For the above three-dimensional control allocation problem, a reachable moment set is established based on spatial geometry. First, the vertices of the reachable moment set are found, and then the boundary surface is established to construct the reachable moment set. The vertices of the reachable moment set are:

[0115]

[0116] According to the above vertices, the boundary surface is constructed, and the three-dimensional reachable moment set is as follows: Figure 3 As shown. Given the vertex coordinates of the reachable moment set, the total volume of the reachable moment set is V Φ =1.492×10 6According to the coordinates of the vertices at the boundary of the achievable torque set, the maximum roll torque that the rotor can provide is 60.8 N·m, the maximum pitch torque is 80.7 N·m, and the maximum thrust is 337.2 N. When the virtual control torque provided by the controller exceeds the boundary of the achievable torque set, actuator saturation will occur.

[0117] In one embodiment, the achievable virtual control torque includes a virtual roll control torque, a virtual pitch control torque, and a virtual heading control torque; the actuator control quantity includes a tilt angle and a rotor speed; and a control allocation algorithm is designed to calculate the actuator control quantity based on the achievable virtual control torque as input, including:

[0118] In the control allocation implementation process, a cascade decoupling control allocation algorithm is designed based on the idea of ​​first determining the tilt angle and then solving the rotor speed;

[0119] Calculate the tilt angles of the three sets of rotors according to the virtual heading control torque;

[0120] According to the tilt angle, the problem of solving the underdetermined equations of the rotor speed in the simplified control allocation model is simplified to the problem of solving the well-determined equations, and the control variable of the rotor speed is obtained.

[0121] In one embodiment, calculating the tilt angles of the three sets of rotors based on the virtual heading control torque includes:

[0122] The tilt angles of the three sets of rotors are calculated based on the virtual heading control torque:

[0123]

[0124] Among them, η represents the reference tilt angle, ∈ is a positive proportional coefficient, and Y r Indicates the virtual heading control torque.

[0125] In one embodiment, the problem of solving the underdetermined set of equations for the rotor speed in the simplified control allocation model is simplified to the problem of solving the well-determined set of equations based on the tilt angle, and the control variable of the rotor speed is obtained, including:

[0126] According to the tilt angle, the problem of solving the underdetermined equations of the rotor speed in the simplified control distribution model is simplified to the problem of solving the well-determined equations, and the rotor speed is obtained as

[0127]

[0128] In one embodiment, the process of correcting the first tension constant and the second tension constant according to the virtual control torque includes:

[0129] The method for correcting the first tension constant and the second tension constant according to the virtual control torque is:

[0130]

[0131] Among them, {P r , T r} represents the virtual control torque, and G represents gravity.

[0132] In a specific embodiment, the control allocation matrix includes a thrust constant, a torque constant, a rotor position parameter, etc. The thrust constant is obtained through a static identification test, but in an actual system, the downwash airflow of the front rotor is affected by factors such as the fuselage arm, ground effect, and wind disturbance; in addition, the tail rotor is installed at the tail of the fuselage and is relatively close to the fuselage, so the aerodynamic characteristics of the tail rotor are inevitably affected to a certain extent. Considering the efficiency loss of the rotor, there is a difference between the actual thrust constant and the measured value. Inaccurate thrust constant will lead to control allocation error, thereby affecting the control performance. In the simulation, k f , k′ f Assuming that the measured value of the front rotor and tail rotor force constant is k f1 =k f *90%, k′ f1 =k′ f *110%, an online correction method is introduced based on the cascade decoupling control allocation. The accuracy of the rotor thrust constant mainly affects the height and pitch control effects. To facilitate analysis and discussion, the aircraft takes off to a given altitude in rotor mode and then tracks a sinusoidal signal in the pitch channel. First, a simulation is performed without control allocation parameter correction, recording the state tracking effect and virtual control quantity {P r , T r}; Then, the online correction method is used for simulation. The simulation results without correction are as follows Figure 5 As shown in the solid curve, the test results show that when the aircraft is hovering stably, the virtual control quantity P r =-6.2N·m, T r =151.5N. Affected by the parameter perturbation, the virtual control quantity in the stable hovering state deviates from the theoretical value. After the online correction method is introduced into the simulation, the test results are as follows: Figure 5 As shown by the dotted line, the online correction coefficient when the aircraft is hovering stably is:

[0133]

[0134] Simulation results show that the online correction method can obtain accurate control distribution parameters and the pitch control effect is significantly improved.

[0135] It should be understood that although Figure 1The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0136] In one embodiment, Figure 6 As shown, a distributed tilt-rotor UAV control distribution device is provided, comprising: a model construction module 602, a three-dimensional achievable torque set construction module 604, a control distribution module 606, a tension constant correction module 608 and a relationship matrix solution module 610, wherein:

[0137] A model building module 602 is configured to build a control allocation model based on the rotor torque generation mechanism; the control allocation matrix is ​​represented as a linear control allocation model, the linear control allocation model including the control allocation matrix and the actuator control variable;

[0138] A three-dimensional reachable moment set construction module 604 is configured to construct an admissible control set based on multiple actuator control variables, perform an affine transformation on the admissible control set using a control allocation matrix to obtain a three-dimensional reachable moment set, determine vertices and multiple reference points of the three-dimensional reachable moment set using the control allocation matrix, and construct a closed three-dimensional reachable moment set based on the vertices and reference points.

[0139] A control allocation module 606 is configured to use a priority-based projection algorithm to find an optimal projection point on the boundary of a closed three-dimensional achievable torque set as a achievable virtual control torque, and to design a cascade decoupling control allocation algorithm based on the achievable virtual control torque to calculate the control variables of the tilt angle and the rotor speed;

[0140] The force constant correction module 608 is configured to simplify the control allocation matrix using the reference tilt angle characteristics in the rotor mode, analyze the control allocation matrix, and obtain perturbation parameters; the perturbation parameters include the first force constant and the second force constant; collect virtual control torque in the hovering flight state, and correct the first and second force constants based on the virtual control torque to obtain accurate first and second force constants;

[0141] The relationship matrix solving module 610 is used to solve the control variable of the rotor speed using the accurate first tension constant and the second tension constant to obtain the rotor speed.

[0142] The specific definitions of a distributed tilt-rotor UAV control distribution device can be found in the definitions of a distributed tilt-rotor UAV control distribution method described above and will not be repeated here. Each module in the above-mentioned distributed tilt-rotor UAV control distribution device can be implemented in whole or in part through software, hardware, or a combination thereof. Each of the above-mentioned modules can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a computer device's memory in software form, so that the processor can call and execute the corresponding operations of each of the above modules.

[0143] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0144] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A distributed tilt-rotor UAV control allocation method, characterized in that: The method comprises: Construct a control allocation model based on the rotor torque generation mechanism; The control allocation model is expressed as a linear control allocation model, wherein the linear control allocation model includes a control allocation matrix and an actuator control variable; Constructing an admissible control set according to a plurality of actuator control quantities, and performing an affine transformation on the admissible control set using the control allocation matrix to obtain a three-dimensional reachable moment set; Determining vertices and a plurality of reference points of the three-dimensional reachable moment set using the control allocation matrix, and constructing a closed three-dimensional reachable moment set based on the vertices and the reference points; The priority-based projection algorithm finds the best projection point on the boundary of the closed three-dimensional reachable moment set as the reachable virtual control moment, including: The priority is the virtual control torque {R r , P r } The allocation priority is higher than the attitude control torque; when the virtual control torque given by the controller is unreachable, that is, outside the three-dimensional reachable torque set, find a corresponding point on the three-dimensional reachable torque set as the best projection point so that the two points are at O r R r P r The projection distance of the plane is minimized; a cascade decoupling control allocation algorithm is designed based on the achievable virtual control torque as input to calculate the control amount of the tilt angle and the rotor speed; The control allocation matrix is ​​simplified by using the reference tilt angle characteristic under the rotor mode, and the control allocation matrix is ​​analyzed to obtain perturbation parameters; the perturbation parameters include a first tension constant and a second tension constant; collecting a virtual control torque in a hovering flight state, and correcting the first tension constant and the second tension constant according to the virtual control torque to obtain accurate first tension constant and second tension constant; The accurate first tension constant and the second tension constant are used to solve the control variable of the rotor speed to obtain the rotor speed.

2. The method according to claim 1, characterized in that A control allocation model is constructed based on the rotor torque generation mechanism, including: According to the rotor torque generation mechanism, the control allocation model is constructed as follows: Among them, R r represents the rolling moment, P r represents the pitching moment, Y r represents the yaw moment, T r represents the tension, η i and ω i Represent the tilt angle and speed of the i-th rotor, k f 、k d represents the thrust constant and torque constant of the forward rotor, represents the position of the rotor in the body coordinate system, k′ f , k′ d represents the tail rotor force constant and torque constant, c*=cos(*), s*=sin(*), η i represents the tilt angle, ω i Indicates the rotor speed.

3. The method according to claim 1, characterized in that Determining the vertices and multiple reference points of the three-dimensional reachable moment set using the control allocation matrix includes: The vertices of the three-dimensional reachable moment set are determined using the control allocation matrix as Among them, B i represents the i-th column vector of the control allocation matrix B, and represents the upper and lower bounds of actuator i; The reference point of the three-dimensional achievable moment set is determined by using the control allocation matrix: Where m represents the total number of executors.

4. The method according to claim 2, characterized in that Before designing a control allocation algorithm based on the achievable virtual control torque as input to calculate the actuator control quantity, the method further includes: According to the symmetry of the distributed tilt-rotor UAV, the two rotors on the right are divided into one group, the two rotors on the left are divided into the second group, and the tail rotor is regarded as the third group. The rotor speed and tilt angle in the group are kept consistent. The control distribution model is simplified to obtain the simplified control distribution model:

5. The method according to claim 2, characterized in that The achievable virtual control torque includes a virtual roll control torque, a virtual pitch control torque and a virtual heading control torque; the actuator control quantity includes a tilt angle and a rotor speed; A control allocation algorithm is designed to calculate the actuator control quantity based on the achievable virtual control torque as input, including: In the control allocation implementation process, a cascade decoupling control allocation algorithm is designed based on the idea of ​​first determining the tilt angle and then solving the rotor speed; Calculating the tilt angles of the three sets of rotors according to the virtual heading control torque; According to the tilt angle, the problem of solving the underdetermined equations of the rotor speed in the simplified control allocation model is simplified to the problem of solving the well-determined equations, and the control variable of the rotor speed is obtained.

6. The method according to claim 5, characterized in that Calculating the tilt angles of the three sets of rotors according to the virtual heading control torque includes: The tilt angles of the three sets of rotors are calculated based on the virtual heading control torque: Among them, η represents the reference tilt angle, ∈ is a positive proportional coefficient, and Y r Indicates the virtual heading control torque.

7. The method according to claim 5, characterized in that According to the tilt angle, the problem of solving the underdetermined set of equations for the rotor speed in the simplified control allocation model is simplified to the problem of solving the well-determined set of equations, and the control variable of the rotor speed is obtained, including: According to the tilt angle, the problem of solving the underdetermined equations of the rotor speed in the simplified control distribution model is simplified to the problem of solving the well-determined equations, and the rotor speed is obtained as follows:

8. The method according to claim 7, characterized in that The process of correcting the first tension constant and the second tension constant according to the virtual control torque includes: The method for correcting the first tension constant and the second tension constant according to the virtual control torque is: Among them, {P r , T r } represents the virtual control torque, and G represents gravity.

9. A distributed tilt-rotor UAV control distribution device, characterized in that: The device comprises: A model building module is used to build a control allocation model based on the rotor torque generation mechanism; the control allocation model is expressed as a linear control allocation model, the linear control allocation model includes a control allocation matrix and an actuator control variable; a three-dimensional reachable moment set construction module, configured to construct an admissible control set based on a plurality of actuator control quantities, perform an affine transformation on the admissible control set using the control allocation matrix to obtain a three-dimensional reachable moment set; determine vertices and a plurality of reference points of the three-dimensional reachable moment set using the control allocation matrix, and construct a closed three-dimensional reachable moment set based on the vertices and reference points; A control allocation module, configured to find an optimal projection point as a reachable virtual control moment on the boundary of the closed three-dimensional reachable moment set based on a priority projection algorithm, comprising: The priority is the virtual control torque {R r , P r } The allocation priority is higher than the attitude control torque; when the virtual control torque given by the controller is unreachable, that is, outside the three-dimensional reachable torque set, find a corresponding point on the three-dimensional reachable torque set as the best projection point so that the two points are at O r R r P r The projection distance of the plane is the smallest; Designing a cascade decoupling control allocation algorithm based on the achievable virtual control torque as input to calculate the control variables of the tilt angle and the rotor speed; a force constant correction module configured to simplify a control allocation matrix using a reference tilt angle characteristic under a rotor mode, analyze the control allocation matrix, and obtain perturbation parameters, wherein the perturbation parameters include a first force constant and a second force constant; collect a virtual control torque in a hovering flight state, and correct the first and second force constants based on the virtual control torque to obtain accurate first and second force constants; The relationship matrix solving module is used to solve the control variable of the rotor speed by using the accurate first tension constant and the second tension constant to obtain the rotor speed.

Images