CN112394740B - Composite anti-interference track tracking control algorithm for advancing process of unmanned rolling machine - Google Patents

Composite anti-interference track tracking control algorithm for advancing process of unmanned rolling machine Download PDF

Info

Publication number
CN112394740B
CN112394740B CN201910740160.2A CN201910740160A CN112394740B CN 112394740 B CN112394740 B CN 112394740B CN 201910740160 A CN201910740160 A CN 201910740160A CN 112394740 B CN112394740 B CN 112394740B
Authority
CN
China
Prior art keywords
disturbance
distance error
angle
course angle
rolling machine
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
CN201910740160.2A
Other languages
Chinese (zh)
Other versions
CN112394740A (en
Inventor
谢辉
宋康
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.)
Tianjin University
Original Assignee
Tianjin 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 Tianjin University filed Critical Tianjin University
Priority to CN201910740160.2A priority Critical patent/CN112394740B/en
Publication of CN112394740A publication Critical patent/CN112394740A/en
Application granted granted Critical
Publication of CN112394740B publication Critical patent/CN112394740B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
    • G05D1/10—Simultaneous control of position or course in three dimensions
    • G05D1/101—Simultaneous control of position or course in three dimensions specially adapted for aircraft

Landscapes

  • Engineering & Computer Science (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention discloses a composite anti-interference track tracking control algorithm of an unmanned rolling machine, which comprises the following steps: calculating an actual distance error in the track tracking of the rolling machine according to the actual position of the rolling machine and the target track of the rolling machine; the anti-interference controller of the distance error calculates a target course angle of the rolling machine through the actual distance error and the total disturbance of the distance error obtained by the instant disturbance observer, so that the track tracking distance error of the rolling machine tends to zero; and the anti-interference controller of the course angle calculates the target steering wheel turning angle of the rolling machine through the target course angle, the course total disturbance obtained by the instant disturbance observer and the course angle disturbance obtained based on the steering system model parameter learner, so that the actual course angle of the rolling machine approaches to the target course angle, and the target steering wheel turning angle is sent to the steering wheel turning angle controller. The control algorithm can solve the problems of fast small-range interference and slow large-range interference.

Description

Composite anti-interference track tracking control algorithm for advancing process of unmanned rolling machine
Technical Field
The invention relates to the technical field of unmanned rolling machine control, in particular to a hierarchical step-by-step self-learning composite anti-interference track tracking control algorithm for an advancing process of an unmanned rolling machine.
Background
The accurate tracking of the target running track by the unmanned rolling machine is the key for ensuring the quality, efficiency and safety of rolling operation. Conventional PID (feedback-integral-derivative) control is one of the most widely used control algorithms. For example, Zhang et al, the university of qinghua, and benzyl Yongming et al, the university of congratulation, both used PID feedback control for trajectory tracking control of the roller compactor. However, the PID feedback control takes control action only after significant trajectory tracking errors occur. The passive control method of 'after adjustment' can easily cause the 'dragon drawing' phenomenon to occur in the track tracking process of the vehicle with the hinged structure of the rolling machine, so that the milling leakage or the over milling of the bin surface is caused.
Therefore, the university of Tongji, Benyongming and the like, put forward a track tracking control algorithm based on fuzzy PID, however, the design of a fuzzy rule is complex, and the research and development efficiency of an unmanned rolling machine control algorithm is influenced. Similarly, the segmented PID parameter tuning can also improve the control accuracy, but still faces the difficulty of complicated parameter tuning. Therefore, the traditional PID feedback control can not meet the requirement of high-precision track tracking control of the rolling machine.
For this reason, model-based trajectory tracking control algorithms have been extensively studied. In the field of traditional passenger vehicles, algorithms such as a preview method, model predictive control, adaptive control and the like are widely researched. Fang et al propose feed forward control based on compensation of the vibratory exciting force of the front steel wheel of the roller compactor to improve the tracking accuracy in the vibration mode. However, the roller is an articulated structure vehicle, has poor lateral stability and complex interaction with the bin floor. Due to the complex condition of the bin surface of the rolling machine, the interference of the stones with different grain diameters is large. When the conventional model-based control is adopted, the deviation of the model from the real object easily causes the deterioration of the control quality.
In view of the above problems, scholars at home and abroad research various control algorithms with high adaptability to model deviations. For example, Dekker et al propose a path tracking control algorithm combining iterative learning control and feedback linearization for autonomous wheeled vehicles. Yang et al have developed a simple iterative learning control algorithm (ILC) for all-wheel-steered vehicles to improve the path-tracking capability of the vehicle. Aiming at the track tracking problem of a rolling machine, the winter and spring of the Yao at Tianjin university combines an active disturbance rejection control method with a preview algorithm, and improves the track tracking effect. Although the active disturbance rejection control algorithm can equate the deviation between the model and the actual object as total disturbance, and use an ESO (total disturbance immediate observer) to perform active observation and compensation, the observation capability of the ESO is affected by sampling frequency and measurement noise, and the adaptability to a large range of uncertainty is relatively limited.
In fact, in the case of a compactor, which is an engineering compacting machine working in an extremely harsh environment, the compaction, friction and resistance characteristics of the surface of the bin are constantly changing during operation, and the characteristics of the steering hydraulic system of the compactor are also constantly changing. The existing control method is difficult to adapt to the control problem which simultaneously comprises fast small-range interference (rolling through stone blocks, ground slippage and the like) and slow large-range interference (bin surface compactness and gradual change of friction characteristics).
Disclosure of Invention
The invention aims to provide a hierarchical, self-learning and composite disturbance rejection trajectory tracking control algorithm for the advancing process of an unmanned rolling machine, aiming at the problem that the control method of the unmanned rolling machine in the prior art cannot comprise fast small-range disturbance and slow large-range disturbance.
The technical scheme adopted for realizing the purpose of the invention is as follows:
a composite disturbance rejection trajectory tracking control algorithm of an unmanned bucker comprises the following steps:
step 1, calculating an actual distance error in track tracking of a rolling machine according to an actual position of the rolling machine (measured by a positioning device assembled on the rolling machine) and a target track of the rolling machine;
step 2, calculating a target course angle of the roller mill by the anti-interference controller of the distance error according to the actual distance error obtained in the step 1 and the total disturbance of the distance error obtained by the instant disturbance observer, so that the track tracking distance error of the roller mill tends to zero;
step 3, calculating a target steering wheel corner of the roller mill by the anti-interference controller of the heading angle through the target heading angle obtained in the step 2, the total heading disturbance obtained by the instant disturbance observer and the heading angle disturbance obtained based on the steering system model parameter learner, so that the actual heading angle of the roller mill approaches to the target heading angle obtained in the step 2, wherein the steering system model parameter learner carries out online estimation on the steering system model parameters of the roller mill according to the real-time mapping relation of the target steering wheel corner and the front and rear body articulation angles;
and 4, sending the target steering wheel corner obtained in the step 3 to a steering wheel corner controller, and performing closed-loop control on the actual steering wheel corner to realize the track tracking of the rolling machine.
In the above technical solution, the disturbance rejection controller for the distance error in step 2 includes a distance error feedback controller and a distance error disturbance suppressor, the instant disturbance observer estimates the total disturbance of the actual distance error in real time by using the actual course angle and the actual distance error, and compensates in real time in the distance error disturbance suppressor, the distance error feedback controller processes the actual distance error, and then calculates the required target course angle of the roller mill by combining the information processed by the distance error disturbance suppressor, so that the track tracking distance error of the roller mill tends to zero.
In the above technical solution, the disturbance rejection controller of the course angle in step 3 includes a course angle feedback controller and a course angle disturbance suppressor, and the course angle disturbance suppressor includes a course angle disturbance suppressor based on a steering system model parameter learner and a course angle disturbance suppressor based on an instant disturbance observer;
the real-time disturbance observer estimates the total disturbance of the course angle in real time by using the actual course angle and the target steering wheel turning angle, and compensates in real time in a course angle disturbance suppressor;
the steering system model parameter learner in the step 4 performs online estimation on the course angle dynamic system model parameters of the roller by using a least square method and utilizing a real-time mapping relation between a target steering wheel corner and front and rear vehicle body hinge angles, and performs real-time compensation in a course angle disturbance suppressor;
and the course angle feedback controller calculates and obtains the target steering wheel rotation angle of the rolling machine according to the target course angle and by combining the information processed by the course angle disturbance suppressor.
In the above technical solution, the actual distance error e in the step 1d：
Figure GDA0003246319020000031
Wherein,
xaand yaEast and north coordinates, x, respectively, of originbAnd ybEast and north coordinates (in meters), x, of the respective end pointsvAnd yvEast and north coordinates (in meters) of the current position of the bucker, respectively;
in the above technical solution, the interference rejection controller of the distance error in step 2 is
Figure GDA0003246319020000032
The distance error feedback controller is
Figure GDA0003246319020000033
u0,outlpIs a virtual control amount in the distance error control, b0,outlpIs the gain of the control input (in radians/sec); wherein: u. of0,outlp＝kp,outlped，kp,outlpIs a proportional control coefficient, preferably kp,outlp＝ωc,out，ωc,outThe bandwidth (in radians/second) of the disturbance rejection controller for the distance error, the value of which is adjusted and determined according to the response speed requirement expected in the actual engineering application, edIs the actual distance error (in meters), b0,outlp＝vmeas，vmeasAs measured vehicle speed (in meters per second);
the distance error disturbance suppressor in the step 2 is
Figure GDA0003246319020000034
Wherein
Figure GDA0003246319020000035
The estimated value of the total disturbance of the distance error is obtained by an instant disturbance observer;
the instant disturbance observer is
Figure GDA0003246319020000036
Wherein
Figure GDA0003246319020000037
Wherein
Figure GDA0003246319020000038
And
Figure GDA0003246319020000039
respectively the actual distance error edThe estimated value of (2) and the estimated value of the total disturbance of the distance error are obtained by the instant disturbance observer through iteration, UoutlpFor control input, i.e. actual course angle, measured by the global positioning system GPS, Uoutlp＝θ1,GPS，f0,outlpIs edIs removed from the dynamic model
Figure GDA00032463190200000310
Part (C) is obtained from the equation of the dynamics of the roller compactor distance error (dynamic models can be derived by those skilled in the art and can be found in the literature Nayl T, Nikolakopoulos G, Gustfsson T]//2012 20th Mediterranean Conference on Control&Automation(MED).IEEE,2012:890-895)。
Figure GDA0003246319020000041
And
Figure GDA0003246319020000042
can be according to the formula
Figure GDA0003246319020000043
And (5) iteration is carried out.
Figure GDA0003246319020000044
β1,outlpAnd beta2,outlpIs the observer gain, preferably, beta1,outlp＝2ωo,outlp,β2,outlp＝ωo,outlp 2ωo,outlpThe bandwidth of the instant disturbance observer (unit is radian/second), and the value of the bandwidth is adjusted and determined according to the expected response speed requirement in the actual engineering application.
The target course angle in the step 2
Figure GDA0003246319020000045
Is a target course angle calculated from two points in the target track
Figure GDA0003246319020000046
The units are radians.
In the above technical solution, the anti-interference controller of the course angle in the step 3 is
Figure GDA0003246319020000047
uinlpIs the required target steering wheel angle, and the course angle feedback controller in the step 3 is
Figure GDA0003246319020000048
Wherein: u. of0,inlpIs a virtual control quantity of the course angle control loop,
Figure GDA0003246319020000049
preferably, k isp,inlp＝ωc,inlp，ωc,inlpBandwidth of the disturbance rejection controller which is a course angle, and the value of the bandwidth is adjusted and determined according to the expected response speed requirement in the actual engineering application, b0,inlpIs the gain of the course angle control input;
the heading angle disturbance suppressor based on the steering system model parameter learner is
Figure GDA00032463190200000410
In which the known disturbance
Figure GDA00032463190200000411
The units are in radians per second,
Figure GDA00032463190200000412
v is the ideal vehicle speed (in meters per second), l1And l2Respectively the length of the front and rear bodies (unit: meter), tausteerThe time constant (unit is second) of the dynamic process of the steering system, gamma is the front and back body articulation angle (unit is radian), a is the static gain coefficient of the steering system, and b is the static intercept coefficient of the steering system;
the course angle disturbance suppressor based on the instant disturbance observer is
Figure GDA00032463190200000413
Wherein
Figure GDA00032463190200000414
Obtaining f for an instantaneous disturbance observerinlpAn estimated value of (d);
the instant disturbance observer is
Figure GDA0003246319020000051
Wherein:
Figure GDA0003246319020000052
u is a control input in the course control loop, i.e. the target steering wheel angle
Figure GDA0003246319020000053
Can be obtained by the feedback of the steering wheel angle control system,
Figure GDA0003246319020000054
is a predicted value (unit: radians) of the actual heading angle of the front body,
Figure GDA0003246319020000055
and
Figure GDA0003246319020000056
all can pass through formula
Figure GDA0003246319020000057
The result of the iteration is that,
Figure GDA0003246319020000058
β1and beta2Is the instant disturbance observer gain, preferably beta1＝2ωo,β2＝ωo 2，ωoThe bandwidth of the instant disturbance observer (unit is radian/second), and the value of the bandwidth is adjusted and determined according to the expected response speed requirement in the actual engineering application.
In the above technical solution, the steering system model parameter learner in step 4 adopts a recursive least square method with a forgetting factor λ to estimate the model parameter
Figure GDA0003246319020000059
Real-time calculations were performed as follows:
Figure GDA00032463190200000510
wherein i is the number of the calculation steps,
Figure GDA00032463190200000511
i.e. an estimated value of Θ, k (i) is an intermediate variable in the steering system model parameter learner, p (i) is an intermediate variable in the steering system model parameter learner, Y ═ γ (i-1),
Figure GDA00032463190200000512
wherein gamma is the hinge angle (unit: radian) of the front and rear vehicle bodies,
Figure GDA00032463190200000513
to target steering wheel angle (in radians),
Figure GDA00032463190200000514
wherein
Figure GDA00032463190200000515
Are each tausteerEstimate of a, b (tau)steerThe unit of (a) is seconds, and a and b are dimensionless). Order:
Figure GDA00032463190200000516
and 3, self-learning of the model parameters of the steering system is realized.
Compared with the prior art, the invention has the beneficial effects that:
1. the invention adopts two control loops of course control (inner loop control) and distance error control (outer loop control) which are connected in series, and keeps the distance difference between the vehicle and another target track near zero by adjusting the turning angle of a steering wheel. For the inner ring controller and the outer ring controller, a composite disturbance rejection method combining three methods of disturbance feedforward disturbance rejection, active disturbance observation disturbance rejection based on an ESO (total disturbance immediate observer) and cumulative disturbance rejection based on model parameter online learning is adopted, and uncertainty of a control process is restrained.
2. The deviation of the vehicle course and distance error prediction model is equivalent to total disturbance, and a real-time disturbance observer is adopted for real-time observation and compensation, so that the robustness of the controller on the deviation of the prediction model can be improved, and the anti-interference capability is improved.
3. The method has the advantages that the information of the control process is utilized to carry out online learning and estimation on the key parameters of the model, the accuracy of prediction model and feedforward control can be continuously improved, the dependency on a disturbance observer is reduced, the continuous improvement of the control effect is realized, and the method is suitable for the problems of much interference and the change of characteristics of the rolling machine and the bin surface along with the time in the operation process of the rolling machine.
Drawings
Fig. 1 shows the overall architecture of the control algorithm of the present invention.
Detailed Description
The present invention will be described in further detail with reference to specific examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.
The invention discloses a hierarchical, self-learning and composite anti-interference track tracking control algorithm for the advancing process of an unmanned rolling machine, which comprises the following steps: 1) designing a composite disturbance rejection controller of the course angle of the rolling machine by taking the steering wheel corner as control input; 2) designing a learner of hydraulic steering system model parameters based on the real-time running information of the rolling machine; 3) combining the course angle composite disturbance rejection controller with a steering system model parameter learner to form a course angle self-learning composite disturbance rejection controller; 4) designing an active disturbance rejection controller for controlling and inputting a distance error by taking a target course angle; 5) the self-learning composite disturbance rejection controller of the course angle and the active disturbance rejection controller of the distance error are connected in series to form the track tracking control algorithm. The algorithm adopts a layered control architecture, combines instant disturbance observation and accumulated parameter learning, and is beneficial to solving the problems of much interference and time variation of characteristics of the rolling machine and the bin surface in the operation process of the rolling machine.
Example 1
A hierarchical, hierarchical and self-learning composite anti-interference track tracking control algorithm for the advancing process of an unmanned rolling machine comprises the following steps:
step 1, calculating an actual distance error in track tracking of a rolling machine according to an actual position of the rolling machine (measured by a positioning device assembled on the rolling machine) and a target track of the rolling machine;
step 2, calculating a target course angle of the roller mill by the anti-interference controller of the distance error according to the actual distance error obtained in the step 1 and the total disturbance of the distance error obtained by the instant disturbance observer, so that the track tracking distance error of the roller mill tends to zero;
step 3, calculating a target steering wheel corner of the roller mill by the anti-interference controller of the heading angle through the target heading angle obtained in the step 2, the total heading disturbance obtained by the instant disturbance observer and the heading angle disturbance obtained based on the steering system model parameter learner, so that the actual heading angle of the roller mill approaches to the target heading angle obtained in the step 2, wherein the steering system model parameter learner carries out online estimation on the steering system model parameters of the roller mill according to the real-time mapping relation of the target steering wheel corner and the front and rear body articulation angles;
and 4, sending the target steering wheel corner obtained in the step 3 to a steering wheel corner controller, and performing closed-loop control on the actual steering wheel corner to realize the track tracking of the rolling machine.
Example 2
Preferably, the disturbance rejection controller of the distance error in the step 2 is used for judging the actual heading angle (theta) of the rolling machine1,GPS) Error from actual distance (e)d) The complex dynamic relationship between the distance and the distance error is simplified into a distance error dynamic system model, and the deviation of the distance error dynamic system model and the actual process is uniformly regarded as the total disturbance (f) of the distance erroroutlp) A 1 to foutlpThe dynamic system model is considered as an extended state in the distance error dynamic system model, that is, the dynamic system model is extended to a dynamic system one order higher than the original dynamic system model. Adopting an instant disturbance observer, and utilizing the actual course angle and the actual distance error information to carry out total disturbance (f) on the distance erroroutlp) And carrying out real-time estimation and real-time compensation in a distance error disturbance suppressor. Aiming at the system processed by the distance error disturbance suppressor, a distance error feedback controller is designed, and the required rolling machine target course angle is calculated
Figure GDA0003246319020000071
So that the track of the roller follows the distance error (e)d) Tending to zero. This closed-loop controller is called the distance error controller (instantaneous disturbance observer + immunity controller of the distance error).
The disturbance rejection controller of the course angle in the step 3 is used for judging the actual course angle (theta) of the rolling machine1,GPS) Angle of rotation with target steering wheel
Figure GDA0003246319020000072
The complex dynamic relationship between the two is also simplified into a course angle dynamic system model, and the deviation of the course angle dynamic system model and the actual process is uniformly regarded as course total disturbance (f)inlp). Will f isinlpThe dynamic system model is considered as an expanded state in the course angle dynamic system model, that is, the dynamic system model is expanded to a dynamic system one step higher than the original dynamic system model. Using an instantaneous disturbance observer, using the actual course angle (theta)1,GPS) And target steering wheel angle
Figure GDA0003246319020000073
Information, total disturbance to course angle (f)inlp) And carrying out real-time estimation and real-time compensation in the course angle disturbance suppressor. Designing a course angle error feedback controller aiming at a system processed by a course angle disturbance suppressor, and calculating the required target steering wheel rotation angle of the rolling machine
Figure GDA0003246319020000074
So that the actual course angle (theta) of the roller1,GPS) Approaching to the target course angle
Figure GDA0003246319020000075
This closed-loop controller is called heading angle controller (instantaneous disturbance observer + heading angle disturbance rejection controller).
The steering system model parameter learner in the step 4 utilizes the target steering wheel corner to improve the adaptability of the control algorithm to the characteristic change of the rolling machine steering system and the bin surface friction and slip characteristics
Figure GDA0003246319020000076
And (3) carrying out on-line estimation on the course angle dynamic system model parameter (theta) of the rolling machine by adopting a least square method according to the real-time mapping relation of the front and rear body articulation angles (gamma). And applying the result of the online estimation of the model parameters to the step 3) to realize the continuous improvement of the control effect.
Example 3
As the basis of the development of a control algorithm, a prediction model of the included angle, the course angle and the position of the front and the rear vehicle bodies of the rolling machine is established. This section, although not part of the present invention, is still written herein for completeness. In the present embodiment, the following simplified heading angle dynamic system model is employed, but is not limited to this model. The predictive model of the influence of the steering wheel of the roller on the front and rear body articulation angles can be represented by (3).
Figure GDA0003246319020000081
Wherein gamma is the hinge angle (unit is radian) of the front and the rear vehicle bodies,
Figure GDA0003246319020000082
the derivative of the front and rear body articulation angle with respect to time,
Figure GDA0003246319020000083
is the target steering angle (in radians) of the steering wheel, tausteerIs the time constant (unit: second) of the dynamic process of the steering system, a is the static gain coefficient of the steering system, and b is the static intercept coefficient of the steering system. Equation (3) can also be modeled according to detailed physical principles, and will not be discussed in this embodiment. The course angle dynamic system model can be simply organized into the form of formula (4):
Figure GDA0003246319020000084
wherein, theta1Is the front vehicle body course angle (unit: radian),
Figure GDA0003246319020000085
is the derivative of the heading angle of the leading body with respect to time, v is the speed of the leading body (in meters per second), l1And l2The length of the front and rear bodies (unit: meter) respectively.
The dynamics of the east (x) and north (y) coordinates of the vehicle can be expressed as:
Figure GDA0003246319020000086
Figure GDA0003246319020000087
wherein
Figure GDA0003246319020000088
Is the derivative of the east coordinate x with respect to time, wherein
Figure GDA0003246319020000089
Is the derivative of the northbound coordinate y with respect to time.
(6) Middle theta10 in the positive east direction, clockwise rotation is negative. To translate this into the angle definition common to GPS feedback (zero north and positive clockwise rotation), the heading angle (θ) of the GPS measurement is defined1,GPS) Comprises the following steps:
Figure GDA00032463190200000810
the specific steps of the controller design are as follows:
the method comprises the following steps: the actual distance error of the track tracking of the roller is calculated with reference to the target track of the roller according to the measured actual position (x, y) of the roller of the global positioning system, i.e. GPS, equipped on the roller.
The calculation of the actual distance error can be obtained by (8):
Figure GDA0003246319020000091
wherein e isdIs the actual distance error (in meters), xaAnd yaEast and north coordinates (in meters), x, respectively, of originbAnd ybEast and north coordinates (in meters), x, of the respective end pointsvAnd yvEast and north coordinates (in meters) of the current location of the bucker, respectively.
Step two: and designing an anti-interference controller for the distance error, and calculating a required target course angle of the rolling machine to enable the track tracking distance error of the rolling machine to tend to be zero.
Vertical distance between front press roller of rolling machine and target track line
Figure GDA0003246319020000092
Satisfies the formula:
Figure GDA0003246319020000093
wherein, WoutlpIs an external disturbance of the control loop,
Figure GDA0003246319020000094
is the target heading angle (in radians) calculated from two points in the target trajectory, which can be calculated according to equation (10).
Figure GDA0003246319020000095
Wherein x isaAnd yaEast and north coordinates, x, respectively, of originbAnd ybEast and north coordinates of the respective end points. For formula (9), define
Figure GDA0003246319020000096
For the virtual control amount, (9) can be converted into:
Figure GDA0003246319020000097
wherein,
Figure GDA0003246319020000098
is an intermediate control quantity, foutlp＝(v-vmeas)uoutlp+WoutlpIs the total perturbation of the distance error (in meters per second), b0,outlp＝vmeasIs the gain of the control input. f. ofoutlpInvolving errors due to measurement of the speed of rotation (v-v)meas) And external interference WoutlpTotal perturbation caused, wherein v and vmeasRespectively, the ideal vehicle speed and the measured vehicle speed.
Suppose that f can be obtained by an observation algorithmoutlpIs estimated value of
Figure GDA0003246319020000099
Then the following distance error immunity controller can be designed for equation (11):
Figure GDA00032463190200000910
wherein,
Figure GDA00032463190200000911
feedback controller for distance error, b0,outlpIs the gain of the control input, u0,outlpIs a virtual control amount in the distance error control,
Figure GDA0003246319020000101
for a distance error disturbance suppressor based on an instantaneous disturbance observer,
Figure GDA0003246319020000102
the estimated value of the total distance error disturbance can be obtained by an instant disturbance observer. Wherein u is0,outlpCan be designed into the simplest proportional controller form:
u0,outlp＝kp,outlped (13)
wherein u is0,outlpIs a virtual control quantity, k, in the distance error controlp,outlpIs the proportional control coefficient to be set.
Bringing (13) into (12) gives:
Figure GDA0003246319020000103
by adjusting omegac,out＝kp,outlpThe dynamic characteristics of the closed loop system can be flexibly changed. Wherein ω isc,outIs the bandwidth (in radians/sec) of the range error control loop (immunity controller for range error), according to uoutlpThe required target course angle
Figure GDA0003246319020000104
The solution can be:
Figure GDA0003246319020000105
to realize
Figure GDA0003246319020000106
Writing (11) to the expanded state form:
Figure GDA0003246319020000107
wherein,
Figure GDA00032463190200001012
is foutlpDerivative with respect to time, houtlpIs unknown. The state space form can be obtained by arranging the formula (11):
Figure GDA0003246319020000109
wherein,
Figure GDA00032463190200001010
Uoutlpfor control input, i.e. the heading angle of the vehicle, measured by the global positioning system GPS, f0,outlpIs edRemoving AX from the dynamic model of (2)outlp+BUoutlpPart (c) is obtained from the dynamic equation of the distance error of the roller (dynamic models can be derived by those skilled in the art). Arranging (11) into an instant disturbance observer form:
Figure GDA00032463190200001011
wherein,
Figure GDA00032463190200001116
is composed of
Figure GDA0003246319020000112
The derivative with respect to time is that of,
Figure GDA0003246319020000113
and
Figure GDA0003246319020000114
respectively, a distance error edAnd the estimated value of the total disturbance of the distance error can be calculated by the formula
Figure GDA0003246319020000115
The result of the iteration is that,
Figure GDA0003246319020000116
β1,outlpand beta2,outlpIs the observer gain. By adjusting LoutlpCan change A-LoutlpC, and further adjusting the transient response speed of the observer. Wherein, the characteristic values are uniformly configured on the bandwidth (omega) of the instant disturbance observero,outlpThe unit is: radian/second) is a simple and practical parameter setting mode, corresponding LoutlpMatrix result is beta1,outlp＝2ωo,outlp,β2,outlp＝ωo,outlp 2。
Step three: and designing an anti-interference controller of the heading angle, and enabling the actual heading angle of the rolling machine to approach to the target heading angle by calculating the required target steering wheel angle of the rolling machine.
And considering a course angle dynamic system model of the rolling machine for calculating the obtained target course angle of the rolling machine in an interference rejection controller for tracking the distance error. For this purpose, bringing (4) into (3) makes available:
Figure GDA0003246319020000117
Figure GDA0003246319020000118
is the derivative of the heading angle of the GPS measurement over time. The method (11) can be further simplified,
Figure GDA0003246319020000119
wherein,
Figure GDA00032463190200001110
is a known disturbance that is a function of,
Figure GDA00032463190200001111
control input gain in course angle control of (1), finlpIs the unknown total disturbance in course angle control (in radians/sec).
It is assumed that an estimate of f can be obtained by an observer
Figure GDA00032463190200001112
Then the following course angle disturbance rejection controller can be designed:
Figure GDA00032463190200001113
wherein u isinlpIs the desired target steering wheel angle (in radians),
Figure GDA00032463190200001114
is a course angle feedback controller, u0,inlpIs a virtual control quantity of a course angle control loop, b0,inlpIs the gain of the heading angle control input,
Figure GDA00032463190200001115
for a heading angle disturbance suppressor based on a predictive model (steering system model parameter learner),
Figure GDA0003246319020000121
for the course angle disturbance suppression based on the instant disturbance observerThe manufacture of the device is carried out,
Figure GDA0003246319020000122
is called an instant disturbance observer. Wherein u is0,inlpThe control law of (2) can be designed into the simplest form of a proportional controller:
Figure GDA0003246319020000123
by adjusting omegac,inlp＝kp,inlpThe dynamic characteristics of a closed loop system can be flexibly changed, wherein omegac,inlpIs the bandwidth (in radians/second) of the disturbance rejection controller for the heading angle. With respect to the instant disturbance observer of the heading angle, one can write (20) as an extended state form:
Figure GDA0003246319020000124
the state space form can be obtained by arranging the formula (23):
Figure GDA0003246319020000125
wherein,
Figure GDA0003246319020000126
is the derivative of X with respect to time, U ═ thetasteerH is finlpThe derivative with respect to time is unknown. Designing an ESO observer
Figure GDA0003246319020000127
Wherein,
Figure GDA0003246319020000128
is the derivative of the estimated value of X with respect to time,
Figure GDA0003246319020000129
β1and beta2Is the ESO observer gain. By adjusting the value of L, the characteristic value of A-LC can be changed, and the performance of the observer can be adjusted. Wherein, the characteristic values are uniformly configured on the bandwidth (omega) of the instant disturbance observero) The method is a simple and practical parameter setting mode. The setting result of the corresponding L matrix is beta1＝2ωo,β2＝ωo 2。
Step four: designing a steering system model parameter learner, carrying out online estimation on the course angle dynamic system model parameters of the rolling machine according to the real-time mapping relation of the target turning angle of the steering wheel, the actual turning angle of the steering wheel, the hinge angles of the front and rear vehicle bodies and the actual course angle, and applying the online estimation to the step 3) to realize continuous improvement of the control effect.
In the models of equations (3) to (4), the time constant, the static gain coefficient, and the static intercept coefficient (τ) of the steering systemsteerA great deal of uncertainty exists with a, b). For example, mill tonnage, bin face friction resistance, and steering system drive torque variations can affect the steering system time constant, and hydraulic and leakage characteristics of a hydraulic steering system can affect the static gain and intercept from steering wheel angle to articulation angle. In the present embodiment, an online estimation method of parameters is described by taking these three parameters as examples. Specifically, converting (3) into a discrete form:
Figure GDA0003246319020000131
where i is the number of discrete sample points and also the number of steps to be counted. The classification (26) can be sorted out as follows:
Figure GDA0003246319020000132
working up (27) into linear form gives:
Y＝ΦT·Θ (28)
wherein,Y＝γ(i-1)，
Figure GDA0003246319020000133
the superscript T denotes transposing the matrix. The square sum of the deviation of the estimated value and the measured value of the included angle of the front and the rear vehicle bodies by the model,
Figure GDA0003246319020000134
minimum as target, using recursion least square method with forgetting factor lambda to estimate the model parameter
Figure GDA0003246319020000135
And carrying out real-time calculation.
Figure GDA0003246319020000136
Wherein i is a number for calculating the number of steps,
Figure GDA0003246319020000137
i.e., an estimated value of Θ, Y ═ γ (i-1),
Figure GDA0003246319020000138
wherein gamma is a hinge angle of the front and rear vehicle bodies,
Figure GDA0003246319020000139
in order to target the steering wheel angle,
Figure GDA00032463190200001310
wherein
Figure GDA00032463190200001311
Are each tausteerA, b. Order:
Figure GDA00032463190200001312
the self-learning of the parameters in the step 3) can be realized.
Step five: and sending the target steering wheel rotation angle of the rolling machine to a steering wheel rotation angle controller, and carrying out closed-loop control on the actual rotation angle of the steering wheel to realize the track tracking of the rolling machine.
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, many modifications and adaptations can be made without departing from the principle of the present invention, and such modifications and adaptations should also be considered as the scope of the present invention.

Claims (8)

1. A composite anti-interference track tracking control algorithm for the advancing process of an unmanned rolling machine is characterized in that: the method comprises the following steps:
step 1, calculating an actual distance error in track tracking of a rolling machine according to an actual position of the rolling machine and a target track of the rolling machine;
step 2, calculating a target course angle of the rolling machine by the anti-interference controller of the distance error through the actual distance error obtained in the step 1 and the total disturbance of the distance error obtained by the instant disturbance observer, so that the track tracking distance error of the rolling machine tends to zero;
the disturbance rejection controller of the distance error in the step 2 comprises a distance error feedback controller and a distance error disturbance suppressor, the instant disturbance observer estimates the total disturbance of the actual distance error in real time by using the actual course angle and the actual distance error, and compensates in real time in the distance error disturbance suppressor, the distance error feedback controller processes the actual distance error, and then calculates the required rolling machine target course angle by combining the information processed by the distance error disturbance suppressor, so that the track tracking distance error of the rolling machine tends to zero;
step 3, calculating a target steering wheel corner of the roller mill by the anti-interference controller of the heading angle through the target heading angle obtained in the step 2, the total heading disturbance obtained by the instant disturbance observer and the heading angle disturbance obtained based on the steering system model parameter learner, so that the actual heading angle of the roller mill approaches to the target heading angle obtained in the step 2, wherein the steering system model parameter learner carries out online estimation on the steering system model parameters of the roller mill according to the real-time mapping relation of the target steering wheel corner and the front and rear body articulation angles;
the disturbance rejection controller of the course angle in the step 3 comprises a course angle feedback controller and a course angle disturbance suppressor, and the course angle disturbance suppressor comprises a course angle disturbance suppressor based on a steering system model parameter learner and a course angle disturbance suppressor based on an instant disturbance observer;
the real-time disturbance observer estimates the total disturbance of the course angle in real time by using the actual course angle and the target steering wheel turning angle, and compensates in real time in a course angle disturbance suppressor;
the steering system model parameter learner in the step 3 utilizes a real-time mapping relation between a target steering wheel corner and front and rear vehicle body hinge angles, adopts a least square method to carry out on-line estimation on course angle dynamic system model parameters of the rolling machine, and carries out real-time compensation in a course angle disturbance suppressor;
the course angle feedback controller calculates and obtains a target steering wheel turning angle of the rolling machine according to the target course angle and by combining the information processed by the course angle disturbance suppressor;
and 4, sending the target steering wheel corner obtained in the step 3 to a steering wheel corner controller, and performing closed-loop control on the actual steering wheel corner to realize the track tracking of the rolling machine.
2. The composite disturbance rejection trajectory tracking control algorithm for the progress of the unmanned roller compactor according to claim 1, wherein: the actual distance error e in step 1d：
Figure FDA0003577202070000021
Wherein,
xaand yaEast and north coordinates, x, respectively, of originbAnd ybRespectively end point ofEast and north coordinates, xvAnd yvEast and north coordinates of the current location of the bucker, respectively.
3. The composite disturbance rejection trajectory tracking control algorithm for the progress of the unmanned roller compactor according to claim 1, wherein: the interference rejection controller of the distance error in the step 2 is
Figure FDA0003577202070000022
The distance error feedback controller is
Figure FDA0003577202070000023
u0,outlpIs a virtual control amount in the distance error control, b0,outlpIs the gain of the control input; wherein: u. of0,outlp＝kp,outlped，kp,outlpIs a proportional control coefficient, edFor actual distance error, b0,outlp＝vmeas，vmeasIs the measured vehicle speed;
the distance error disturbance suppressor in the step 2 is
Figure FDA0003577202070000024
Wherein
Figure FDA0003577202070000025
The estimated value of the total disturbance of the distance error is obtained by an instant disturbance observer;
the instant disturbance observer is
Figure FDA00035772020700000217
Wherein
Figure FDA0003577202070000027
Wherein
Figure FDA0003577202070000028
And
Figure FDA0003577202070000029
respectively the actual distance error edThe estimated value of (2) and the estimated value of the total disturbance of the distance error are obtained by the instant disturbance observer through iteration, UoutlpFor control input, i.e. actual course angle, measured by the global positioning system GPS, Uoutlp＝θ1,GPS，f0,outlpIs edIn the dynamic model of (1)
Figure FDA00035772020700000210
The part of (2) is obtained by a dynamic equation of the distance error of the rolling machine,
Figure FDA00035772020700000211
and
Figure FDA00035772020700000212
can be according to the formula
Figure FDA00035772020700000213
The result of the iteration is that,
Figure FDA00035772020700000214
β1,outlpand beta2,outlpIs the observer gain.
4. The composite disturbance rejection trajectory tracking control algorithm for the unmanned bucker advancement process according to claim 3, wherein: k is a radical ofp,outlp＝ωc,out，ωc,outBandwidth of the disturbance rejection controller for distance errors, beta1,outlp＝2ωo,outlp,β2,outlp＝ωo,outlp 2，ωo,outlpThe observer bandwidth is perturbed instantaneously.
5. The composite disturbance rejection trajectory tracking control algorithm for the unmanned bucker advancement process according to claim 3, wherein: the target course angle in the step 2
Figure FDA00035772020700000215
Figure FDA00035772020700000216
Is a target course angle calculated from two points in the target track
Figure FDA0003577202070000031
6. The composite disturbance rejection trajectory tracking control algorithm for the progress of the unmanned roller compactor according to claim 1, wherein: the anti-interference controller of the course angle in the step 3 is
Figure FDA0003577202070000032
uinlpIs the required target steering wheel angle, and the course angle feedback controller in the step 3 is
Figure FDA0003577202070000033
Wherein: u. of0,inlpIs a virtual control quantity of the course angle control loop,
Figure FDA00035772020700000317
b0,inlpis the gain of the course angle control input;
the heading angle disturbance suppressor based on the steering system model parameter learner is
Figure FDA0003577202070000034
In which the known disturbance
Figure FDA0003577202070000035
v is the ideal vehicle speed, l1And l2Vehicle length, tau, of front and rear bodies, respectivelysteerIs the time constant of the dynamic process of the steering system, gamma is the front and rear body articulation angle, and a is the steering systemB is the static intercept coefficient of the steering system;
the course angle disturbance suppressor based on the instant disturbance observer is
Figure FDA0003577202070000036
Wherein
Figure FDA0003577202070000037
Obtaining f for an instantaneous disturbance observerinlpAn estimated value of (d);
the instant disturbance observer is
Figure FDA0003577202070000038
Wherein:
Figure FDA0003577202070000039
u is a control input in the course control loop, i.e. the target steering wheel angle
Figure FDA00035772020700000310
Can be obtained by the feedback of the steering wheel angle control system,
Figure FDA00035772020700000311
for a predicted value of the actual heading angle of the front body,
Figure FDA00035772020700000312
and
Figure FDA00035772020700000313
all can pass through formula
Figure FDA00035772020700000314
The result of the iteration is that,
Figure FDA00035772020700000315
β1and beta2Is the instant disturbance observer gain.
7. The composite disturbance rejection trajectory tracking control algorithm for the unmanned bucker advancement process according to claim 6, wherein: k is a radical ofp,inlp＝ωc,inlp，ωc,inlpIs the bandwidth of the immunity controller for the heading angle,
Figure FDA00035772020700000316
ωothe observer bandwidth is perturbed instantaneously.
8. The composite disturbance rejection trajectory tracking control algorithm for the unmanned bucker advancement process according to claim 7, wherein: the steering system model parameter learner in the step 3 adopts a recursive least square method with a forgetting factor lambda to estimate the model parameters
Figure FDA0003577202070000041
Real-time calculations were performed as follows:
Figure FDA0003577202070000042
wherein i is a number for calculating the number of steps,
Figure FDA0003577202070000043
i.e. an estimated value of Θ, k (i) is an intermediate variable in the steering system model parameter learner, p (i) is an intermediate variable in the steering system model parameter learner, Y ═ γ (i-1),
Figure FDA0003577202070000044
wherein gamma is a hinge angle of the front and rear vehicle bodies,
Figure FDA0003577202070000045
in order to target the steering wheel angle,
Figure FDA0003577202070000046
wherein
Figure FDA0003577202070000047
Are each tausteerThe estimated values of a, b, let:
Figure FDA0003577202070000048
and 3, self-learning of the model parameters of the steering system is realized.
CN201910740160.2A 2019-08-12 2019-08-12 Composite anti-interference track tracking control algorithm for advancing process of unmanned rolling machine Active CN112394740B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910740160.2A CN112394740B (en) 2019-08-12 2019-08-12 Composite anti-interference track tracking control algorithm for advancing process of unmanned rolling machine

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910740160.2A CN112394740B (en) 2019-08-12 2019-08-12 Composite anti-interference track tracking control algorithm for advancing process of unmanned rolling machine

Publications (2)

Publication Number Publication Date
CN112394740A CN112394740A (en) 2021-02-23
CN112394740B true CN112394740B (en) 2022-06-17

Family

ID=74602206

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910740160.2A Active CN112394740B (en) 2019-08-12 2019-08-12 Composite anti-interference track tracking control algorithm for advancing process of unmanned rolling machine

Country Status (1)

Country Link
CN (1) CN112394740B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113682296A (en) * 2021-08-25 2021-11-23 中汽创智科技有限公司 Vehicle transverse control method and device based on cascade PID and vehicle
CN115309058B (en) * 2022-06-10 2023-06-30 哈尔滨理工大学 Finite time composite learning control method for power positioning ship

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103076804A (en) * 2012-12-27 2013-05-01 深圳先进技术研究院 Automatic guide vehicle, automatic guide system and navigation method based on ultrasonic range finder
CN104502991A (en) * 2014-12-27 2015-04-08 芜湖赛特施工设备有限公司 Track tracking detection device for construction site
CN105137997A (en) * 2015-09-22 2015-12-09 清华大学 Automatic driving system and method of water conservancy construction vibration roller
CN105735093A (en) * 2014-12-12 2016-07-06 宝马格有限公司 Method for controlling work train
WO2018138320A1 (en) * 2017-01-27 2018-08-02 Wheel.Me As A rolling arrangement and method for autonomously repositioning devices with integrated rolling arrangement
CN108897311A (en) * 2018-06-14 2018-11-27 天津大学 A kind of unmanned system of building a dam of roller cluster
CN108914750A (en) * 2018-06-14 2018-11-30 天津大学 A kind of vibrating roller press Unmanned Systems
CN108951374A (en) * 2018-07-10 2018-12-07 北京踏歌艾尔机器人科技有限公司 Roll the control method and device of engineering truck
CN109799830A (en) * 2019-03-06 2019-05-24 山东交通学院 Impact road roller automatic driving control system and method

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9796421B1 (en) * 2016-04-07 2017-10-24 GM Global Technology Operations LLC Autonomous vehicle lateral control for path tracking and stability
US10372134B2 (en) * 2017-06-30 2019-08-06 Intel Corporation Methods and apparatus to implement nonlinear control of vehicles moved using multiple motors

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103076804A (en) * 2012-12-27 2013-05-01 深圳先进技术研究院 Automatic guide vehicle, automatic guide system and navigation method based on ultrasonic range finder
CN105735093A (en) * 2014-12-12 2016-07-06 宝马格有限公司 Method for controlling work train
CN104502991A (en) * 2014-12-27 2015-04-08 芜湖赛特施工设备有限公司 Track tracking detection device for construction site
CN105137997A (en) * 2015-09-22 2015-12-09 清华大学 Automatic driving system and method of water conservancy construction vibration roller
WO2018138320A1 (en) * 2017-01-27 2018-08-02 Wheel.Me As A rolling arrangement and method for autonomously repositioning devices with integrated rolling arrangement
CN108897311A (en) * 2018-06-14 2018-11-27 天津大学 A kind of unmanned system of building a dam of roller cluster
CN108914750A (en) * 2018-06-14 2018-11-30 天津大学 A kind of vibrating roller press Unmanned Systems
CN108951374A (en) * 2018-07-10 2018-12-07 北京踏歌艾尔机器人科技有限公司 Roll the control method and device of engineering truck
CN109799830A (en) * 2019-03-06 2019-05-24 山东交通学院 Impact road roller automatic driving control system and method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
Accurate trajectory tracking with disturbance-resistant and heading estimation method for self-driving vibratory roller;Xie Hui;《IFAC-PapersOnLine》;20181231;第51卷(第31期);第754-758页 *
Autonomous Automobile Trajectory Tracking for Off-Road Driving: Controller Design, Experimental Validation and Racing;Gabriel M. Hoffmann;《 2007 American Control Conference》;20070730;第2296-2301页 *
基于模糊算法的无人驾驶压路机自动碾压控制;卞永明;《同济大学学报( 自然科学版)》;20171231;第45卷(第12期);第1830-1838页 *

Also Published As

Publication number Publication date
CN112394740A (en) 2021-02-23

Similar Documents

Publication Publication Date Title
CN108614426B (en) Multi-mobile-robot formation robust control method based on disturbance observer
Xia et al. Lateral path tracking control of autonomous land vehicle based on ADRC and differential flatness
CN110780594B (en) Path tracking method and system of intelligent vehicle
CN108227491B (en) Intelligent vehicle track tracking control method based on sliding mode neural network
CN107415939B (en) Steering stability control method for distributed driving electric automobile
CN109857124B (en) Unmanned ship accurate path tracking control method based on speed and course dual guidance
Xia et al. Optimal robust trajectory tracking control of a X-rudder AUV with velocity sensor failures and uncertainties
Shi et al. Planning the trajectory of an autonomous wheel loader and tracking its trajectory via adaptive model predictive control
CN108919837B (en) Second-order sliding mode control method of automatic driving vehicle based on visual dynamics
Liang et al. Finite-time velocity-observed based adaptive output-feedback trajectory tracking formation control for underactuated unmanned underwater vehicles with prescribed transient performance
CN106292287A (en) A kind of UUV path following method based on adaptive sliding-mode observer
CN112394740B (en) Composite anti-interference track tracking control algorithm for advancing process of unmanned rolling machine
CN102880060A (en) Self-adaptive index time varying slip form posture control method of reentry flight vehicle
CN108958246A (en) Trajectory Tracking Control method of the unmanned vehicle in U-Turn scene
CN111703432A (en) Real-time estimation method for sliding parameters of intelligent tracked vehicle
CN113110048A (en) Nonlinear system output feedback adaptive control system and method adopting HOSM observer
Shi et al. Distributed circumnavigation control of autonomous underwater vehicles based on local information
Rigatos A nonlinear optimal control approach for tracked mobile robots
Ziye et al. Tracking control of unmanned tracked vehicle in off-road conditions with large curvature
Yang et al. Path tracking control of an articulated road roller with sideslip compensation
Lu et al. Adaptive heading control strategy for unmanned ground vehicle with variable wheelbase based on robust-active disturbance rejection control
Chen et al. Neural network compensator-based robust iterative learning control scheme for mobile robots nonlinear systems with disturbances and uncertain parameters
CN112987555B (en) Composite anti-interference track tracking control algorithm for retreating process of unmanned rolling machine
CN112965498A (en) Intelligent sweeping machine running track tracking control method
Xie et al. Parameter self‐learning feedforward compensation‐based active disturbance rejection for path‐following control of self‐driving forklift trucks

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