A controller parameter tuning method and system for a quad-rotor unmanned aerial vehicle in a forward flight mode

Abstract

The application discloses a controller parameter setting method and system for a quad-rotor unmanned aerial vehicle in a forward flight mode, the method comprising the following steps: S1, frequency domain identification is performed on the quad-rotor unmanned aerial vehicle in the forward flight mode, and a forward flight dynamics model of the quad-rotor unmanned aerial vehicle is obtained; and S2, based on the forward flight dynamics model, pole placement analytical method is used to set parameters of an attitude control cascade PID controller. The method provided by the application has the advantages that in terms of time domain response, control instruction tracking is smooth, there is no static error and oscillation phenomenon, and the response speed is relatively fast, and the rising time is about 0.3s; in terms of frequency domain response, the system is stable in convergence, the open loop amplitude margin and the phase margin are sufficient, the relative stability is relatively strong, and the method has certain robustness.

Description

Technical Field

[0001] This invention relates to the field of aircraft technology, and in particular to a method and system for tuning controller parameters in the forward flight mode of a quadcopter unmanned aerial vehicle. Background Technology

[0002] A quadcopter drone is a four-channel input, six-degree-of-freedom underactuated system. Its flight controller mostly uses Pixhawk products. Its native firmware uses a cascaded PID algorithm to achieve stable and precise control of the multirotor's attitude and position. By adjusting the deviations of attitude angular motion information (angle and angular velocity) or flight position motion information (position and velocity) using proportional, integral, and derivative algorithms, the steady-state and dynamic characteristics of the control system are improved, thereby achieving good control effects such as fast response, no overshoot, and disturbance rejection.

[0003] In engineering applications, most designers today, when solving practical quadcopter drone control problems, directly adopt a continuous experimental and iterative approach of "manual adjustment + flight verification" to obtain the final PID parameters. In fact, most literature on PID control for quadcopter drones does not cover parameter tuning; after deriving the mathematical model, it directly provides the specific PID parameters, resulting in suboptimal control performance.

[0004] Furthermore, traditional research on the dynamics modeling of quadrotor UAVs, both domestically and internationally, is mostly based on the assumption of hovering flight conditions. However, during forward flight, both the propeller dynamics and fuselage dynamics undergo certain changes, making them unsuitable for controller parameter design in forward flight mode. As the missions of quadrotor UAVs expand in civilian and military fields such as combat strikes and racing, higher demands are placed on the stability and speed of their control systems in forward flight mode. However, currently, there is a lack of systematic research on quadrotor UAV models in forward flight mode, both domestically and internationally. There is an urgent need to develop a complete and accurate modeling method and experimental system for quadrotor UAV forward flight.

[0005] Therefore, existing controller design methods based on hovering mode and manual parameter tuning have certain technical defects and cannot meet the control performance requirements of high-speed forward flight processes with wide application scenarios. Summary of the Invention

[0006] To address the problems existing in the prior art, this invention provides a method and system for tuning controller parameters in the forward flight mode of a quadcopter unmanned aerial vehicle.

[0007] In a first aspect, the present invention provides a method for tuning controller parameters in the forward flight mode of a quadcopter unmanned aerial vehicle, comprising:

[0008] S1: Frequency domain identification of the quadrotor UAV in forward flight mode is performed to obtain the forward flight dynamics model of the quadrotor UAV.

[0009] S2: Based on the forward flight dynamics model, the parameters of the attitude control cascade PID controller are tuned using the pole placement analytical method.

[0010] Secondly, the present invention provides a controller parameter tuning system for the forward flight mode of a quadcopter unmanned aerial vehicle, comprising:

[0011] Model building module: It performs frequency domain identification on the quadcopter UAV in forward flight mode to obtain the forward flight dynamics model of the quadcopter UAV;

[0012] Parameter tuning module: Based on the forward flight dynamics model, it uses the pole placement analytical method to tune the parameters of the attitude control cascade PID controller.

[0013] The beneficial effects of the controller parameter tuning method and system for a quadcopter unmanned aerial vehicle in forward flight mode according to the present invention include:

[0014] (1) The controller parameter tuning method for the forward flight mode of a quadcopter UAV provided by the present invention can establish a relatively accurate dynamic model of the forward flight mode of a quadcopter UAV.

[0015] (2) The controller parameter tuning method for the forward flight mode of the quadrotor UAV provided by the present invention, based on the pole placement analytical method of frequency domain analysis, can theoretically realize the parameter design of the cascade PID controller for attitude control of the quadrotor UAV. At the same time, combined with the actual use requirements and the conventional characteristics of the cascade PID controller, it proposes reasonable design indicators for parameter design.

[0016] (3) The controller parameter tuning method for the forward flight mode of the quadcopter UAV provided by the present invention has obvious advantages over manual parameter tuning. In terms of time domain response, the control command tracking is smooth, without steady-state error and oscillation, and the response speed is fast, with a rise time of about 0.3s. In terms of frequency domain response, the system converges and is stable, with sufficient open-loop amplitude margin Gm and phase margin Pm, relatively strong stability, and certain robust performance. Attached Figure Description

[0017] Figure 1 A flowchart illustrating a method for tuning controller parameters in the forward flight mode of a quadcopter unmanned aerial vehicle according to the present invention is shown.

[0018] Figure 2 This diagram illustrates the structure of a controller parameter tuning system for the forward flight mode of a quadcopter unmanned aerial vehicle (UAV) according to the present invention.

[0019] Figure 3 a) A diagram showing the swept-frequency signal input data of the forward flight mode of the quadcopter UAV, an experimental example of the present invention;

[0020] Figure 3b) A schematic diagram of the structure of a quadcopter UAV, an experimental example of the present invention, is shown;

[0021] Figure 3 c) A schematic diagram showing the main data recording of the forward flight mode of the quadcopter UAV in the experimental example of the present invention;

[0022] Figure 3 d) Shows the fitting data graph of the identification results of the forward flight mode of the quadcopter UAV in the experimental example of the present invention;

[0023] Figure 3 e) A time-domain verification data diagram of the forward flight mode of the quadcopter UAV, an experimental example of the present invention, is shown;

[0024] Figure 4 The figure shows the step response curve of the roll channel of the quadcopter UAV in the forward flight mode of the present invention under a roll angle excitation of 15°.

[0025] Figure 5 The open-loop Bode plot of the roll path of the quadcopter UAV in the forward flight mode of the experimental example of the present invention is shown. Detailed Implementation

[0026] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby providing a clearer and more explicit definition of the scope of protection of the present invention.

[0027] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0028] Figure 1 A flowchart illustrating a controller parameter tuning method for a quadcopter unmanned aerial vehicle (UAV) in forward flight mode according to the present invention is shown. The method may include:

[0029] S1: Frequency domain identification of the quadrotor UAV in forward flight mode is performed to obtain the forward flight dynamics model of the quadrotor UAV.

[0030] In a preferred embodiment of the present invention, S1 may include the following sub-steps:

[0031] S1-1: In the forward flight mode of a quadcopter drone, control the quadcopter drone to reach the desired forward flight speed.

[0032] S1-2: When the desired forward speed is reached, the frequency domain of the roll, pitch, yaw and vertical channels of the quadcopter UAV is identified using a preset frequency sweep signal.

[0033] In this invention, after the quadcopter UAV reaches the expected forward flight state, i.e., when it is at a stable desired forward flight speed, the roll, pitch, yaw, and vertical channels are fully excited using a preset frequency sweep signal, and the command input (the remote controller input representing the desired Euler angle and the preset frequency sweep signal input) and the output of the cascade PID controller (the control commands for the roll, pitch, yaw, and vertical channels, [δ]) are recorded. lat ,δ lon ,δ ped ,δ col The data includes signals / data such as flight status (acceleration, attitude angle / angular rate, flight altitude, GPS (position / velocity)). Then, using CIFER software and a frequency domain transfer function identification algorithm, the open-loop dynamic model of the quadcopter UAV in forward flight mode is calculated.

[0034] In a preferred embodiment of the present invention, the preset frequency sweep signal is represented by Equation 1:

[0035]

[0036] Among them, R sweep This represents a swept frequency signal, where A represents the amplitude and T represents the frequency sweep signal. rec ω represents the duration of the frequency sweep signal. min and ω max These represent the minimum and maximum frequencies of the swept frequency signal, respectively. C1 and C2 are constants.

[0037] Preferably, A is 0–0.3491 rad, meaning the attitude angle input will be adjusted between 0 and 20° depending on the forward flight speed conditions. Considering the sufficiency of command response and the limitation of communication distance after the quadcopter UAV maneuvers, T rec It lasts for 40 to 60 seconds.

[0038] In this invention, since the cross-pass frequency of the quadcopter UAV is between 10 and 20 rad / s, ω is preferred. min =0.6 rad / s and ω max = 62.8 rad / s.

[0039] In this invention, due to the system identification principle and the actual situation of quadcopter UAV in the frequency domain identification process, C1 is preferably 4 and C2 is preferably 0.0187.

[0040] In this invention, by performing frequency domain identification on the quadcopter UAV and fitting the recorded data in the input-output frequency domain, a forward flight dynamics model of the quadcopter UAV is obtained. In the forward flight mode, the dynamics model of the quadcopter UAV is a multiple-input multiple-output system, which can be decoupled into a longitudinal model and a lateral model. In the longitudinal model, the pitch axis and altitude axis are coupled, and in the lateral model, the roll axis and yaw axis are coupled, as shown in Equations 3 and 4.

[0041] Longitudinal model:

[0042]

[0043] Lateral model:

[0044]

[0045] In the above two equations, u represents the linear velocity along the x-axis, w represents the linear velocity along the z-axis, v represents the linear velocity along the y-axis, q represents the pitch rate, p represents the roll rate, r represents the yaw rate, and θ represents the pitch angle. ψ represents the roll angle, δ represents the yaw angle, and δ represents the roll angle. lon Indicates the control command for the pitch channel, δ col Indicates the control command for the vertical channel, δ lat The control command for the roll channel, δ ped The control command for the yaw channel, X u M represents the derivative of the aerodynamic force X with respect to u along the x-axis of the machine system. u M represents the derivative of the aerodynamic torque M with respect to u along the y-axis of the machine system. w Let w0 represent the derivative of the aerodynamic torque M with respect to w in the y-axis direction of the aircraft system, w0 represent the instantaneous equilibrium velocity of the aircraft system in the z-axis direction at the expected forward velocity, u0 represent the instantaneous equilibrium velocity of the aircraft system in the x-axis direction at the expected forward velocity, and θ0 represent the pitch angle in the instantaneous equilibrium direction at the expected forward velocity. The aerodynamic force X along the x-axis of the machine system is related to δ. lon The derivative, The aerodynamic torque M of the machine system in the y-axis direction is expressed as a function of δ. lon The derivative, The aerodynamic force Z relative to δ along the z-axis of the machine system is represented. col The derivative of Y v L represents the derivative of the aerodynamic force Y with respect to v along the y-axis of the machine system. v N represents the derivative of the aerodynamic torque L with respect to v along the x-axis of the engine system. r Let N be the derivative of the aerodynamic force N with respect to r along the z-axis of the machine system. The aerodynamic force Y in the y-axis direction of the machine system is related to δ. lat The derivative, The aerodynamic torque L in the x-axis direction of the engine system is expressed as a function of δ. lat The derivative, The aerodynamic force N in the z-axis direction of the machine system is relative to δ. ped The derivative of .

[0046] Since the flight attitude of a quadcopter drone is controlled by a cascaded PID controller adjusting the speeds of its four motors, optimizing the cascaded PID controller is necessary to enhance its stability and reliability. Therefore, this invention proposes using the pole placement analytical method to tune the parameters of the cascaded PID control, thereby achieving optimal control system output.

[0047] In this invention, in order to quickly analyze the parameters of the cascade PID controller using the pole placement analytical method, in a preferred embodiment of this invention, S1 further includes: decoupling the forward flight dynamics model of the quadcopter UAV to obtain the open-loop transfer functions of the roll, pitch, yaw and vertical channels.

[0048] In other words, by decoupling Equations 3 and 4, and based on the transformation relationship between the state space and the transfer function, the forward flight dynamics model is decoupled into the following four single-channel SISO linear models, namely the open-loop transfer functions of the roll, pitch, yaw, and vertical channels, i.e., Equation 2:

[0049]

[0050] Among them, G roll Let G represent the transfer function of the roll channel, s represent the independent variable in the complex frequency domain, and G represent the transfer function of the roll channel. pitch G represents the open-loop transfer function of the pitch channel. yaw G represents the open-loop transfer function of the yaw channel. ver This represents the open-loop transfer function of the vertical channel.

[0051] By excitation with a preset frequency sweep signal, a relatively accurate forward flight dynamics model of a quadcopter UAV can be obtained.

[0052] The method for tuning controller parameters in the forward flight mode of a quadcopter UAV according to the present invention further includes:

[0053] S2: Based on the forward flight dynamics model, the parameters of the cascade PID controller are tuned using the pole placement analytical method.

[0054] In a preferred embodiment of the present invention, S2 may include the following sub-steps:

[0055] S2-1: Based on the actual situation of the quadcopter UAV and the time-frequency characteristics of the cascade PID controller, determine the desired closed-loop transfer function of the quadcopter UAV to meet the design specifications.

[0056] In this invention, taking the roll channel as an example, the transfer function of the cascade PID controller is as follows:

[0057]

[0058] Among them, K P K I K D These are the control parameters for the proportional, integral, and derivative components, respectively, with K being the outer-loop control parameter. According to G... roll The time-frequency characteristics of the cascaded PID controller, taking into account the uncorrected open-loop transfer function G roll The molecule contains a zero and a differential element. The differential element exactly cancels out the integral element in the PID controller, while the zero... It will still exist, and it is worth noting that this zero-point part will still exist in the numerator when solving the entire closed-loop transfer function of the roll channel, and will not change with the control parameters. Therefore, the closed-loop transfer function of the roll channel is obtained by rearranging:

[0059]

[0060] Among them, A i (i = 0, 1, 2, 3, 4) are the simplification coefficients in the denominator during the solution process, derived from the four parameters (K) of the cascade PID controller. P K I K D It consists of K and kinetic coefficients.

[0061] S2-2: Based on the inner and outer loops and design specifications of the cascade PID controller, determine the poles in the closed-loop transfer function, and use the pole placement analytical method to tune the parameters of the cascade PID controller.

[0062] As can be seen from Equation 6 above, the closed-loop system has three zeros, two of which are determined by the inner-loop PID parameters, and the third zero is... Therefore, by designing a low-frequency pole in the denominator to a value close to that constant zero, its component in the time-domain response can be weakened as much as possible, thus achieving the effect of approximate zero-pole cancellation.

[0063] After cancellation, the pole distribution of the system's closed-loop transfer function consists of a low-frequency dominant pole and a pair of high-frequency conjugate poles. Considering the desired time-domain and frequency-domain characteristics, the present invention selects to configure the final closed-loop system poles as a low-frequency dominant pole and a pair of high-frequency conjugate poles. That is, the corrected closed-loop transfer function becomes:

[0064]

[0065] Based on the principle that the corresponding coefficients in the characteristic equation (i.e., the denominator) of the closed-loop transfer function are equal, the analytical expression between the PID parameters to be designed and the desired design index can be obtained, as shown in the following equation:

[0066]

[0067] Where ω is the undamped natural frequency of the high-frequency conjugate pole, ζ is the damping ratio of the non-dominant second-order oscillatory element, τ1 is the time constant of the low-frequency dominant pole, and τ2 is the pole time constant that cancels out the zeros in the molecule, with τ1 < τ2, and

[0068] Therefore, it can be seen that the parameters (K) of the cascade PID controller P K I K D The parameters (K) of the cascade PID controller are obtained by solving for the time constant, damping ratio, and undamped natural frequency. In other words, the parameters (K) can be adjusted by changing the time constants τ1 and τ2, the damping ratio ζ, and the undamped natural frequency ω. P K I K D The tuning of K).

[0069] Similarly, the other two attitude control channels, namely pitch and yaw, can also be designed and tuned according to the above steps since the linearized model is similar to the roll channel, and will not be elaborated here.

[0070] In a preferred embodiment of the present invention, the design and selection process for time constants τ1 and τ2, damping ratio ζ, and undamped natural frequency ω is as follows:

[0071] (1) The time constant τ1 mainly determines the transient response speed of the control system. Considering the relationship between the response speeds of position control and attitude control, as well as the actual situation of the quadcopter UAV, the preferred value of the time constant τ1 is 0.3-0.5s.

[0072] (2) The damping ratio ζ mainly determines the oscillation characteristics of the second-order element. When ζ is 0.7, the amplitude characteristic of the frequency response is the smoothest, so ζ is preferably 0.7.

[0073] (3) The undamped natural frequency ω affects the system's gain and response speed. When the time constants τ1, τ2, and damping ratio ζ are determined, the undamped natural frequency ω and the system's open-loop crossover frequency ω... c Correlation, i.e., ω c =f(ω), which can be adjusted based on the frequency domain Bode plot of the system after the preliminary design, and the undamped natural frequency ω is finally determined through multiple rounds of iteration.

[0074] Preferably, the controller parameter tuning method for the forward flight mode of the quadcopter UAV of the present invention may further include:

[0075] S3: Perform time-domain characteristic analysis and parameter optimization iteration on the tuned cascade PID controller.

[0076] In this invention, after tuning the parameters of the cascaded PID controller, the time-domain and frequency-domain characteristics of the system are analyzed, and then the optimal parameters are finally determined after multiple tuning iterations.

[0077] Figure 2 This diagram illustrates a controller parameter tuning system for the forward flight mode of a quadcopter UAV, provided by the present invention. This system can be applied to terminals or servers capable of data processing, such as cloud or local servers. The system mainly includes:

[0078] Model building module 201: It performs frequency domain identification on the quadrotor UAV in forward flight mode to obtain the forward flight dynamics model of the quadrotor UAV;

[0079] Parameter tuning module 202: Based on the forward flight dynamics model, it uses the pole placement analytical method to tune the parameters of the attitude control cascade PID controller.

[0080] The controller parameter tuning system for the forward flight mode of a quadcopter UAV provided by this invention can be used to execute the controller parameter tuning method for the forward flight mode of a quadcopter UAV described above. Its implementation principle and technical effect are similar, and will not be repeated here.

[0081] Preferably, the model building module 201 and the parameter tuning module 202 in the controller parameter tuning system for the forward flight mode of a quadcopter UAV of the present invention can be directly in hardware, in a software module executed by a processor, or in a combination of both.

[0082] Software modules may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in this art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium.

[0083] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or any combination thereof. A general-purpose processor can be a microprocessor, but alternatively, it can be any conventional processor, controller, microcontroller, or state machine. The processor can also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors incorporating a DSP core, or any other such configuration. Alternatively, the storage medium can be integrated with the processor. The processor and storage medium can reside in an ASIC. The ASIC can reside in the user terminal. Alternatively, the processor and storage medium can reside as discrete components in the user terminal.

[0084] In one embodiment of the present invention, a real-time simulation system for a quadcopter unmanned aerial vehicle (UAV) is provided, which executes the controller parameter tuning method for the forward flight mode of a quadcopter UAV described in the present invention.

[0085] In one embodiment of the present invention, a computer-readable storage medium stores computer instructions that are operated to perform the controller parameter tuning method for the forward flight mode of a quadcopter UAV described in the present invention.

[0086] In one embodiment of the present invention, a program product includes a computer program stored in a readable storage medium, at least one processor can read the computer program from the readable storage medium, and at least one processor executes the computer program to perform the controller parameter tuning method for the forward flight mode of a quadcopter UAV described in the present invention.

[0087] In the several embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0088] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0089] Experimental Example

[0090] Taking a quadcopter drone as an example, equipped with a Pixhawk Cub 2.0 flight controller, a frequency domain identification experiment was conducted under the condition of a forward flight speed of 10 m / s. The experimental process and results are as follows: Figure 3 As shown in (a~d); Figure 3 a) is a graph of the sweep frequency signal input data; Figure 3 b) is a schematic diagram of the structure of a quadcopter drone; Figure 3 c) is a schematic diagram of the main data records; Figure 3 d) Fitted data plot for the identification results; Figure 3 e) is a time-domain verification data graph.

[0091] The preset sweep frequency signal is represented by Equation 1:

[0092]

[0093] Among them, R sweep This represents a frequency sweep signal, A = 0.17, T rec =60s, ω min =0.6 rad / s and ω max =62.8 rad / s, C1 = 4, C2 = 0.0187.

[0094] By processing the input and output data using CIFER software, the forward flight dynamics model of the quadcopter UAV is obtained, as shown in Equations 3 and 4:

[0095] Longitudinal model:

[0096]

[0097] Lateral model:

[0098]

[0099] Decoupling equations 3 and 4 yields the open-loop transfer functions for roll, pitch, yaw, and vertical channels, i.e., equation 2:

[0100]

[0101] The model measurement results of the frequency domain identification experiment at a forward speed of 10 m / s are shown in Table 1, including aerodynamic derivatives and control derivatives. Table 1 shows that the relative uncertainty (CR%) of each model parameter is less than 20%, and the relative insensitivity (Insen%) is less than 10%, indicating the accuracy of the model identification. Specific results are shown in Table 1.

[0102] Table 1. Parameter identification results of the forward flight dynamics model at 10 m / s

[0103]

[0104] Taking the roll channel as an example, according to G roll By combining the time-frequency characteristics of the cascaded PID controller, the closed-loop transfer function of the roll channel is obtained:

[0105]

[0106] The final closed-loop system poles are configured as a low-frequency dominant pole and a pair of high-frequency conjugate poles, meaning the corrected closed-loop transfer function becomes:

[0107]

[0108] Thus, the analytical expression between the PID parameters to be designed and the desired design index is obtained, as shown in the following formula:

[0109]

[0110] Where ω is the undamped natural frequency of the high-frequency conjugate pole, chosen to be 5 rad / s, ζ = 0.7, and τ1 = 0.3 s.

[0111] The final four parameters K of the cascade PID controller for the roll channel are obtained. P K I K D The values ​​of K are 0.119, 0.888, 0.004 and 2.549, respectively.

[0112] Similarly, the four parameters of the cascaded PID controllers for the pitch and yaw channels can be obtained, and the specific results are shown in Table 2.

[0113] Table 2. Parameter tuning results and manual parameter tuning results of the cascade PID controller of the present invention.

[0114]

[0115] In addition, based on the MATLAB / Simulink environment, the parameters obtained by the two different methods were used to perform time-domain and frequency-domain simulation analysis of the control system. The fourth-order Runge-Kutta method with a fixed step size was selected as the numerical iterative algorithm, with a step size of 0.001s.

[0116] Taking the roll-through channel as an example, such as Figure 4 The figure shows two sets of parameters (K of this invention). P K I K D K and K are 0.119, 0.888, 0.004 and 2.549 respectively, and K is manually tuned. P K I K D The step response results of the roll channel under a 15° Euler angle excitation with K values ​​of 0.1, 0.09, 0.005, and 4, respectively. From... Figure 4 As can be seen from the data, the roll channel rise time t of the two methods is... 63 The steady-state time is approximately 0.3s, which is quite close. However, the response of the present invention is smoother, without significant changes in the output response. In terms of steady-state performance, both methods show no steady-state error, but the steady-state time t of the present invention is... 90 The transition time is 1.33s, which is significantly better than the 3.57s achieved by manual parameter tuning in the experiment, and there is no obvious oscillation during the transition process.

[0117] This is because the response curve based on the tuning parameters of this invention is equivalent to a time constant of exactly t. 63 The response of the first-order dynamic model is thus a smooth step response in the time domain, with no steady-state error and no overshoot or oscillation. The response results using the higher-order dynamic model in this invention exhibit a typical separation point, namely t... 63 Near the location, in t 63 Previously, it was more suitable for short-cycle, rapid responses because the transition process was fast and ideal, but for cycles exceeding t... 63 Subsequently, the system is affected by long-period poles, causing the steady-state process to slow down. In summary, this invention exhibits smooth instruction tracking, no steady-state error or oscillation in the time domain, and a fast response speed with a rise time of approximately 0.3s.

[0118] Meanwhile, to better analyze the stability performance of the control system in the frequency domain, this invention plots its open-loop Bode plot, as shown below. Figure 5 The figure shows two sets of parameters (K of this invention). P K I K D K and K are 0.119, 0.888, 0.004 and 2.549 respectively, and K is manually tuned. P K I K D The open-loop Bode plot frequency domain characteristic curves of the roll channel are shown for K values ​​of 0.1, 0.09, 0.005, and 4, respectively. For this invention, the gain margin Gm is much greater than 6 dB, enabling the system to converge quickly. Simultaneously, it possesses a large phase margin Pm of 97.3° at a frequency of 2.73 rad / s. This is because the breakpoint of the open-loop model for the quadcopter UAV attitude control is at the outer loop feedback point, not at the actuator dynamics point, i.e., the point where the system's open-loop characteristics are worst. Therefore, sufficient phase margin must be reserved in the design to ensure relative stability. Overall, the system of this invention has sufficient open-loop gain and phase margins, strong relative stability, and certain robustness.

[0119] The present invention has been described above with reference to preferred embodiments; however, these embodiments are merely exemplary and illustrative. Various substitutions and modifications can be made to the present invention based on these embodiments, all of which fall within the scope of protection of the present invention.

Claims

1. A method for tuning controller parameters in the forward flight mode of a quadcopter unmanned aerial vehicle (UAV), characterized in that, include: S 1: Frequency domain identification is performed on the quadrotor UAV in forward flight mode to obtain the forward flight dynamics model of the quadrotor UAV. The forward flight dynamics model of the quadrotor UAV is decoupled to obtain the open-loop transfer functions of roll, pitch, yaw and vertical channels. The open-loop transfer functions of the roll, pitch, yaw, and vertical channels are expressed by Equation 2: Formula 2 in, G roll This represents the open-loop transfer function of the roll channel. p Indicates the roll rate. δ lat This indicates the control command for the roll-through channel. s Represents the independent variable in the complex frequency domain. L δlat Indicator System x Axial aerodynamic torque L right δ lat The derivative of L v Indicator System x Axial aerodynamic torque L right y linear velocity of the shaft v The derivative of Y δlat Indicator System y Axial aerodynamic forces Y right δ lat The derivative of Y v Indicator System y Axial aerodynamic forces Y right v The derivative of w 0 represents the instantaneous equilibrium of the aircraft system at the desired forward flight speed. z Flight speed in the axial direction θ 0 represents the pitch angle at instantaneous equilibrium when the desired forward speed is reached; G pitch This represents the open-loop transfer function of the pitch channel. q Indicates pitch rate, δ lon Indicates control commands for the pitch channel. M δlon Indicator System y Axial aerodynamic torque M right δ lon The derivative of M u Indicator System y Axial aerodynamic torque M right x linear velocity of the shaft u The derivative of Indicator System x Axial aerodynamic forces X right δ lon The derivative of X u Indicator System x Axial aerodynamic forces X right u The derivative of M w Indicator System y Axial aerodynamic torque M right z linear velocity of the shaft w The derivative of u 0 represents the instantaneous equilibrium of the aircraft system at the desired forward flight speed. x Flight speed in the axial direction; G yaw This represents the open-loop transfer function of the yaw path. r Indicates the yaw rate. δ ped Control commands indicating the yaw channel. N ped Indicator System z Axial aerodynamic forces N right δ ped The derivative of N r Indicator System z Axial aerodynamic forces N right r The derivative; G ver Represents the open-loop transfer function of the vertical channel. w express z The linear velocity of the shaft, δ col Indicates control commands for the vertical channel. Z col Indicator System z Axial aerodynamic forces Z right δ col The derivative; S 1 includes: S 1-1: In the forward flight mode of the quadcopter drone, control the quadcopter drone to reach the desired forward flight speed; S 1-2: At the desired forward flight speed, the roll, pitch, yaw, and vertical channels of the quadcopter UAV are identified in the frequency domain using a preset frequency sweep signal. The preset frequency sweep signal is represented by Equation 1: Set 1 in, R sweep Indicates a frequency sweep signal. A Indicates amplitude. T rec Indicates the duration of the frequency sweep signal. ω min and ω max These represent the minimum and maximum frequencies of the swept frequency signal, respectively. , C 1 and C 2 It is a constant value; S 2: Based on the aforementioned forward flight dynamics model, the attitude control cascade is performed using the pole placement analytical method. PID The controller parameters are tuned, including: S 2-1: Based on the actual situation and attitude control cascade of the quadcopter UAV PID The time-frequency characteristics of the controller are used to determine the desired closed-loop transfer function of the quadcopter UAV; S 2-2: Based on cascade PID The controller's inner and outer loops and design specifications are determined, the pole distribution is identified, and the cascade configuration is analyzed using the pole placement analytical method. PID The controller parameters are tuned; In S2, the closed-loop transfer function of the roll channel Represented as: ， in, K P , K I , K D and K For cascade PID The controller has four parameters. A i ( i =0,1,2,3,4) are the simplification coefficients of the denominator during the solution process; ， in, ω The undamped natural frequency of the high-frequency conjugate poles. The damping ratio of the non-dominant second-order oscillatory element. The time constant of the low-frequency dominant pole. The pole time constant that cancels out the zeros in the molecule; The closed-loop transfer functions of the pitch and yaw channels correspond to the closed-loop transfer function of the roll channel.

2. The controller parameter tuning method for the forward flight mode of a quadcopter UAV according to claim 1, characterized in that, C 1 The value is 4. C 2 The value is 0.0187.

3. The controller parameter tuning method for the forward flight mode of a quadcopter UAV according to claim 1, characterized in that, The cascade PID The controller parameters are calculated based on the time constant of the dominant pole, the damping ratio of the non-dominant second-order oscillatory element, and the undamped natural frequency.

4. The controller parameter tuning method for the forward flight mode of a quadcopter UAV according to claim 1, characterized in that, Also includes: S 3: Adjust the cascaded configuration PID The controller performs time-domain characteristic analysis and parameter optimization iteration.

Images