CN113900372A - Unmanned ship course keeping method based on neural network active disturbance rejection control - Google Patents
Unmanned ship course keeping method based on neural network active disturbance rejection control Download PDFInfo
- Publication number
- CN113900372A CN113900372A CN202111034047.6A CN202111034047A CN113900372A CN 113900372 A CN113900372 A CN 113900372A CN 202111034047 A CN202111034047 A CN 202111034047A CN 113900372 A CN113900372 A CN 113900372A
- Authority
- CN
- China
- Prior art keywords
- course
- unmanned ship
- neural network
- disturbance rejection
- active disturbance
- 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.)
- Pending
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 64
- 238000000034 method Methods 0.000 title claims abstract description 31
- 230000001105 regulatory effect Effects 0.000 claims abstract description 8
- 230000007704 transition Effects 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 7
- 230000015572 biosynthetic process Effects 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 230000002596 correlated effect Effects 0.000 claims description 3
- 230000010354 integration Effects 0.000 claims description 3
- 238000012886 linear function Methods 0.000 claims description 3
- 238000003786 synthesis reaction Methods 0.000 claims description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 abstract description 4
- 230000000694 effects Effects 0.000 description 9
- 230000006870 function Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 230000000875 corresponding effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000010355 oscillation Effects 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B11/00—Automatic controllers
- G05B11/01—Automatic controllers electric
- G05B11/36—Automatic controllers electric with provision for obtaining particular characteristics, e.g. proportional, integral, differential
- G05B11/42—Automatic controllers electric with provision for obtaining particular characteristics, e.g. proportional, integral, differential for obtaining a characteristic which is both proportional and time-dependent, e.g. P. I., P. I. D.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Feedback Control In General (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
A unmanned ship course keeping method based on active disturbance rejection control of a neural network comprises the following steps: step A: positioning the orientation of a hull and the orientation of a boat head of the unmanned boat through a gyrocompass to determine the actual course of the unmanned boat; and B: judging whether a course deviation angle exists according to a preset course and an actual course of the unmanned ship, and if so, inputting the course deviation angle into a neural network active disturbance rejection controller; and C: the rotation speeds of a left motor and a right motor of the unmanned ship are regulated and controlled through the neural network active disturbance rejection controller, and the orientation of the propeller is regulated through the left motor and the right motor of the unmanned ship, so that the unmanned ship keeps the course. The invention can detect the actual course of the unmanned ship in real time, compare the actual course with the set course, and then adjust the rotating speed of the left motor and the right motor to realize the angle control, thereby keeping the course of the unmanned ship from deviating due to the external force of the water surface, and further avoiding the problem of increasing the traveling distance and the traveling time caused by the air route re-planning.
Description
Technical Field
The invention relates to the technical field of unmanned boats, in particular to an unmanned boat course keeping method based on active disturbance rejection control of a neural network.
Background
Unmanned ship, a instrument that is applied to marine operations, unmanned ship very easily receives the exogenic action influence on the surface of water, leads to unmanned ship self course to receive the skew. In the face of interference of wind waves to the advancing route of the unmanned ship, two main ideas exist for solving the problem at present: one method is to plan a local path in the process of unmanned ship moving, so that the unmanned ship can return to the original channel after receiving interference offset, but the method can increase the moving time of the unmanned ship and increase the energy consumption. Another approach is to create a course holder and add drive to keep the drone from deviating from the channel, thus eliminating the problems associated with channel changes.
The course controller mainly comprises a PID controller, a backstepping controller, robust control and the like at present. The linear control represented by the PID needs to be based on an accurate mathematical model, but because the unmanned ship is subjected to internal interference and is also easily subjected to interference of external factors such as wind, waves, flow and the like, the coefficient of the traditional PID controller needs to be continuously adjusted, and the control effect is not ideal. And non-linear controllers such as: the backstepping controller, the robust control and the like have good control effects when the system model has high precision, but the parameters and the structure of the mathematical model of the unmanned ship have uncertainty due to the continuous changes of the loading state, the draught, the navigational speed and the like of the unmanned ship, and the actual control effect cannot be expected. And a dual motor propelled course balancer. When encountering transverse wind waves, the reaction effect is poor, the navigation channel cannot be returned to the navigation channel at the first time, and the navigation channel returns to the preset course.
Disclosure of Invention
The invention aims to provide a unmanned ship course keeping method based on neural network active disturbance rejection control aiming at the defects in the background technology, the invention can detect the actual course of the unmanned ship in real time, compare the actual course with the set course, and then adjust the rotating speed of a left motor and a right motor to realize angle control, so that the course of the unmanned ship can be kept from being deviated due to the external force of the water surface, and the problem of increasing the advancing distance and the advancing time caused by air route re-planning can be solved.
In order to achieve the purpose, the invention adopts the following technical scheme:
a unmanned ship course keeping method based on active disturbance rejection control of a neural network comprises the following steps:
step A: positioning the orientation of a hull and the orientation of a boat head of the unmanned boat through a gyrocompass to determine the actual course of the unmanned boat;
and B: judging whether a course deviation angle exists according to a preset course and an actual course of the unmanned ship, and if so, inputting the course deviation angle into a neural network active disturbance rejection controller;
and C: the rotation speeds of a left motor and a right motor of the unmanned ship are regulated and controlled through the neural network active disturbance rejection controller, and the orientation of the propeller is regulated through the left motor and the right motor of the unmanned ship, so that the unmanned ship keeps the course.
Preferably, in the step B, the determining whether the heading deviation angle exists includes:
when the actual course is consistent with the preset course, no course deviation angle exists;
when the actual course is positioned at the right side of the preset course, a course deviation angle exists, and the current course deviation angle is a positive value;
when the actual course is positioned on the left side of the preset course, a course deviation angle exists, and the current course deviation angle is a negative value;
and carrying out angle difference on the preset course and the actual course to obtain a course deviation angle.
Preferably, the step C specifically includes:
step C1: inputting the current course deviation angle into the input end of the neural network active disturbance rejection controller;
step C2: the neural network active disturbance rejection controller calculates input parameters to obtain output values;
step C3: the output value is output to the unmanned ship to control the rotation speed difference of a left motor and a right motor of the unmanned ship, so that the unmanned ship can steer in real time;
step C4: and acquiring a new course deviation angle, judging whether the current course deviation angle is 0, if so, maintaining the current course, and if not, executing the steps C1 to C3.
Preferably, the neural network active disturbance rejection controller comprises a differential tracker, a nonlinear state feedback module, an extended observer and a neural network module;
the differential tracker is used for arranging a transition process according to the preset course and the limitation of a controlled object and providing each order derivative of the transition process;
the extended observer is used for reconstructing the neural network active disturbance rejection controller through an input value and an output value so as to compensate the disturbance borne by the neural network active disturbance rejection controller in real time;
the nonlinear state feedback module is used for forming the actual control quantity of the unmanned ship through the fitting state variable error and the disturbance compensation of the extended observer;
the neural network module is used for setting the parameters in the nonlinear state feedback module.
Preferably, the differential tracker includes arranging for the transition and providing the first derivatives of the transition using the following equations;
wherein:
x1a differential representing a heading;
x2a differential amount representing a differential value of the heading;
k represents an amount of time, x1(k +1) denotes the current (k +1) time x1Value of (a), x1(k) Represents the last time (k time) x1A value of (d);
h represents an integration step;
flan denotes the steepest control synthesis function, equivalent fh；
r represents the tracking speed.
Preferably, the extended observer compensates the disturbance suffered by the neural network active disturbance rejection controller in real time by using the following formula;
wherein:
e represents the error of the system state variable;
z1、z2an observation representing a system state variable;
z3represents an estimate of the total disturbance of the system;
k represents an amount of time, i.e., time k;
y represents the actual output quantity of the unmanned ship at the moment k;
alpha is a non-linear parameter, and can be generally taken as alpha1＝1，α2＝0.5，α3＝0.25；
β0i(i ═ 1, 2, and 3) represent output error correction gain values;
δ represents the linear segment interval length;
fal (e, α, δ) represents a non-linear function, which is specifically
The sgn () function is a step function.
Preferably, the nonlinear state feedback module forms the actual control quantity of the unmanned ship through the following formula;
wherein:
u0an output value representing a nonlinear state error feedback;
u represents u0Obtained after compensation by an extended observerOutputting a control value;
e1、e2an error signal indicative of a scheduled transition;
α4、α5represents: is a non-linear parameter in general4＝0.75，α5＝1.25；
δ3、δ4A certain length representing a linear segment interval;
β1i(i ═ 1, 2) denotes adjustable parameters;
b is a constant value, and the value of b is positively correlated with the delay degree of the control object.
Compared with the prior art, the invention has the following technical effects:
the invention can detect the actual course of the unmanned ship in real time, compare the actual course with the set course, and then adjust the rotating speed of the left motor and the right motor to realize the angle control, thereby keeping the course of the unmanned ship from deviating due to the external force of the water surface, and further avoiding the problem of increasing the traveling distance and the traveling time caused by the air route re-planning. Compared with the existing fuzzy active disturbance rejection controller, the neural network active disturbance rejection controller selected by the invention reduces the requirements on the experience of research personnel and further improves the stability of a control system.
Drawings
FIG. 1 is a flow chart of an unmanned boat heading maintenance method according to one embodiment of the invention;
FIG. 2 is a block diagram of a neural network active disturbance rejection controller according to one embodiment of the present invention;
FIG. 3 is a flow chart of the LM-RBF neural network of one embodiment of the present invention;
fig. 4 is a schematic structural diagram of an RBF neural network according to an embodiment of the present invention.
Detailed Description
The technical scheme of the invention is further explained by the specific implementation mode in combination with the attached drawings.
The active disturbance rejection technology is improved and developed from a PID technology, compared with the traditional PID control, the quality and the control precision of the active disturbance rejection control are higher, the internal disturbance and the external disturbance are classified into total disturbance by the active disturbance rejection, the total disturbance borne by the system is observed, estimated and compensated by an ESO (electronic service organization) in the active disturbance rejection control, and the nonlinear and uncertain system containing unknown disturbance is changed into a standard linear system. The unmanned ship system mainly closer to the actual unmanned ship system can also enable the unmanned ship system to have stronger anti-interference capability and better control effect. However, the problem of parameters exists in the pure active disturbance rejection control, and the parameters are frequently required to be adjusted in practical application, so that the method of combining the neural network and the active disturbance rejection control is adopted, and partial parameters of the active disturbance rejection controller are adjusted through the self-learning adaptive capacity, so that the control effect is better. Therefore, the requirements on the experience of research personnel are reduced, and the stability of the control system is further improved. In addition, the control method of the four-motor double-propeller adopted by the design can change the orientation of the propeller by adjusting the motor above the propeller when the side wind waves are faced, so that a side thrust is generated to resist the force from the side surface, meanwhile, the original course can be returned by means of the rotation speed difference, and the effect is better.
As shown in fig. 1, the method specifically comprises the following steps:
step A: positioning the orientation of a hull and the orientation of a boat head of the unmanned boat through a gyrocompass to determine the actual course of the unmanned boat;
and B: judging whether a course deviation angle exists according to a preset course and an actual course of the unmanned ship, and if so, inputting the course deviation angle into a neural network active disturbance rejection controller;
preferably, in the step B, the determining whether the heading deviation angle exists includes:
when the actual course is consistent with the preset course, no course deviation angle exists;
when the actual course is positioned at the right side of the preset course, a course deviation angle exists, and the current course deviation angle is a positive value;
when the actual course is positioned on the left side of the preset course, a course deviation angle exists, and the current course deviation angle is a negative value;
and carrying out angle difference on the preset course and the actual course to obtain a course deviation angle.
The method comprises the following steps of obtaining a course deviation angle by subtracting a preset course from an actual course, inputting the course deviation angle into a neural network active disturbance rejection controller after obtaining the course deviation angle, outputting the output of the active disturbance rejection controller to an unmanned ship, and controlling left and right propellers of the unmanned ship and a motor above the propellers, thereby adjusting the steering of the unmanned ship, and specifically comprises the following steps:
and C: the rotation speeds of a left motor and a right motor of the unmanned ship are regulated and controlled through the neural network active disturbance rejection controller, and the orientation of the propeller is regulated through the left motor and the right motor of the unmanned ship, so that the unmanned ship keeps the course.
Preferably, the step C specifically includes:
step C1: inputting the current course deviation angle into the input end of the neural network active disturbance rejection controller;
step C2: the neural network active disturbance rejection controller calculates input parameters to obtain output values;
step C3: the output value is output to the unmanned ship to control the rotation speed difference of a left motor and a right motor of the unmanned ship, so that the unmanned ship can steer in real time;
step C4: and acquiring a new course deviation angle, judging whether the current course deviation angle is 0, if so, maintaining the current course, and if not, executing the steps C1 to C3.
Different heading deviation angles represent different states of the unmanned vehicle. The course retainer realizes real-time steering of the unmanned ship through the rotation speed difference of the left motor and the right motor, then the new actual course of the unmanned ship is measured through the gyro compass and fed back to the input end, and the course of the unmanned ship is controlled by obtaining the deviation angle again and is circulated. When the actual course of the unmanned ship is consistent with the preset course, the deviation angle is 0, the output of the active disturbance rejection controller is not changed any more, and the existing course is maintained to run.
Preferably, as shown in fig. 2, the neural network active disturbance rejection controller includes a differential tracker, a nonlinear state feedback module, an extended observer, and a neural network module;
the differential tracker is used for arranging a transition process according to the preset course and the limitation of a controlled object and providing each order derivative of the transition process, so that the overshoot caused by the sudden change of the preset course in the classical control can be effectively improved;
preferably, the differential tracker includes arranging for the transition and providing the first derivatives of the transition using the following equations;
wherein:
x1a differential representing a heading;
x2a differential amount representing a differential value of the heading;
k represents an amount of time, x1(k +1) denotes the current (k +1) time x1Value of (a), x1(k) Represents the last time (k time) x1A value of (d);
h represents an integration step;
flan denotes the steepest control synthesis function, equivalent fh；
r represents the tracking speed.
Wherein r and h are parameters needing to be set;
further, the extended observer is a core part of the neural network active disturbance rejection controller, and reconstructs the neural network active disturbance rejection controller through an input value and an output value so as to compensate the disturbance borne by the neural network active disturbance rejection controller in real time;
the extension state observer is used for estimating and observing the total disturbance of the system in real time, and an unmanned ship system with unknown disturbance and uncertain interior is compensated into a standard system with linear integrators connected in series through proper compensation.
Preferably, the extended observer compensates the disturbance suffered by the neural network active disturbance rejection controller in real time by using the following formula;
wherein:
e represents the error of the system state variable;
z1、z2an observation representing a system state variable;
z3represents an estimate of the total disturbance of the system;
k represents an amount of time, i.e., time k;
y represents the actual output quantity of the unmanned ship at the moment k;
alpha is a non-linear parameter, and can be generally taken as alpha1＝1，α2＝0.5，α3＝0.25；
β0i(i ═ 1, 2, and 3) represent output error correction gain values;
δ represents the linear segment interval length;
fal (e, α, δ) represents a non-linear function, which is specifically
The sgn () function is a step function;
wherein, beta0iAnd (i is 1, 2 and 3) and delta are parameters needing to be adjusted.
Further, the nonlinear state feedback module is used for forming an actual control quantity of the unmanned ship through fitting state variable errors and disturbance compensation of an extended observer;
preferably, the nonlinear state feedback module forms the actual control quantity of the unmanned ship through the following formula;
wherein:
u0an output value representing a nonlinear state error feedback;
u represents u0An output control value is obtained after compensation is carried out by the extended observer;
e1、e2an error signal indicative of a scheduled transition;
α4、α5represents: is a non-linear parameter in general4＝0.75，α5＝1.25；
δ3、δ4A certain length representing a linear segment interval;
β1i(i ═ 1, 2) denotes adjustable parameters;
b is a constant value, and the value of b is positively correlated with the delay degree of the control object.
Further, as shown in fig. 3, the neural network module is configured to adjust parameters in the nonlinear state feedback module.
The neural network active disturbance rejection controller has many parameters which need to be set, but the neural network module is really needed to be set as long as beta in the nonlinear state feedback module1i(i-1, 2), because the two parameters have great influence on the performance of the neural network active disturbance rejection controller, setting the two parameters can enable the neural network active disturbance rejection controller to obtain more ideal effect;
an RBF neural network is selected, and the RBF neural network is a feedforward type neural network. The structure of the three-layer optical network is shown in figure 4:
the method is an n-m-1 type neural network, wherein n is the number of input nodes, m is the number of implicit nodes, and the number of output nodes is 1. The input layer to the hidden layer is non-linearly mapped, and the base function is selected from standard Gaussian function
Where j is the node serial number of the hidden layer, i.e., j is 1, 2.. m, and X is the input quantity X of the input layer [ X1 × 2.. xn ]]，cjIs the center vector of the jth node in the network, and
to find the Euclidean distance, σ, of the twohThe width of the h node of the hidden layer. From hidden layer to output layer is
wjThe weighted value is multiplied by the corresponding hidden layer output and then added to output.
The inputs to the neural network are X ═ u (k), y (k-2), y (k-1)]T；
Wherein u (k) and y (k) are input and output of the control system respectively, and the performance optimization index of the neural network is
According to the method, an LM algorithm (a nonlinear least square optimization algorithm) is added on the basis of an RBF neural network, compared with the traditional RBF, the LM algorithm is replaced by the nonlinear least square optimization algorithm, and the problems of poor stability and low convergence speed of a gradient descent method are solved. The method introduces a momentum factor alpha to achieve the aim of weakening oscillation in parameter adjustment. Beta to be adjusted after introduction of LM algorithm1iThe iterative algorithm of (f ═ 1, 2) is as follows:
in the formula
The jacobian information expression identified by the neural network is:
eta is the learning rate, and a is a dynamic factor, and the value of the dynamic factor is related to the error index, namely:
the technical principle of the present invention is described above in connection with specific embodiments. The description is made for the purpose of illustrating the principles of the invention and should not be construed in any way as limiting the scope of the invention. Based on the explanations herein, those skilled in the art will be able to conceive of other embodiments of the present invention without inventive effort, which would fall within the scope of the present invention.
Claims (7)
1. A unmanned ship course keeping method based on active disturbance rejection control of a neural network is characterized by comprising the following steps: the method comprises the following steps:
step A: positioning the orientation of a hull and the orientation of a boat head of the unmanned boat through a gyrocompass to determine the actual course of the unmanned boat;
and B: judging whether a course deviation angle exists according to a preset course and an actual course of the unmanned ship, and if so, inputting the course deviation angle into a neural network active disturbance rejection controller;
and C: the rotation speeds of a left motor and a right motor of the unmanned ship are regulated and controlled through the neural network active disturbance rejection controller, and the orientation of the propeller is regulated through the left motor and the right motor of the unmanned ship, so that the unmanned ship keeps the course.
2. The unmanned ship heading keeping method based on the neural network active disturbance rejection control as claimed in claim 1, wherein:
in the step B, judging whether the course deviation angle exists comprises the following steps:
when the actual course is consistent with the preset course, no course deviation angle exists;
when the actual course is positioned at the right side of the preset course, a course deviation angle exists, and the current course deviation angle is a positive value;
when the actual course is positioned on the left side of the preset course, a course deviation angle exists, and the current course deviation angle is a negative value;
and carrying out angle difference on the preset course and the actual course to obtain a course deviation angle.
3. The unmanned ship heading keeping method based on the neural network active disturbance rejection control as claimed in claim 1, wherein:
the step C specifically comprises the following steps:
step C1: inputting the current course deviation angle into the input end of the neural network active disturbance rejection controller;
step C2: the neural network active disturbance rejection controller calculates input parameters to obtain output values;
step C3: the output value is output to the unmanned ship to control the rotation speed difference of a left motor and a right motor of the unmanned ship, so that the unmanned ship can steer in real time;
step C4: and acquiring a new course deviation angle, judging whether the current course deviation angle is 0, if so, maintaining the current course, and if not, executing the steps C1 to C3.
4. The unmanned ship heading keeping method based on the neural network active disturbance rejection control as claimed in claim 1, wherein:
the neural network active disturbance rejection controller comprises a differential tracker, a nonlinear state feedback module, an extended observer and a neural network module;
the differential tracker is used for arranging a transition process according to the preset course and the limitation of a controlled object and providing each order derivative of the transition process;
the extended observer is used for reconstructing the neural network active disturbance rejection controller through an input value and an output value so as to compensate the disturbance borne by the neural network active disturbance rejection controller in real time;
the nonlinear state feedback module is used for forming the actual control quantity of the unmanned ship through the fitting state variable error and the disturbance compensation of the extended observer;
the neural network module is used for setting the parameters in the nonlinear state feedback module.
5. The unmanned ship heading keeping method based on the neural network active disturbance rejection control as claimed in claim 4, wherein:
the differential tracker includes arranging for a transition and providing the order derivatives of the transition using the following equations;
wherein:
x1a differential representing a heading;
x2a differential amount representing a differential value of the heading;
k represents an amount of time, x1(k +1) denotes the current (k +1) time x1Value of (a), x1(k) Represents the last time (k time) x1A value of (d);
h represents an integration step;
flan denotes the steepest control synthesis function, equivalent fh；
r represents the tracking speed.
6. The unmanned ship heading keeping method based on the neural network active disturbance rejection control as claimed in claim 5, wherein:
the extended observer compensates the disturbance borne by the neural network active disturbance rejection controller in real time by using the following formula;
wherein:
e represents the error of the system state variable;
z1、z2an observation representing a system state variable;
z3represents an estimate of the total disturbance of the system;
k represents an amount of time, i.e., time k;
y represents the actual output quantity of the unmanned ship at the moment k;
alpha is a non-linear parameter, and can be generally taken as alpha1＝1，α2＝0.5，α3＝0.25；
β0i(i ═ 1, 2, and 3) represent output error correction gain values;
δ represents the linear segment interval length;
fal (e, α, δ) represents a nonlinear function.
7. The unmanned ship heading keeping method based on the neural network active disturbance rejection control as claimed in claim 6, wherein:
the nonlinear state feedback module forms the actual control quantity of the unmanned ship through the following formula;
wherein:
u0an output value representing a nonlinear state error feedback;
u represents u0An output control value is obtained after compensation is carried out by the extended observer;
fal (e, α, δ) represents a non-linear function, which is specifically
The sgn () function is a step function;
e1、e2an error signal indicative of a scheduled transition;
α4、α5represents; is a non-linear parameter in general4＝0.75，α5＝1.25；
δ3、δ4A certain length representing a linear segment interval;
β1i(i ═ 1, 2) denotes adjustable parameters;
b is a constant value, and the value of b is positively correlated with the delay degree of the control object.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111034047.6A CN113900372A (en) | 2021-09-03 | 2021-09-03 | Unmanned ship course keeping method based on neural network active disturbance rejection control |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111034047.6A CN113900372A (en) | 2021-09-03 | 2021-09-03 | Unmanned ship course keeping method based on neural network active disturbance rejection control |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113900372A true CN113900372A (en) | 2022-01-07 |
Family
ID=79188591
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111034047.6A Pending CN113900372A (en) | 2021-09-03 | 2021-09-03 | Unmanned ship course keeping method based on neural network active disturbance rejection control |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113900372A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115616907A (en) * | 2022-09-22 | 2023-01-17 | 上海海事大学 | Unmanned ship course intelligent planning method and controller |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105785999A (en) * | 2016-04-27 | 2016-07-20 | 大连海事大学 | Unmanned surface vehicle course motion control method |
| CN110209054A (en) * | 2019-06-11 | 2019-09-06 | 大连海事大学 | Unmanned ships and light boats course Active Disturbance Rejection Control system based on RBF neural |
| CN111897347A (en) * | 2020-08-27 | 2020-11-06 | 广东工业大学 | Course retainer of double-motor-propelled unmanned ship based on neural network PID control and retaining method |
| CN112558481A (en) * | 2020-12-17 | 2021-03-26 | 广东工业大学 | Course angle active disturbance rejection control method in ship control system and related device |
| CN112947477A (en) * | 2021-03-23 | 2021-06-11 | 哈尔滨工程大学 | Unmanned ship course active disturbance rejection control system and control method |
-
2021
- 2021-09-03 CN CN202111034047.6A patent/CN113900372A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105785999A (en) * | 2016-04-27 | 2016-07-20 | 大连海事大学 | Unmanned surface vehicle course motion control method |
| CN110209054A (en) * | 2019-06-11 | 2019-09-06 | 大连海事大学 | Unmanned ships and light boats course Active Disturbance Rejection Control system based on RBF neural |
| CN111897347A (en) * | 2020-08-27 | 2020-11-06 | 广东工业大学 | Course retainer of double-motor-propelled unmanned ship based on neural network PID control and retaining method |
| CN112558481A (en) * | 2020-12-17 | 2021-03-26 | 广东工业大学 | Course angle active disturbance rejection control method in ship control system and related device |
| CN112947477A (en) * | 2021-03-23 | 2021-06-11 | 哈尔滨工程大学 | Unmanned ship course active disturbance rejection control system and control method |
Non-Patent Citations (3)
| Title |
|---|
| 姜涛: "基于RBF神经网络的船舶航向自抗扰控制" * |
| 范兴宇: "基于RBF神经网络的无人水面艇自抗扰控制" * |
| 赵顺利，李伟，张文拴: "基于径向基函数神经网络的船舶航迹自抗扰控制" * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115616907A (en) * | 2022-09-22 | 2023-01-17 | 上海海事大学 | Unmanned ship course intelligent planning method and controller |
| US11977383B2 (en) | 2022-09-22 | 2024-05-07 | Shanghai Maritime University | Intelligent course planning method and controller for unmanned surface vehicle |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109460043B (en) | Multimode nonsingular terminal sliding mode based ship track active disturbance rejection control method | |
| CN111580523B (en) | Unmanned ship path tracking active disturbance rejection control method based on sideslip angle compensation | |
| CN108845576B (en) | Thrust distribution method based on combination of particle swarm optimization and sequence quadratic programming | |
| CN108267955B (en) | Motion control method for autonomous berthing of unmanned ship | |
| Li et al. | Line-of-sight-based global finite-time stable path following control of unmanned surface vehicles with actuator saturation | |
| CN109656142B (en) | Cascade structure model-free self-adaptive guidance method for unmanned ship | |
| CN111897347B (en) | Course maintaining method of double-motor propulsion unmanned ship based on neural network PID control | |
| CN113031614B (en) | Ocean vessel course control composite optimization oil-saving method | |
| Zhang et al. | LVS guidance principle and adaptive neural fault-tolerant formation control for underactuated vehicles with the event-triggered input | |
| CN113341693A (en) | Course control method based on asymmetric phase difference and amplitude of flapping wings | |
| CN113467231A (en) | Unmanned ship path tracking method based on sideslip compensation ILOS guidance law | |
| CN113900372A (en) | Unmanned ship course keeping method based on neural network active disturbance rejection control | |
| CN109116727B (en) | PID type first-order full-format model-free self-adaptive cruise control algorithm based on low-pass filter | |
| CN115951581A (en) | High-speed unmanned ship path tracking control method based on improved EMPC | |
| He et al. | Finite time course keeping control for unmanned surface vehicles with command filter and rudder saturation | |
| Tang et al. | Robust fixed-time trajectory tracking control of the dynamic positioning ship with actuator saturation | |
| CN113342012A (en) | Course control method of sliding and flapping integrated aircraft | |
| Shao et al. | Sideslip angle observation-based LOS and adaptive finite-time path following control for sailboat | |
| CN116820081A (en) | Under-actuated unmanned ship path tracking control method based on model-free self-adaptive sliding mode | |
| CN109828462A (en) | Wave glider becomes under the speed of a ship or plane adaptive bow to controller and control method | |
| CN114660942A (en) | Combined type helicopter full-mode self-adaptive control method | |
| CN112180918B (en) | Synchronous control method and system for double mobile robots | |
| Shao et al. | Sailboat path following control based on LOS with sideslip angle observation and finite-time backstepping | |
| Xia et al. | Neural network based nonlinear model predictive control for ship path following | |
| JPS59220496A (en) | Automatic steering system for vessel |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20220107 |