Control method of unmanned aerial vehicle operation holder device based on fuzzy active disturbance rejection

Abstract

The invention discloses a control method of an unmanned aerial vehicle operation holder device based on fuzzy active disturbance rejection. The control method comprises the following steps: acquiring attitude Euler angle data of an operation load; a fuzzy active disturbance rejection control system is designed, the control system comprises a tracking differentiator, a linear extension observer, a fuzzy control module and a linear error compensation controller, attitude Euler angle data are input, and a speed target value of a three-axis motor is calculated and output; designing an FOC motor control algorithm comprising a sliding mode PI ring and a current control ring; the sliding mode PI ring receives a speed target value from the fuzzy active disturbance rejection control system and outputs a current control signal to a current control ring of an inner ring; the current control loop combines the current of the motor and the rotor position of the motor to obtain q-axis current so as to control the electromagnetic output torque of the motor; and the motor controller performs angle rotation action according to the speed target value. The stability of the holder can be improved, and the buffeting performance of the holder can be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle control, and in particular to a control method for an unmanned aerial vehicle operation gimbal device based on fuzzy self-interference rejection. Background Art

[0002] With the rapid development of drone technology and its wide application in various fields, from aerial photography to agricultural plant protection, from power line inspection to express delivery, drones are gradually changing the operation mode of traditional industries. As an important component of drones, the gimbal is responsible for carrying and stabilizing mission loads such as cameras and sensors. Its control method is crucial to improving the efficiency and quality of drone operations.

[0003] The gimbal device consists of a gimbal and a load. During actual operation, the gimbal needs to face multiple challenges from the external environment, such as wind and vibration, which directly affect the stability and precise positioning of the load. In addition, different application scenarios have different requirements for gimbal control. For example, aerial photography pursues smooth lens movement, while agricultural spraying pays more attention to the accuracy of the spraying direction. These differentiated requirements put forward higher standards and more complex requirements for the motor control technology of the gimbal. Faced with technical challenges and market demands, more advanced and intelligent gimbal control methods have gradually become the focus of industry development. Summary of the invention

[0004] 1. Technical problems to be solved:

[0005] In view of the above technical problems, the present invention provides a control method for a UAV operation gimbal device based on fuzzy self-disturbance rejection, and designs a fuzzy self-disturbance rejection algorithm for the gimbal control part of the device to optimize the control effect of the gimbal; at the same time, an improved sliding mode PI controller is designed in the gimbal motor control part to replace the speed controller in the FOC algorithm, so as to improve the stability of the gimbal and reduce the jitter of the gimbal.

[0006] 2. Technical solution:

[0007] A control method for a UAV operation gimbal device based on fuzzy self-disturbance rejection, characterized by comprising:

[0008] Step 1: Obtaining the position data of the workload collected by the attitude sensor of the UAV, wherein the position data is expressed in the form of attitude Euler angles;

[0009] Step 2: Design a fuzzy ADAS control system; the fuzzy ADAS control system includes a tracking differentiator, a linear extension observer, a fuzzy control module, and a linear error compensation controller; the attitude sensor collects the Euler angle data of the workload and inputs it into the tracking differentiator for signal tracking and differentiation processing, and outputs an angle tracking signal and its differential signal; the real-time speed of the three-axis motor of the gimbal is detected by a magnetic encoder, and the real-time speed of the three-axis motor is input into the linear extension observer; the linear extension observer observes the real-time speed, outputs a speed observation signal and its differential signal and a total disturbance signal; the system error is calculated according to the angle tracking differential signal and the speed observation differential signal, and the system error is input into the linear error compensation controller; the linear error compensator calculates and outputs the speed target value of the three-axis motor according to the system error and the total disturbance signal;

[0010] Step 3: Design an improved sliding mode PI controller for the three-axis FOC motor, including an outer sliding mode PI loop and an inner current control loop; the sliding mode PI loop receives the speed target value from the fuzzy anti-disturbance control system and outputs a current control signal to the inner current control loop; the inner current control loop collects the current of the brushless DC motor, performs Clark transformation and Park transformation in combination with the motor rotor position information, obtains the q-axis current and d-axis current in the synchronous rotating coordinate system, and controls the electromagnetic output torque of the motor by closed-loop controlling the value of the q-axis current;

[0011] Step 4: The three-axis FOC motor controller obtains the required rotation angle of the motor based on the received speed target value of the three-axis motor and the rotor information obtained from the magnetic encoder, and controls the motor to move according to the obtained rotation angle.

[0012] Furthermore, the attitude Euler angle data of the workload position collected by the attitude sensor in step 1 is transmitted to the gimbal controller via the SPI bus; the main control core model of the gimbal controller is STM32F103RC.

[0013] Furthermore, step 2 specifically includes:

[0014] S21: The expression of the tracking differentiator is as follows:

[0015]

[0016] In the above formula, v 1( k+1) represents the tracking signal of the current round of Euler angles, v1(k) represents the tracking signal of the Euler angles obtained in the previous round of calculations; h represents the integration step; v2(k+1) represents the differential signal of the current round of Euler angles, v2(k) represents the differential signal of the Euler angles obtained in the previous round of calculations; h0 is the filtering parameter, f his the comprehensive function of the fastest control, v(k) represents the current input Euler angle signal, and r is the parameter for adjusting the tracking speed;

[0017] S22: The expression of the linear expansion observer is:

[0018]

[0019] In the above formula, It represents the speed observation signal output by the previous round of linear expansion observer; represents the velocity differential signal output by the previous linear expansion observer, is the total output disturbance of the previous round of linear expansion observer output; Indicates the current speed observation signal; Indicates the current speed differential signal; is the total disturbance signal currently output; β1, β2, β3 are the preset speed observation gain coefficient, differential observation gain coefficient and total disturbance observation gain coefficient respectively. By adjusting the three gain coefficients, the tracking effect of the state variable can be controlled; u is the speed target value of the three-axis motor obtained by the previous round of calculation, and b0 is the system coefficient;

[0020] S23: The expression of the fuzzy control module is:

[0021]

[0022] In the above formula, b1 and b2 are the initial gain values ​​of the linear error compensator in the current round; Δb1 and Δb2 are the gain compensation values ​​of the fuzzy control output. and is the gain value after fuzzy control compensation;

[0023] S24: The expression of the linear error compensator is:

[0024]

[0025] In the above formula, e1 represents the actual state error, v1(k+1) represents the tracking signal of the current wheel Euler angle, represents the current speed observation signal; e2 represents the differential error; v2(k+1) represents the differential signal of the current wheel Euler angle; represents the current speed differential signal; u0 represents the previous stage speed value; e1 represents the actual state error; e2 represents the differential error; b1 and b2 are the gain coefficients of the current wheel of the linear error compensator; u is the speed target value, u0 represents the previous stage speed value, represents the total output disturbance, and b0 represents the system coefficient.

[0026] Furthermore, in step 3, the control of the sliding mode PI loop adopts a reaching law, which is expressed as follows:

[0027]

[0028] Where |x1| is the absolute value of the error term, s is the sliding surface, is the derivative of s, k1 and k2 are parameters;

[0029] The designed sliding surface is as follows:

[0030] S=Cx1+x2 (6)

[0031] In the formula, x1 is the angle error, x2 is the speed error, and C is the angle parameter;

[0032] Combining the above formula with the motor equation, we can get:

[0033]

[0034] Where D is the output parameter; tanh() represents the hyperbolic tangent; τ represents time;

[0035] Substitute the speed target value output by the three-axis FOC motor controller into the above solution to obtain the current round of calculation The value is used as the q-axis current i of the current loop input q The value is input into the current loop.

[0036] Furthermore, in step three, the inner current control loop specifically includes: a current PID module, an inverse park operation module, an SVPWM module, a motor, and a Clark Park operation module connected in sequence; the speed target value passes through the sliding mode PI module and is output as the q-axis current expected value, and then passes through the current PID module, the inverse park operation module, the SVPWM module, and the inverse Clark transformation module of the inner current control module in sequence and then returns to the current PID module.

[0037] Furthermore, the inner current control loop adopts i d =0 control mode, the current loop i q with i d After the current PID module, i d The expected value is 0, and the q-axis current i q The expected value of is calculated by the sliding mode PI control loop; the current PID module calculates the voltage target value V of the q-axis and d-axis q 、V d , combined with the current rotor angle of the rotor V q 、V d After performing the inverse park operation, we get V in the stationary coordinate system. α and Vβ ; V α and V β After another reverse Clark transformation, it becomes the desired voltage V of the three-phase motor a 、V b 、V c ; Then it is updated to the PWM output of each phase through the SVM mechanism, and finally acts on the motor through the three-phase inverter bridge.

[0038] 3. Beneficial effects:

[0039] (1) This method proposes a control method for a UAV operating gimbal device based on fuzzy anti-disturbance control. A fuzzy anti-disturbance control system is designed for the stability control part of the gimbal device. The fuzzy anti-disturbance control algorithm composed of a tracking differentiator, a linear expansion observer, a fuzzy control module and a linear error compensation controller can change the controller parameter value in real time according to the position information of the gimbal, realize the full parameter adjustment of the motor, ensure that the error between the system speed and the given speed is very small under any circumstances, and effectively avoid overshoot of the control system. At the same time, a fuzzy algorithm is introduced on the basis of the linear anti-disturbance control algorithm, so that the gain function in the anti-disturbance control algorithm can be adaptively adjusted according to external disturbances, thereby improving the rapidity of system response, enhancing the robustness of the system, and improving the dynamic performance and steady-state performance of the system.

[0040] (2) This method proposes a control method for a UAV operating gimbal device based on fuzzy self-disturbance rejection. An improved FOC control algorithm is designed for the motor control part of the gimbal, and an improved sliding mode PI controller is used to replace the speed controller in the traditional FOC algorithm. Sliding mode control is a nonlinear control strategy that can effectively deal with system uncertainty and external interference. It forces the system state to remain stable in unpredictable environmental changes through the design of the switching surface, making the gimbal more robust, faster in response mode, smaller in static error, easy to achieve parameter adaptive adjustment, and improve the overall stability of the gimbal system. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is the system block diagram of the fuzzy active disturbance rejection control system in this method;

[0042] Figure 2 Schematic diagram of an improved sliding mode PI controller of a three-axis FOC motor in the present method;

[0043] Figure 3 The overall flow chart of this method is shown in FIG. DETAILED DESCRIPTION

[0044] The present invention will be described in detail below with reference to the accompanying drawings.

[0045] As attached Figure 1 To Attachment Figure 3 As shown, a control method for a UAV operation gimbal device based on fuzzy anti-disturbance control is characterized by comprising:

[0046] Step 1: Obtaining the position data of the workload collected by the attitude sensor of the UAV, wherein the position data is expressed in the form of attitude Euler angles;

[0047] Step 2: Design a fuzzy ADAS control system; the fuzzy ADAS control system includes a tracking differentiator, a linear extension observer, a fuzzy control module, and a linear error compensation controller; the attitude sensor collects the Euler angle data of the workload and inputs it into the tracking differentiator for signal tracking and differentiation processing, and outputs an angle tracking signal and its differential signal; the real-time speed of the three-axis motor of the gimbal is detected by a magnetic encoder, and the real-time speed of the three-axis motor is input into the linear extension observer; the linear extension observer observes the real-time speed, outputs a speed observation signal and its differential signal and a total disturbance signal; the system error is calculated according to the angle tracking differential signal and the speed observation differential signal, and the system error is input into the linear error compensation controller; the linear error compensator calculates and outputs the speed target value of the three-axis motor according to the system error and the total disturbance signal;

[0048] Step 3: Design an improved sliding mode PI controller for the three-axis FOC motor, including an outer sliding mode PI loop and an inner current control loop; the sliding mode PI loop receives the speed target value from the fuzzy anti-disturbance control system and outputs a current control signal to the inner current control loop; the inner current control loop collects the current of the brushless DC motor, performs Clark transformation and Park transformation in combination with the motor rotor position information, obtains the q-axis current and d-axis current in the synchronous rotating coordinate system, and controls the electromagnetic output torque of the motor by closed-loop controlling the value of the q-axis current;

[0049] Step 4: The three-axis FOC motor controller obtains the required rotation angle of the motor based on the received speed target value of the three-axis motor and the rotor information obtained from the magnetic encoder, and controls the motor to move according to the obtained rotation angle.

[0050] Furthermore, the attitude Euler angle data of the workload position collected by the attitude sensor in step 1 is transmitted to the gimbal controller via the SPI bus; the main control core model of the gimbal controller is STM32F103RC.

[0051] Furthermore, step 2 specifically includes:

[0052] S21: The expression of the tracking differentiator is as follows:

[0053]

[0054] In the above formula, v 1( k+1) represents the tracking signal of the current round of Euler angles, v1(k) represents the tracking signal of the Euler angles obtained in the previous round of calculations; h represents the integration step; v2(k+1) represents the differential signal of the current round of Euler angles, v2(k) represents the differential signal of the Euler angles obtained in the previous round of calculations; h0 is the filtering parameter, f h is the comprehensive function of the fastest control, v(k) represents the current input Euler angle signal, and r is the parameter for adjusting the tracking speed;

[0055] Among them, the fastest control comprehensive function is expressed as follows:

[0056]

[0057] In the above formula, d, a0, a1, a2, μ, f s (μ,d),f s (a, d), where a is the intermediate quantity in the derivation process of the fastest control comprehensive function.

[0058] The tracking differentiator (TD) is designed to solve the impact of drastic changes in the input signal on the control system and provide the controller with accurate information on the signal change rate. Compared with the traditional differentiator, the TD has better noise immunity and can effectively avoid the error accumulation and signal oscillation problems introduced by noise.

[0059] S22: The expression of the linear expansion observer is:

[0060]

[0061] In the above formula, It represents the speed observation signal output by the previous round of linear expansion observer; represents the velocity differential signal output by the previous linear expansion observer, is the total output disturbance of the previous round of linear expansion observer output; Indicates the current speed observation signal; Indicates the current speed differential signal; is the total disturbance signal currently output; β1, β2, β3 are the preset speed observation gain coefficient, differential observation gain coefficient and total disturbance observation gain coefficient respectively. By adjusting the three gain coefficients, the tracking effect of the state variable can be controlled; u is the speed target value of the three-axis motor obtained by the previous round of calculation, and b0 is the system coefficient;

[0062] The Linear Extended State Observer (LESO) is a special form of the Extended State Observer (ESO) for state estimation and disturbance compensation of dynamic systems. The core purpose of LESO is to compensate for the nonlinearity and uncertainty of the system by observing the state of the system and unknown disturbances in real time, thereby improving the robustness and control performance of the system. LESO is mainly used to actively suppress the impact of external disturbances and uncertainties on the system, especially in some complex systems that are difficult to model. It achieves the purpose of precise control by online observation of unknown disturbances and system dynamic characteristics and adjusting the output of the controller in real time.

[0063] S23: The expression of the fuzzy control module is:

[0064]

[0065] In the above formula, b1 and b2 are the initial gain values ​​of the linear error compensator in the current round; Δb1 and Δb2 are the gain compensation values ​​of the fuzzy control output. and is the gain value after fuzzy control compensation;

[0066] Fuzzy control does not require a more accurate mathematical model, and can control complex controlled objects through fuzzy rules. The values ​​of gains b1 and b2 in the linear error compensation controller (LESF) can be adjusted through fuzzy control. The output gain value of fuzzy control can be determined by selecting membership functions and formulating fuzzy rules. However, since the gain after fuzzy processing is a fuzzy quantity, a defuzzification calculation is required before the fuzzy gain is input into the linear error compensation controller (LESF) to convert the fuzzy gain into an accurate gain. Commonly used defuzzification algorithms include the maximum membership method, cutting method, area centroid method, and center average method.

[0067] In one embodiment, the area centroid method can be used to perform defuzzification calculation on the fuzzy quantity, and its mathematical expression is as follows:

[0068]

[0069] In the formula, y is the output variable, μ(y) is the membership function of the output variable y, and y exact It is the exact value after defuzzification calculation.

[0070] S24: The expression of the linear error compensator is:

[0071]

[0072] In the above formula, e1 represents the actual state error, v1(k+1) represents the tracking signal of the current wheel Euler angle, represents the current speed observation signal; e2 represents the differential error; v2(k+1) represents the differential signal of the current wheel Euler angle; represents the current speed differential signal; u0 represents the previous stage speed value; e1 represents the actual state error; e2 represents the differential error; b1 and b2 are the gain coefficients of the current wheel of the linear error compensator; u is the speed target value, u0 represents the previous stage speed value, represents the total output disturbance, and b0 represents the system coefficient.

[0073] The linear error compensation controller (LESF) is a controller based on the feedback control principle, which is mainly used for error compensation and adjustment in dynamic systems. The core idea is to measure the error of the system (i.e. the deviation between the actual output and the expected output of the system) and use the error value to feed back to the controller for adjustment, thereby achieving the purpose of precise control.

[0074] Furthermore, in step 3, the control of the sliding mode PI loop adopts a reaching law, which is expressed as follows:

[0075]

[0076] Where |x1| is the absolute value of the error term, s is the sliding surface, is the derivative of s, k1 and k2 are parameters;

[0077] The designed sliding surface is as follows:

[0078] S=Cx1+x2 (6)

[0079] In the formula, x1 is the angle error, x2 is the speed error, and C is the angle parameter;

[0080] Combining the above formula with the motor equation, we can get:

[0081]

[0082] Where D is the output parameter; tanh() represents the hyperbolic tangent; τ represents time;

[0083] Substituting the speed target value output by the three-axis FOC motor controller into the above solution, we can get the current round of calculation The value is used as the q-axis current i of the current loop input q The value is input into the current loop.

[0084] In this method, if Figure 2 As shown, the brushless DC motor of the actuator adopts the FOC motor drive algorithm; the overall cascade PID control structure is adopted, the outer loop is a sliding mode PI loop, and the inner loop is a current control loop.

[0085] The mathematical expression of the PID algorithm after the analog PID is discretized is as follows:

[0086]

[0087] Where: K P is the proportionality coefficient; K I is the integral coefficient; K D is the differential coefficient;

[0088] e(n) = current expected value - current actual value; e(i) is the expected value at time i - the actual value at time i.

[0089] The sliding mode PI loop of the motor drive takes the motor speed value sent by the main control core after the above-mentioned linear anti-disturbance control algorithm as the target, combines the differential of the angle data of the encoder feedforward as the current speed, and outputs the current control signal to the current control loop through sliding mode PI control. The current control loop collects the current of the brushless DC motor, combines the motor rotor position information obtained from the magnetic encoder, performs a Clark transformation and a Park transformation, and can obtain the q-axis current i in the synchronous rotating coordinate system. q and the d-axis current i d The magnitude of the electromagnetic torque is related to i q It is proportional to the value of iq, so the electromagnetic output torque of the motor can be controlled by closed-loop control of the value of iq.

[0090] Furthermore, in step three, the inner current control loop specifically includes: a current PID module, an inverse park operation module, an SVPWM module, a motor, and a Clark Park operation module connected in sequence; the speed target value passes through the sliding mode PI module and is output as the q-axis current expected value, and then passes through the current PID module, the inverse park operation module, the SVPWM module, and the inverse Clark transformation module of the inner current control module in sequence and then returns to the current PID module.

[0091] Furthermore, the inner current control loop adopts i d =0 control mode, the current loop i q with i d After the current PID module, i d The expected value is 0, and the q-axis current i q The expected value of is calculated by the sliding mode PI control loop; the current PID module calculates the voltage target value V of the q-axis and d-axis q 、V d , combined with the current rotor angle of the rotor V q 、V d After performing the inverse park operation, we get V in the stationary coordinate system. α and V β ; Vα and V β After another reverse Clark transformation, it becomes the desired voltage V of the three-phase motor a 、V b 、V c ; Then it is updated to the PWM output of each phase through the SVM mechanism, and finally acts on the motor through the three-phase inverter bridge.

[0092] Although the present invention has been disclosed as above in terms of preferred embodiments, they are not intended to limit the present invention. Anyone skilled in the art can make various changes or modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection defined by the claims of this application.

Claims

1. A control method for a UAV operation gimbal device based on fuzzy anti-disturbance control, characterized in that: include: Step 1: Obtaining the position data of the workload collected by the attitude sensor of the UAV, wherein the position data is expressed in the form of attitude Euler angles; Step 2: Design a fuzzy anti-disturbance control system; the fuzzy anti-disturbance control system includes a tracking differentiator, a linear expansion observer, a fuzzy control module, and a linear error compensation controller; the attitude sensor collects the Euler angle data of the workload and inputs it into the tracking differentiator for signal tracking and differential processing, and outputs an angle tracking signal and its differential signal; the real-time speed of the three-axis motor of the gimbal is detected by a magnetic encoder, and the real-time speed of the three-axis motor is input into the linear expansion observer; The linear expansion observer observes the real-time speed and outputs the speed observation signal, its differential signal and the total disturbance signal; The system error is calculated according to the angle tracking differential signal and the speed observation differential signal, and the system error is input into the linear error compensation controller; the linear error compensator calculates and outputs the speed target value of the three-axis motor according to the system error and the total disturbance signal; Step 3: Design an improved sliding mode PI controller for the three-axis FOC motor, including an outer sliding mode PI loop and an inner current control loop; the sliding mode PI loop receives the speed target value from the fuzzy anti-disturbance control system and outputs a current control signal to the inner current control loop; the inner current control loop collects the current of the brushless DC motor, performs Clark transformation and Park transformation in combination with the motor rotor position information, obtains the q-axis current and d-axis current in the synchronous rotating coordinate system, and controls the electromagnetic output torque of the motor by closed-loop controlling the value of the q-axis current; Step 4: The three-axis FOC motor controller obtains the required rotation angle of the motor based on the received speed target value of the three-axis motor and the rotor information obtained from the magnetic encoder, and controls the motor to move according to the obtained rotation angle.

2. The control method of the unmanned aerial vehicle operation gimbal device based on fuzzy anti-disturbance control according to claim 1 is characterized in that: In step 1, the attitude Euler angle data of the workload position collected by the attitude sensor is transmitted to the gimbal controller via the SPI bus; the main control core model of the gimbal controller is STM32F103RC.

3. The control method of the unmanned aerial vehicle operation gimbal device based on fuzzy anti-disturbance control according to claim 1 is characterized in that: Step 2 specifically includes: S21: The expression of the tracking differentiator is as follows: In the above formula, v 1( k+1) represents the tracking signal of the current round of Euler angles, v1(k) represents the tracking signal of the Euler angles obtained in the previous round of calculations; h represents the integration step; v2(k+1) represents the differential signal of the current round of Euler angles, v2(k) represents the differential signal of the Euler angles obtained in the previous round of calculations; h0 is the filtering parameter, f h is the comprehensive function of the fastest control, v(k) represents the current input Euler angle signal, and r is the parameter for adjusting the tracking speed; S22: The expression of the linear expansion observer is: In the above formula, It represents the speed observation signal output by the previous round of linear expansion observer; represents the velocity differential signal output by the previous linear expansion observer, is the total output disturbance of the previous round of linear expansion observer output; Indicates the current speed observation signal; Indicates the current speed differential signal; is the total disturbance signal currently output; β1, β2, β3 are the preset speed observation gain coefficient, differential observation gain coefficient and total disturbance observation gain coefficient respectively. By adjusting the three gain coefficients, the tracking effect of the state variable can be controlled; u is the speed target value of the three-axis motor obtained by the previous round of calculation, and b0 is the system coefficient; S23: The expression of the fuzzy control module is: In the above formula, b1 and b2 are the initial gain values ​​of the linear error compensator in the current round; Δb1 and Δb2 are the gain compensation values ​​of the fuzzy control output. and is the gain value after fuzzy control compensation; S24: The expression of the linear error compensator is: In the above formula, e1 represents the actual state error, v1(k+1) represents the tracking signal of the current wheel Euler angle, represents the current speed observation signal; e2 represents the differential error; v2(k+1) represents the differential signal of the current wheel Euler angle; represents the current speed differential signal; u0 represents the previous stage speed value; e1 represents the actual state error; e2 represents the differential error; b1 and b2 are the gain coefficients of the current wheel of the linear error compensator; u is the speed target value, u0 represents the previous stage speed value, represents the total output disturbance, and b0 represents the system coefficient.

4. The control method of the unmanned aerial vehicle operation gimbal device based on fuzzy anti-disturbance control according to claim 3 is characterized in that: In step 3, the control of the sliding mode PI loop adopts the reaching law, which is expressed as follows: In the above formula, |x1| is the absolute value of the error term, s is the sliding surface, is the derivative of s, k1 and k2 are parameters; The designed sliding surface is as follows: S=Cx1+x2 (6) In the above formula, x1 is the angle error, x2 is the speed error, and C is the angle parameter; Combining the motor equation, we can get: In the above formula, D is the output parameter; tanh() represents the hyperbolic tangent; τ represents time; Substitute the speed target value output by the three-axis FOC motor controller into the above solution to obtain the current round of calculation The value is used as the q-axis current i of the current loop input q The value is input into the current loop.

5. The control method of the unmanned aerial vehicle operation gimbal device based on fuzzy anti-disturbance control according to claim 4 is characterized in that: In step 3, the inner current control loop specifically includes: a current PID module, an inverse park operation module, an SVPWM module, a motor, and a Clark Park operation module connected in sequence; the speed target value passes through the sliding mode PI module and is output as the q-axis current expected value, and then passes through the current PID module, the inverse park operation module, the SVPWM module, and the inverse Clark transformation module of the inner current control module in sequence and then returns to the current PID module.

6. The control method of the unmanned aerial vehicle operation gimbal device based on fuzzy anti-disturbance control according to claim 5, characterized in that: The inner current control loop adopts i d =0 control mode, the current loop i q with i d After the current PID module, i d The expected value is 0, and the q-axis current i q The expected value of is calculated by the sliding mode PI control loop; the current PID module calculates the voltage target value V of the q-axis and d-axis q 、V d , combined with the current rotor angle of the rotor V q 、V d After performing the inverse park operation, we get V in the stationary coordinate system. α and V β ; V α and V β After another reverse Clark transformation, it becomes the desired voltage V of the three-phase motor a 、V b 、V c ; Then it is updated to the PWM output of each phase through the SVM mechanism, and finally acts on the motor through the three-phase inverter bridge.

Images