CN106647781B - Control method based on Repetitive control compensation fuzzy neuron PID quadrotor - Google Patents

Control method based on Repetitive control compensation fuzzy neuron PID quadrotor Download PDF

Info

Publication number
CN106647781B
CN106647781B CN201610947397.4A CN201610947397A CN106647781B CN 106647781 B CN106647781 B CN 106647781B CN 201610947397 A CN201610947397 A CN 201610947397A CN 106647781 B CN106647781 B CN 106647781B
Authority
CN
China
Prior art keywords
control
fuzzy
pid
quadrotor
compensation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201610947397.4A
Other languages
Chinese (zh)
Other versions
CN106647781A (en
Inventor
赵帅
罗晓曙
钟海鑫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guangxi Normal University
Original Assignee
Guangxi Normal University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guangxi Normal University filed Critical Guangxi Normal University
Priority to CN201610947397.4A priority Critical patent/CN106647781B/en
Publication of CN106647781A publication Critical patent/CN106647781A/en
Application granted granted Critical
Publication of CN106647781B publication Critical patent/CN106647781B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G05CONTROLLING; REGULATING
    • G05DSYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
    • G05D1/08Control of attitude, i.e. control of roll, pitch, or yaw
    • G05D1/0808Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft
    • G05D1/0816Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft to ensure stability
    • G05D1/0825Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft to ensure stability using mathematical models

Abstract

For a kind of control method based on Repetitive control compensation fuzzy neuron PID quadrotor, comprising the following steps: S10: establish the kinetic model of quadrotor drone;S20: based on compensation fuzzy neuron PID control is repeated, S21: design generates fuzzy reasoning (rule) by neural network and is capable of the network structure of self adjusting PID parameter;S22: compensation control is repeated.Repetitive controller based on internal model principle is embedded by method provided by the invention to be generated in the closed-loop control of fuzzy reasoning self adjusting PID based on neural network, it is formed based on repetition compensation fuzzy neuron PID control, so that system is still within closed loop states, fuzzy neuron PID controls to adjust output error in real time, system at steady state when, compensating controller is repeated to be adjusted, to enable output signal under lower state to track input signal well, when disturbing, fuzzy neuron PID adjusts input signal, reduce signal errors, promote the tracking accuracy of aerocraft system.

Description

Control method based on Repetitive control compensation fuzzy neuron PID quadrotor
Technical field
The present invention relates to unmanned vehicle technical fields, and in particular to the flight controlling party of quadrotor unmanned vehicle Method.
Background technique
Quadrotor has 6 freedom degrees, drives propeller to generate lift, thrust by four individual motors, thus So that quadrotor is realized hovering and change of flight posture, is that a kind of multiple-input and multiple-output, close coupling, drive lacking are non-thread Property system.And flight system requires non-overshoot (or overshoot is smaller) and can quickly track input instruction, when stable state without Static error has strong anti-interference ability and when to system parameter variations, to there is stronger robustness.PID control (Proportional-integral-derivative Control) because it is simple, stability is good, preferable robustness and skill Art is still the control algolithm of current most of aircraft first choices with respect to other control algolithm comparative maturities.But due to four rotations By externalities etc. in the uncertainty and flight course of rotor aircraft system itself, make parameter in PID control without Method self-adjusting, to influence the flight attitude of aircraft, and in PID control aerocraft system tracking accuracy it is lower.
Summary of the invention
Present invention seek to address that the technical problems existing in the prior art.
A kind of effective control method is provided, enables aircraft when by external disturbance in real time to flight The flight attitude of device carries out self-adjusting, improves and improve the tracking accuracy of flight system.
Present invention aims at the PID ginsengs of normal PID lgorithm in the flight attitude control method for solving quadrotor drone Several self-adjusting process errors is big and the problem of tracking accuracy deficiency, propose based on Repetitive control compensation fuzzy neuron from The control method of PID is adjusted, to improve the flying quality of quadrotor.
To achieve the above object, proposed by the present invention a kind of based on the flight of Repetitive control compensation fuzzy neuron PID quadrotor The control method of device, includes the following steps:
S10: the kinetic model of quadrotor drone is established
According to the flight attitude of quadrotor, aircraft is established by Newton-Euller method and coordinate conversion matrix Dynamic (dynamical) mathematical model, wherein kinetics equation such as formula (1):
In formula, m is the quality of quadrotor, and g is acceleration of gravity, μx、μy、μzFor three X-axis, Y-axis, Z axis directions Coefficient of air resistance, Jx、Jz、JzFor quadrotor around X-axis, Y-axis, Z axis rotary inertia, IrFor quadrotor Rotary inertia of the rotor relative to rotary shaft, l is distance w of the rotor centers point to seat quadrotor mass centre1、w2、 w3For the angular speed of aircraft, x, y, z is the position of aircraft, Ωi(i=1,2,3,4) is the revolving speed of each rotor, θ, φ, ψ For 3 attitude angles (pitching, rolling, yaw) of aircraft.
By the kinetics equation of the quadrotor of formula (1) be converted into four independent control channel Ui (i=1,2,3, 4), by controlling this four independent control channels, this four channels respectively by height Repetitive control compensation fuzzy neuron PID, Roll Repetitive control compensation fuzzy neuron PID, pitching Repetitive control compensation fuzzy neuron PID, yaw Repetitive control compensation nerve Fuzzy composition;
S20: based on repetition compensation fuzzy neuron PID control
S21: design generates fuzzy reasoning (rule) by neural network and is capable of the network structure of self adjusting PID parameter
The Neural Fuzzy system for designing a dual input, singly exporting, and set its using one way propagation multilayer before To neural network, its input data successively successively passes through each hidden layer node, finally from the output section of output layer from input layer Point obtains output data, wherein defined below between each layer:
First layer is neuron node, indicates the input signal of fuzzy controller, is completed to error e and error rate The receiving of ec;
The second layer indicates the Linguistic Value of input signal linguistic variable, is the blurring to input data, i.e., by input data It is converted into fuzzy quantity, is expressed as a membership function;
The process of the fuzzy reasoning of third layer and the 4th layer of completion fuzzy system, this two layers expression fuzzy control rule, In, third layer completes the fuzzy former piece of fuzzy rule, and the consequent of the 4th layer of completion fuzzy rule carries out fuzzy reasoning and exports mould Paste amount;
Layer 5 completes de-fuzzy, by fuzzy quantity sharpening, and exports control amount;
Fuzzy neuron self adjusting PID leads to according to the size of input signal deviation e and ec, direction and variation tendency feature It crosses neuro-fuzzy inference and makes corresponding decision, on-line tuning pid parameter kp,ki,kdIt is wanted with meeting different moments to the difference of parameter It asks, wherein PID controller is parameter increase formula controller, and fuzzy neuron is added on the basis of initializing PID controller parameter Controller determines optimal k required for PID control to pid parameter on-line tuning, as algorithm abovep,ki,kdParameter, thus real Parameter self-tuning is showed;
S22: compensation control is repeated
Repetitive controller based on internal model principle is embedded into, the closed loop control of fuzzy reasoning self adjusting PID is generated based on neural network In system, formed based on repetition compensation fuzzy neuron PID control.
This method is converted into four independent control channels by the kinetics equation of quadrotor, and passes through design It is realized as the network structure that neural network generates fuzzy reasoning (rule) and is capable of self adjusting PID parameter to needed for PID control Best kp,ki,kdParameter self-tuning;The tracking error of output signal can be eliminated by repeating compensation control, then system is made to exist Be not in distortion under the control output of load, promote the gesture stability stability and control precision of aircraft.
Specifically, in the Repetitive controller based on internal model principle, in addition to the mistake at the current time being added in controlled device It has also been superimposed the error signal of last moment outside difference signal, has formed the positive feedback with Time Delay, in Time Delay series connection one Low-pass filter, while to reduce the gain that Repetitive controller acts on high band.
Further, the control signal that the Repetitive controller obtains need to just export after delay time t, and setting repeats Compensating controller exports the output of low-pass filter in the delay time t again after repeating PID, and before exporting again It compensates.
Control method based on Repetitive control compensation fuzzy neuron PID quadrotor of the invention, it is former based on internal model The Repetitive controller of reason is by the compensation to control error, to reduce error, to reduce the steady-state error of control system, inhibits negative The disturbance of load has also been superimposed upper a period of time other than the error signal at the current time being added in controlled device in repeated controlling system The error signal at quarter.Since Repetitive controller is made of the positive feedback with Time Delay, the open-loop transfer function of system exists Contain numerous pole in the imaginary axis, therefore system goes to zero to the systematic error of any input signal and interference signal, robustness compared with By force, and in Time Delay one low-pass filter of series connection, while to reduce the gain that Repetitive controller acts on high band, in turn It ensure that system stability, even if internal system is interfered, releasing required for influence of the interference to output signal by prolonging In slow that time t, the Repetitive controller based on internal model principle is embedded into, fuzzy reasoning self-adjusting is generated based on neural network In PID closed-loop control, is formed and interfered out based on repetition compensation fuzzy neuron PID control so that system is still within closed loop states In delay time t after now, fuzzy neuron PID controls to adjust output error in real time, system at steady state when, The main function for repeating compensating controller is to be adjusted, so that output signal under lower state be enable to track input well Signal, when disturbing, fuzzy neuron PID can adjust input signal, so that signal errors be made to reduce, promote flight The tracking accuracy of device system.
Detailed description of the invention
Above-mentioned and/or additional aspect of the invention and advantage will become from the description of the embodiment in conjunction with the following figures Obviously and it is readily appreciated that, in which:
Fig. 1 is quadrotor unmanned vehicle overall structure diagram of the invention;
Fig. 2 is that quadrotor unmanned vehicle main modular of the invention constitutes schematic diagram;
Fig. 3 is the Neural Fuzzy system construction drawing in quadrotor control method of the invention;
Fig. 4 is the fuzzy neuron PID control system structure principle chart in quadrotor control method of the invention;
Fig. 5 is the repeated controlling system schematic diagram in quadrotor control method of the invention;
Fig. 6 is in quadrotor control method of the invention based on the fuzzy neuron PID control system for repeating compensation System block diagram;
Fig. 7 is the quadrotor roll angle system emulation block diagram in quadrotor control method of the invention;
Fig. 8 is noiseless Traditional PID, fuzzy neuron PID, repeats to compensate the rank of the roll angle under fuzzy neuron PID control Jump response comparison diagram;
Fig. 9 is to have Traditional PID under lasting interference, fuzzy neuron PID, repeat the rolling under compensation fuzzy neuron PID control The attitude angle comparison diagram of aircraft under the step response Self-tuning System at angle;
Figure 10 is Traditional PID, fuzzy neuron PID, repeats compensation fuzzy neuron PID tracing property test comparison chart;
Figure 11 is that the repetition of quadrotor control method of the invention compensates fuzzy neuron PID robustness test chart;
Figure 12 is the PID robustness test chart of quadrotor posture PID control method;
Figure 13 is the fuzzy neuron PID robustness test chart of quadrotor posture fuzzy neuron PID control method.
Specific embodiment
To better understand the objects, features and advantages of the present invention, with reference to the accompanying drawing and specific real Applying mode, the present invention is further described in detail.It should be noted that in the absence of conflict, the implementation of the application Feature in example and embodiment can be combined with each other.
In the following description, numerous specific details are set forth in order to facilitate a full understanding of the present invention, still, the present invention may be used also Implement in a manner of using other than the one described here, therefore, protection scope of the present invention is not by following public tool The limitation of body embodiment.
It is further described referring to quadrotor unmanned vehicle of the Fig. 1-2 to the embodiment of the present invention.
As depicted in figs. 1 and 2, quadrotor unmanned vehicle 100 includes body 10 and the aircraft being fixed on body 10 Control module 20, further includes the motor drive module 60 being fixed on 10 4 cantilevers of body and the rotor 70 being driven by motor, In addition, as shown in Fig. 2, quadrotor unmanned vehicle further include be mounted on the body 10 and respectively with the flying vehicles control The navigation of the connection of module 20, Inertial Measurement Unit 40, power module further include logical with the communication connection of flying vehicles control module 20 Believe module 30 and the power module 50 for flying vehicles control module for power supply;Navigation, Inertial Measurement Unit 40 are using high-precision GPS navigation carries out tracking and positioning to quadrotor unmanned vehicle 100, and provides location information simultaneously to the main controller module 20 Navigation, and modify during the navigation process and solidify baud rate, in addition it can save the setting up procedure of baud rate, inertia measurement portion Divide includes the three axis accelerometer connecting respectively with the main control module 20, used gyroscope and magnetometer, the inertia measurement Unit is for three axis accelerometers of sense aircraft, rolling angular speed, pitch rate, yawrate information and ground magnetic strength Information is spent, these information collectively form the flight attitude data of quadrotor;The communication module 30 mainly in real time with Earth station carries out control signal and status signal data exchange comprising remote controler, PPM decoder and PPM receiver, it is described PPM encoder is connect with the remote controler, and four channel control signals of remote controler wirelessly pass after being encoded by PPM encoder The PPM receiver is passed, the PPM receiver is connect with the control module;The motor drive module 60 includes being used for It drives four motors of four rotors 70 on aircraft and controls four motors (motor 1, motor 2,3 and of motor respectively Motor 4) work electricity adjust (electron speed regulator), it is described electricity adjust is connect with the flight control modules 20 with receive motor control believe Number, the control amount that the motor drive module 60 is adjusted according to flying vehicles control module 20 to electricity, and then control turning for 4 motors Speed, and the revolving speed of 4 motors of real-time measurement, the lift for generating 4 rotors by the change of motor speed and torque generate phase The variation answered.
Navigation, Inertial Measurement Unit 40 can provide the current location information of quadrotor drone and three axis accelerate The posture information of the compositions such as degree, rolling angular speed, pitch rate and yawrate, flying vehicles control module 20 is quadrotor The core of 100 control system of unmanned plane, effect are responsible for the posture information and real-time resolving of acquisition aircraft, further according to The flight information issued by remote controler is detected, in conjunction with based on Repetitive control compensation fuzzy neuron PID quadrotor The control program of control method calculates actual output motor and controls signal to electricity tune, and then electricity is adjusted believes according to the control of acquisition The revolving speed of number 4 motors of control, to realize the control of the lift and torques that generate to 4 rotors, motor can be controlled by PWM Its revolving speed is made to reach and control the size of power caused by each rotor and torque.
The control module of the quadrotor unmanned vehicle is based on Repetitive control compensation fuzzy neuron PID quadrotor Control method COMPREHENSIVE CALCULATING real-time attitude information and control signal message after output motor control signal to control unmanned flight The method of device the following steps are included:
S10: the kinetic model of quadrotor drone is established
According to the flight attitude of quadrotor, aircraft is established by Newton-Euller method and coordinate conversion matrix Dynamic (dynamical) mathematical model, wherein kinetics equation such as formula (1):
In formula, m is the quality of quadrotor, and g is acceleration of gravity, μx、μy、μzFor three X-axis, Y-axis, Z axis directions Coefficient of air resistance, Jx、Jz、JzFor quadrotor around X-axis, Y-axis, Z axis rotary inertia, IrFor quadrotor Rotary inertia of the rotor relative to rotary shaft, l is distance w of the rotor centers point to seat quadrotor mass centre1、w2、 w3For the angular speed of aircraft, x, y, z is the position of aircraft, Ωi(i=1,2,3,4) is the revolving speed of each rotor, θ, φ, ψ For 3 attitude angles (pitching, rolling, yaw) of aircraft.
By the kinetics equation of the quadrotor of formula (1) be converted into four independent control channel Ui (i=1,2,3, 4), by controlling this four independent control channels, this four channels respectively by height Repetitive control compensation fuzzy neuron PID, Roll Repetitive control compensation fuzzy neuron PID, pitching Repetitive control compensation fuzzy neuron PID, yaw Repetitive control compensation nerve Fuzzy composition;
S20: based on repetition compensation fuzzy neuron PID control
S21: design generates fuzzy reasoning (rule) by neural network and is capable of the network structure of self adjusting PID parameter
As shown in figure 3, the Neural Fuzzy system for designing a dual input, singly exporting, and set it and passed using unidirectional The multilayer feedforward neural network broadcast, its input data successively successively pass through each hidden layer node, finally from output from input layer The output node of layer obtains output data, wherein defined below between each layer:
First layer is neuron node, indicates the input signal of fuzzy controller, is completed to error e and error rate The receiving of ec;
The second layer indicates the Linguistic Value of input signal linguistic variable, is the blurring to input data, i.e., by input data It is converted into fuzzy quantity, is expressed as a membership function;
The process of the fuzzy reasoning of third layer and the 4th layer of completion fuzzy system, this two layers expression fuzzy control rule, In, third layer completes the fuzzy former piece of fuzzy rule, and the consequent of the 4th layer of completion fuzzy rule carries out fuzzy reasoning and exports mould Paste amount;
Layer 5 completes de-fuzzy, by fuzzy quantity sharpening, and exports control amount;
As shown in figure 4, fuzzy neuron self adjusting PID is according to the size of error e and error rate ec input signal, side To and variation tendency feature, corresponding decision, on-line tuning pid parameter k are made by neuro-fuzzy inferencep,ki,kdTo meet Different moments require the difference of parameter, and wherein PID controller is parameter increase formula controller, in initialization PID controller ginseng Pid parameter on-line tuning is determined required for PID control most as algorithm above plus neurofuzzy controller on the basis of number Good kp,ki,kdParameter, to realize parameter self-tuning;
S22: compensation control is repeated
Repetitive controller based on internal model principle is embedded into, the closed loop control of fuzzy reasoning self adjusting PID is generated based on neural network In system, formed based on repetition compensation fuzzy neuron PID control.Repetitive controller is that Inoue is based on internal model principle in proposition in 1981 Come, if as shown in figure 5, internal model principle is exactly that input signal is included in a stable closed-loop system in simple terms, it is controlled The output of object can error-free tracking input signal, Repetitive controller has preferable tracking signal capabilities and robustness, base To reduce the steady-state error of control system, press down in compensation of the Repetitive controller to control error of internal model principle to reduce error Make the disturbance of load.On being also superimposed other than the error signal at the current time being added in controlled device in repeated controlling system The error signal at one moment, both the past deviation in figure.Since Repetitive controller is made of the positive feedback with Time Delay, The open-loop transfer function of system contains numerous pole in the imaginary axis, therefore system is to the system of any input signal and interference signal Error goes to zero, and robustness is stronger, but stability is difficult to ensure;Additionally while Repetitive controller can make output signal tracking defeated Enter signal, can itself there are also problems, shown in Fig. 6, the control signal that Repetitive controller obtains does not export not instead of at once, delay It sometime just exports, it is assumed that internal system has interference, and releasing influence of the interference to output signal then at least will be by delay That time.In delay time after interference appearance, system will not be adjusted, and system is in open loop shape within this section of delay time State.Repetitive controller is embedded into and is generated in the closed-loop control of fuzzy reasoning self adjusting PID based on neural network, as shown in fig. 6, in mind It is generated through network and a repetition compensating controller is added in the closed-loop control of fuzzy reasoning self adjusting PID, formed and compensated based on repetition Fuzzy neuron PID control has low-pass filter Q (s) the * e of time delay process at Time Delay series connection one-L, while to subtract Small Repetitive controller acts on the gain of high band, and r is input signal, e error, de/dt of the remote controler to flying vehicles control module Error change amount, e1 are that error passes through the output of the low-pass filter with time delay process, e2 is e1 by repeating the defeated of PID Out, ue is the output of Repetitive controller, the output of up fuzzy neuron PID control, the control output that u is flying vehicles control module 20, y For the quadrotor unmanned vehicle current pose information that Inertial Measurement Unit 40 detects, compensator T(s) * e-LBy Repetitive controller Output ue carry out positive feedback.Compensating controller is repeated in Fig. 6 can eliminate the tracking error of output signal, then system be made to exist It is not in distortion under the output of load, fuzzy neuron PID then controls to adjust output error in real time;System is in stabilization When under state, the main function for repeating compensating controller is to be adjusted, to keep output signal under lower state fine Ground tracks input signal, and when disturbing, fuzzy neuron PID can adjust input signal, so that signal errors be made to subtract It is small.
In order to verify control program effect feasibility proposed by the present invention, the four-rotor aircraft control system built is utilized Carry out simulating, verifying.Effect of the invention is further illustrated below with reference to Fig. 9 to 13.
Control performance comparative experiments
Corresponding conventional PID controller and fuzzy neuron PID controller are devised, is mended with proposed by the present invention based on repetition It repays the quadrotor under fuzzy neuron self-regulated PID control and compares experiment.In an experiment, it completes first without dry In the case where disturbing, repeat to compensate under fuzzy neuron self-regulated PID control and regulatory PID control and fuzzy neuron PID control The step response of quadrotor roll angle is tested, and roll angle system emulation procedural block diagram is as shown in Figure 7.Corresponding emulation is real It is as shown in Figure 8 to test result.Then it carries out in the case where there is lasting interference, repeats to compensate fuzzy neuron self-regulated PID control and Traditional PID Quadrotor Immunity Performance comparison under control and fuzzy neuron PID control, as shown in Figure 9.It repeats to compensate neural mould The quadrotor robust performance comparison under self-regulated PID control and regulatory PID control and fuzzy neuron PID control is pasted, As shown in Figure 11,12,13.The tracing property of quadrotor under three of the above control algolithm is detected under the input of sinusoidal signal It can test, test results are shown in figure 10.
In noiseless lower contrast simulation as a result, as shown in Figure 8, it can be seen that repeat the super of compensation fuzzy neuron PID control Tune amount is significantly less than the overshoot of fuzzy neuron PID and the overshoot of regulatory PID control, and its regulating time is less than neural mould The regulating time of PID and the regulating time of regulatory PID control are pasted, by simulation comparison result it is found that repeating compensation control mind There is preferable response dynamics performance through fuzzy-adaptation PID control.
The contrast simulation in the case where there is interference is as a result, as shown in Figure 9, it can be seen that repeats to compensate when duration interference is added Fuzzy neuron PID control wants small relative to the disturbance of fuzzy neuron PID control, and PID control is bright by the variation that is influenced is disturbed It is aobvious, therefore compensation fuzzy neuron PID is repeated with preferable anti-interference.
It is inputted in sine, the simulation result of three kinds of controllers, as shown in Figure 10, repeats the control for compensating fuzzy neuron PID Control precision of the precision better than fuzzy neuron PID and the control precision for being substantially better than Traditional PID.Thus repeat compensation fuzzy neuron PID controller has preferable tracking performance in the case where changing system parameter.It is observed by emulation and repeats compensation fuzzy neuron PID control The robustness of device processed can learn that repetition compensates fuzzy neuron PID control by emulation experiment comparison as shown in Figure 11,12,13 The robustness of device is substantially better than the robustness of conventional PID controller.
By emulation experiment it is found that repeat compensate fuzzy neuron self-regulated PID control under flight system overshoot compared with It is small and can quickly track input instruction, without static error when stable state, has strong anti-interference ability and system is joined When number variation, there is preferable robustness.
These are only the preferred embodiment of the present invention, is not intended to restrict the invention, for those skilled in the art For member, the invention may be variously modified and varied.It is all within creativeness spirit of the invention and principle, it is made any Modification, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.

Claims (3)

1. a kind of control method based on Repetitive control compensation fuzzy neuron PID quadrotor, comprising the following steps:
S10: the kinetic model of quadrotor drone is established
According to the flight attitude of quadrotor, the dynamic of aircraft is established by Newton-Euller method and coordinate conversion matrix The mathematical model of mechanics, wherein kinetics equation such as formula (1):
In formula, m is the quality of quadrotor, and g is acceleration of gravity, μx、μy、μzFor X-axis, Y-axis, three directions of Z axis sky Vapour lock force coefficient, Jx、Jz、JzFor quadrotor around X-axis, Y-axis, Z axis rotary inertia, IrFor the rotation of quadrotor Rotary inertia of the wing relative to rotary shaft, l are distance w of the rotor centers point to seat quadrotor mass centre1、w2、w3For The angular speed of aircraft, x, y, z are the position of aircraft, ΩiFor the revolving speed of each rotor, wherein i=1,2,3,4, θ, φ, ψ Respectively pitch attitude angle, roll attitude angle, the yaw-position angle of aircraft,
Four independent control channel Ui are converted by the kinetics equation of the quadrotor of formula (1), wherein i=1,2,3, 4, by controlling this four independent control channels, this four channels by height Repetitive control compensation fuzzy neuron PID, are turned over respectively Repetitive control compensation fuzzy neuron PID, pitching Repetitive control compensation fuzzy neuron PID are rolled, Repetitive control compensation nerve mould is yawed Paste PID composition;
S20: based on repetition compensation fuzzy neuron PID control
S21: design generates fuzzy inference rule by neural network and is capable of the network structure of self adjusting PID parameter
The Neural Fuzzy system for designing a dual input, singly exporting, and set its using one way propagation multilayer before Godwards Through network, its input data is successively successively passed through each hidden layer node, is finally obtained from the output node of output layer from input layer To output data, wherein defined below between each layer:
First layer is neuron node, indicates the input signal of fuzzy controller, is completed to error e and error rate ec Receive;
The second layer indicates the Linguistic Value of input signal linguistic variable, is the blurring to input data, i.e., converts input data At fuzzy quantity, it is expressed as a membership function;
The process of the fuzzy reasoning of third layer and the 4th layer of completion fuzzy system, this two layers expression fuzzy control rule, wherein the The fuzzy former piece of three layers of completion fuzzy rule, the consequent of the 4th layer of completion fuzzy rule carry out fuzzy reasoning and export fuzzy quantity;
Layer 5 completes de-fuzzy, by fuzzy quantity sharpening, and exports control amount;
Fuzzy neuron self adjusting PID passes through mind according to the size of input signal deviation e and ec, direction and variation tendency feature Corresponding decision, on-line tuning pid parameter k are made through fuzzy reasoningp,ki,kdIt is required with meeting different moments to the difference of parameter, Wherein PID controller is parameter increase formula controller, and nerve fuzzy control is added on the basis of initializing PID controller parameter Device determines optimal k required for PID control to pid parameter on-line tuning, as algorithm abovep,ki,kdParameter, to realize Parameter self-tuning;
S22: compensation control is repeated
Repetitive controller based on internal model principle is embedded into, the closed-loop control of fuzzy reasoning self adjusting PID is generated based on neural network In, it is formed based on repetition compensation fuzzy neuron PID control.
2. the control method according to claim 1 based on Repetitive control compensation fuzzy neuron PID quadrotor, It is characterized in that, in the Repetitive controller based on internal model principle, in addition to the error signal at the current time being added in controlled device It has also been superimposed the error signal of last moment outside, has formed the positive feedback with Time Delay, in Time Delay one low pass filtered of series connection Wave device, while to reduce the gain that Repetitive controller acts on high band.
3. the control method according to claim 2 based on Repetitive control compensation fuzzy neuron PID quadrotor, It is characterized in that, the control signal that the Repetitive controller obtains need to just export after delay time t, and setting repeats compensating controller In the delay time t output of low-pass filter is exported again after repeating PID, and is compensated before exporting again.
CN201610947397.4A 2016-10-26 2016-10-26 Control method based on Repetitive control compensation fuzzy neuron PID quadrotor Active CN106647781B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610947397.4A CN106647781B (en) 2016-10-26 2016-10-26 Control method based on Repetitive control compensation fuzzy neuron PID quadrotor

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610947397.4A CN106647781B (en) 2016-10-26 2016-10-26 Control method based on Repetitive control compensation fuzzy neuron PID quadrotor

Publications (2)

Publication Number Publication Date
CN106647781A CN106647781A (en) 2017-05-10
CN106647781B true CN106647781B (en) 2019-09-06

Family

ID=58821304

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610947397.4A Active CN106647781B (en) 2016-10-26 2016-10-26 Control method based on Repetitive control compensation fuzzy neuron PID quadrotor

Country Status (1)

Country Link
CN (1) CN106647781B (en)

Families Citing this family (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108958281B (en) * 2017-05-19 2021-04-02 上海交通大学 Unmanned aerial vehicle suspension transportation stability analysis and control method based on differential quadrature method
CN107168047B (en) * 2017-05-26 2020-10-16 河北汉光重工有限责任公司 Method for improving control performance of repeated compensation PID control system
CN109308074A (en) * 2017-07-28 2019-02-05 深圳禾苗通信科技有限公司 A kind of compensation method and system of drone center of unmanned aerial vehicle offset
CN107272723B (en) * 2017-08-02 2020-06-19 郑州轻工业学院 Five-gear fuzzy height control method for unmanned butterfly aircraft
CN107608215A (en) * 2017-10-18 2018-01-19 齐鲁工业大学 A kind of mapping UAV Attitude control method based on RBF neural
KR102021384B1 (en) * 2017-11-14 2019-09-20 주식회사 셈웨어 Method for creating dynamic modeling of drone based on artificial intelligence
CN108762283B (en) * 2018-05-08 2021-04-13 哈尔滨工业大学 Attitude control method, device and system of satellite platform with rotating load
CN108663940B (en) * 2018-08-20 2019-09-24 西北工业大学 Aircraft neural network lea rning control method based on the compound estimation of lump
CN108762098B (en) * 2018-08-20 2019-11-01 西北工业大学 Non-minimum phase aircraft neural network control method based on Hybrid Learning
CN108897227B (en) * 2018-08-20 2019-09-24 西北工业大学 Non-linear strict feedback systems overall situation finite time neural network control method
CN108828957B (en) * 2018-08-20 2019-11-05 西北工业大学 Aircraft overall situation finite time neural network control method based on handover mechanism
CN109062234B (en) * 2018-08-20 2019-11-05 西北工业大学 A kind of non-minimum phase aircraft Hybrid Learning sliding-mode control
CN109062049B (en) * 2018-08-20 2019-09-24 西北工业大学 Strict feedback systems neural network control method based on the compound estimation of lump
CN109116738B (en) * 2018-09-27 2021-04-13 杭州电子科技大学 Two-degree-of-freedom internal model control analysis method of industrial heating furnace
CN110347042A (en) * 2019-07-15 2019-10-18 广西师范大学 A kind of three axis holder sliding-mode controls based on discrete tracking-differentiator and RBF neural compensation
CN111026146B (en) * 2019-12-24 2021-04-06 西北工业大学 Attitude control method for composite wing vertical take-off and landing unmanned aerial vehicle
CN111399376B (en) * 2020-03-07 2021-03-19 中国地质大学(武汉) Two-dimensional repetitive controller design optimization method of T-S fuzzy system

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6445962B1 (en) * 1999-03-15 2002-09-03 Fisher Rosemount Systems, Inc. Auto-tuning in a distributed process control environment
CN102945046A (en) * 2012-11-15 2013-02-27 中国兵器工业计算机应用技术研究所 Control method of unmanned aircraft
CN103853050A (en) * 2014-03-19 2014-06-11 湖北蔚蓝国际航空学校有限公司 PID optimization control method of four-rotor aircraft
CN105242679A (en) * 2015-10-22 2016-01-13 电子科技大学 Method for designing control system of four rotor aircraft
CN105700538B (en) * 2016-01-28 2019-04-12 武汉光庭信息技术股份有限公司 Track follower method based on neural network and pid algorithm

Also Published As

Publication number Publication date
CN106647781A (en) 2017-05-10

Similar Documents

Publication Publication Date Title
Zuo et al. Augmented L 1 adaptive tracking control of quad-rotor unmanned aircrafts
Liu et al. Robust three-loop trajectory tracking control for quadrotors with multiple uncertainties
Chen et al. Nonlinear control of underactuated systems subject to both actuated and unactuated state constraints with experimental verification
Sreenath et al. Trajectory generation and control of a quadrotor with a cable-suspended load-a differentially-flat hybrid system
Bangura et al. Real-time model predictive control for quadrotors
CN105867401B (en) The spacecraft attitude fault tolerant control method of single-gimbal control moment gyros
Sabatino Quadrotor control: modeling, nonlinearcontrol design, and simulation
Shao et al. Robust dynamic surface trajectory tracking control for a quadrotor UAV via extended state observer
Cabecinhas et al. A nonlinear quadrotor trajectory tracking controller with disturbance rejection
Santos et al. Intelligent fuzzy controller of a quadrotor
Do et al. Nonlinear formation control of unicycle-type mobile robots
Voos Nonlinear control of a quadrotor micro-UAV using feedback-linearization
Madani et al. Backstepping control for a quadrotor helicopter
Omari et al. Nonlinear control of VTOL UAVs incorporating flapping dynamics
Chen et al. A reconfiguration scheme for quadrotor helicopter via simple adaptive control and quantum logic
Szafranski et al. Different approaches of PID control UAV type quadrotor
Huang et al. Adaptive tracking control of underactuated quadrotor unmanned aerial vehicles via backstepping
Derafa et al. Super twisting control algorithm for the attitude tracking of a four rotors UAV
Chen et al. Robust backstepping sliding-mode control and observer-based fault estimation for a quadrotor UAV
Fang et al. Adaptive integral backstepping control of a micro-quadrotor
Voos Nonlinear state-dependent Riccati equation control of a quadrotor UAV
Cutler Design and control of an autonomous variable-pitch quadrotor helicopter
Das et al. Dynamic inversion with zero-dynamics stabilisation for quadrotor control
Madani et al. Adaptive control via backstepping technique and neural networks of a quadrotor helicopter
Mian et al. Nonlinear flight control strategy for an underactuated quadrotor aerial robot

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant