CN115373287A - Adaptive parameter model prediction path tracking control method for articulated tractor - Google Patents

Adaptive parameter model prediction path tracking control method for articulated tractor Download PDF

Info

Publication number
CN115373287A
CN115373287A CN202210961404.1A CN202210961404A CN115373287A CN 115373287 A CN115373287 A CN 115373287A CN 202210961404 A CN202210961404 A CN 202210961404A CN 115373287 A CN115373287 A CN 115373287A
Authority
CN
China
Prior art keywords
model
articulated
tractor
vehicle
tracking
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
Application number
CN202210961404.1A
Other languages
Chinese (zh)
Inventor
张庆
周保成
赵建柱
尤泳
王德成
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.)
China Agricultural University
Original Assignee
China Agricultural 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 China Agricultural University filed Critical China Agricultural University
Priority to CN202210961404.1A priority Critical patent/CN115373287A/en
Publication of CN115373287A publication Critical patent/CN115373287A/en
Pending legal-status Critical Current

Links

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
    • G05B17/00—Systems involving the use of models or simulators of said systems
    • G05B17/02—Systems involving the use of models or simulators of said systems electric
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention relates to a tracking control method for a self-adaptive parameter model prediction path of a articulated tractor, which comprises the following steps: the method comprises the following steps: according to the structure of the articulated steering semi-track tractor, a multi-body dynamic model of the articulated steering semi-track tractor is built in the RecurDyn; step two: constructing a multi-body dynamic model of the articulated steering semi-track tractor in the step one, and performing front wheel steering and articulated steering kinematic analysis to obtain a kinematic model; step three: designing to obtain a path tracking MPC controller according to the kinematics model obtained in the step two and by combining model prediction control, and obtaining input and output of the controller; step four: performing self-adaptive real-time optimization on the time domain parameters of the path tracking controller designed in the third step by adopting a genetic algorithm; step five: creating an m file for the co-simulation; step six: constructing a joint simulation model in Simulink; step seven: and (3) carrying out simulation comparison on a traditional model prediction controller and the designed adaptive parameter model prediction, and verifying the effectiveness of the adaptive parameter model. The invention optimizes the time domain parameters of the traditional MPC through the genetic algorithm, so that the time domain parameters can realize self-adaptive optimization according to the information such as the vehicle position and the like in real time, the accuracy of path tracking is improved, and the environmental applicability of path tracking is increased.

Description

Adaptive parameter model prediction path tracking control method for articulated tractor
Technical Field
The invention relates to the technical field of automatic driving of agricultural vehicles, in particular to a tracking control method for a self-adaptive parameter model prediction path of a articulated steering tractor.
Background
With the rapid development of satellite navigation, information wireless transmission, sensors and control technologies in recent years, the technical research and demonstration application of autonomous navigation agricultural equipment are rapidly promoted. The agricultural machinery automatic navigation control technology has become an important technology for liberating productivity and realizing agricultural automation. The articulated steering semi-crawler tractor has the advantages that the articulated steering semi-crawler tractor front and rear tractor bodies can deflect relatively to realize steering, the articulated steering semi-crawler tractor has the characteristics of small and exquisite whole machine, compact structure, good trafficability, small turning radius and easiness in operation, has the advantages of high operating efficiency and good traveling performance when being used in an orchard and a small-spacing farmland, and is widely concerned. Therefore, the research on the automatic navigation of the articulated steering tractor matched with the front wheel steering has great significance for realizing intellectualization and no humanization of orchards in China. Agricultural machinery path tracking control is the key of the automatic navigation technology of agricultural equipment. The articulated vehicle has a complex structure and strong nonlinearity, and compared with a common vehicle, the articulated vehicle has more difficulty and more challenge in path tracking control. The currently commonly used path tracking control algorithms include a PID control algorithm, a pure tracking control algorithm, a linear feedback control algorithm, a Stanley control algorithm, a model predictive control algorithm (MPC), an optimal control algorithm and the like. The model predictive control has the advantages of predicting future tracks and processing multiple constraint conditions, and is widely applied to the aspect of vehicle path tracking control. However, the time domain parameters of the conventional MPC controller are fixed at different speeds and under different road conditions, which reduces the applicability of the MPC controller and makes it difficult to satisfy the tracking effects of different environments.
The method comprises the steps of designing a vehicle track tracking MPC controller based on a single-track vehicle transverse dynamics model, analyzing the influence of the vehicle track tracking MPC controller on the performance of the MPC controller by taking a transverse deviation weight coefficient as an example under the working condition of tracking a linear target track, and converting the determination problem of the MPC controller weight coefficient into a multi-objective optimization problem by taking the response time and the tracking deviation of an intelligent vehicle tracking target track as the minimum targets to obtain an optimized weight coefficient. The scheme has the defects of low dynamic adjustment precision, low environment adaptability, and low optimized calculation response speed and precision.
A path tracking control method of a variable-parameter intelligent networked automobile is based on a model prediction control principle and designs a path tracking controller of the intelligent networked automobile. The method comprises the steps of firstly, taking a vehicle model of a 3-freedom model as a control system; after the system is linearized, determining a quadratic objective function of the system, and determining a matrix form according to a function form; and then, performing off-line simulation on Carsim and Matlab/Simulink platforms, and determining simulation parameters suitable for the path tracking controller under each typical working condition. The method determines the input parameters of different road sections in advance, so that the vehicle controller selects different control parameters according to the characteristics of the input path and acts on a control system, and the method has low environmental suitability, poor system continuity and poor real-time performance.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a tracking control method for a self-adaptive parameter model prediction path of a articulated steering tractor, which solves the problems that the traditional MPC controller is difficult to meet the tracking effect of different environments due to fixed time domain parameters under different speeds and different road conditions, and the problem that Carsim software adopted in the current path tracking experiment simulation lacks a crawler model.
In order to achieve the above purposes, the technical scheme adopted by the invention is as follows:
a tracking control method for a self-adaptive parameter model prediction path of a articulated tractor comprises the following steps:
the method comprises the following steps: according to the structure of the articulated steering semi-track tractor, a multi-body dynamic model of the articulated steering semi-track tractor is built in the RecurDyn;
step two: constructing a multi-body dynamic model of the articulated steering semi-track tractor in the step one, and performing front wheel steering and articulated steering kinematic analysis to obtain a kinematic model;
step three: designing a path tracking MPC controller according to the kinematics model obtained in the step two and combining model prediction control to obtain input and output of the controller;
step four: adopting a genetic algorithm to carry out self-adaptive real-time optimization on the time domain parameters of the path tracking controller designed in the step three;
step five: creating an m file for the co-simulation;
step six: constructing a joint simulation model in Simulink;
step seven: and (3) carrying out simulation comparison on a traditional model prediction controller and the designed adaptive parameter model prediction, and verifying the effectiveness of the adaptive parameter model.
Further, the first step specifically comprises:
modeling a front wheel, a front axle, an engine, a transfer case, a universal joint, a bowing oil tank, a rear case body and a suspension leg, and converting the model into an stp file which can be imported by RecurDyn;
importing the stp model file into RecurDyn, and then carrying out parametric modeling on the crawler chassis by a Track module in the RecurDyn, wherein the crawler chassis part comprises: the rear tension wheel, the driving wheel, the loading wheel and the front tension wheel;
the constraint pairs among all parts in the model are arranged, and the number of the constraint pairs is 13, 21, 4 and 8, wherein the number of the fixed pairs is 21, the number of the moving pairs is 4, and the number of the ball pairs is 8.
Further, the front wheel steering kinematic model in the step two is as shown in the formula (4):
Figure BDA0003793339530000031
wherein v represents the central speed of the rear axle of the articulated tractor,
Figure BDA0003793339530000032
representing the component of the velocity v on the X-axis,
Figure BDA0003793339530000033
indicating the component of velocity v on the Y-axis, psi represents the dog-leg tractor heading angle,
Figure BDA0003793339530000034
indicating the change rate of the course angle, and delta indicating the rotation angle of the front wheel of the articulated steering tractor;
the bending steering kinematic model in the step two is as the formula (10):
Figure BDA0003793339530000035
in the formula, v 1 The speed of the center of the rear axle is shown,
Figure BDA0003793339530000036
representing the velocity v 1 The component on the X-axis is,
Figure BDA0003793339530000037
representing the velocity v 1 Component in the Y axis, θ 1 The azimuth angle of the rear vehicle body is indicated,
Figure BDA0003793339530000038
the derivative of the azimuth angle of the rear vehicle body with respect to time is represented,
Figure BDA0003793339530000039
the angle of the knuckle steer is shown,
Figure BDA00037933395300000310
representing the derivative of the articulated steering angle with respect to time, delta the maximum value of the front wheel angle, L 1 Indicating the distance of the hinge point from the rear axle, L 2 Indicating the distance from the hinge point to the front axle and omega indicating the rate of change of the articulated steering angle.
Further, the objective function for optimizing the system state quantity and the control quantity in step three is as shown in equation (11):
Figure BDA0003793339530000041
where xi (t) is a state space expression, U (t-1) is a system input, Δ U is a system control increment, Δ η (t + i | t) = η (t + i | t) - η r (t + i | t), which is the difference between the actual output and the reference output, Q, R and ρ are the weight matrices, and ε is the relaxation factor.
Further, the fitness function of the genetic algorithm in step four is as shown in formula (12):
Figure BDA0003793339530000042
the flow of the genetic algorithm in step four: initializing a population, then assigning and rounding each individual of the population according to a time domain parameter range, calculating the fitness corresponding to each individual through a fitness function by combining the speed and pose information of a vehicle, outputting an optimal parameter when a termination condition is met, and if the terminal condition is not met, selecting, crossing and mutating to obtain a new population, and performing iterative computation again.
Further, the parameter setting of the genetic algorithm in step four comprises: the population size is 200, the probability of cross and variation is 0.6 and 0.1, the evolution times are 20, the time domain value range is predicted (0,60), and the time domain value range is controlled (0,30).
Further, the fifth step specifically includes:
in a Communicator module of a RecurDyn software, plant _ in is used for creating system input, the input comprises the rotating speed of a driving wheel, the rotating angle of a front wheel and a bowing steering angle, plant _ out is used for creating system output, and the output comprises a front axle horizontal coordinate, a front axle vertical coordinate and a vehicle course angle;
after the I/O is created, the Simulink module in the Recurdyn software Communicator module generates the.m file for the co-simulation.
Further, the joint simulation model in the sixth step includes:
the system comprises a RecurDyn vehicle model module, a vehicle tracking module and a vehicle tracking module, wherein the RecurDyn vehicle model module outputs the current position and the current course angle of a vehicle;
the system comprises a nearest target point searching module, a tracking target point tracking module and a tracking target point tracking module, wherein the nearest target point searching module finds a nearest tracking target point according to the current position of a vehicle;
the MPC controller controls the vehicle corner according to the latest target and outputs the vehicle corner, the course error and the transverse error when tracking the last target point;
and the genetic algorithm module is used for solving optimal Np and Nc according to the output of the MPC controller and controlling the vehicle corner by the MPC controller when tracking the next target point.
Further, the path taken by the simulation verification in the seventh step includes: u-shaped, 8-shaped and multi-concave-convex curve.
The invention provides a new method for tracking and simulating a path of a articulated steering semi-track tractor by combining RecurDyn with MATLAB/Simulink. The time domain parameters of the traditional MPC are optimized through a genetic algorithm, so that the time domain parameters can be adaptively optimized in real time according to information such as vehicle positions and the like, the accuracy of path tracking is improved, and the environmental applicability of path tracking is increased.
Drawings
The invention has the following drawings:
FIG. 1 is a flow diagram of a technical solution;
FIG. 2 is a multi-body dynamic model diagram of a articulated steering semi-track tractor;
FIG. 3 is a front wheel steering kinematics diagram;
FIG. 4 is a graph of a bow-steer kinematics relationship;
FIG. 5 is a time domain parameter optimization schematic;
FIG. 6 is a diagram of a joint simulation model;
FIG. 7 is a schematic diagram of adaptive parametric model predictive controller path tracking control;
FIG. 8 is a diagram of the tracking effect of U font curve path;
FIG. 9 is a graph of a curve path tracking effect in font form;
FIG. 10 is a diagram illustrating a multi-concave-convex curve path tracking effect;
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
As shown in FIG. 1, the adaptive parametric model prediction path tracking control method for the articulated tractor comprises the following steps:
step 1: according to the structure of the articulated steering semi-track tractor, a multi-body dynamic model of the articulated steering semi-track tractor is built in the RecurDyn.
Step 2: and (4) constructing a multi-body dynamic model of the articulated steering semi-track type tractor in the step (1), and performing front wheel steering and articulated steering kinematic analysis to obtain a kinematic model.
And 3, step 3: and (3) designing to obtain a path tracking controller according to the kinematic model obtained in the step (2) and by combining model prediction control, and obtaining the input and the output of the controller.
And 4, step 4: and (3) performing self-adaptive real-time optimization on the time domain parameters of the path tracking controller designed in the step (3) by adopting a GA genetic algorithm.
And 5: an m file was created for a combined simulation of reccurdyn and MATLAB/Simulink.
Step 6: a joint simulation model was constructed in Simulink.
And 7: and (3) carrying out simulation comparison on the traditional MPC controller and the designed adaptive parameter model prediction, and verifying the effectiveness of the MPC controller.
Wherein:
the step 1 of constructing the multi-body dynamic model specifically comprises the following steps:
the multi-body dynamic model of the articulated steering semi-track tractor is shown in figure 2. In Catia software, front wheels 1, a front axle 2, an engine 3, a transfer case 4, a universal joint 5, a articulated oil tank 6, a rear case 7 and suspension legs 8 are modeled, and the model is converted into an stp file which can be imported by RecurDyn. Importing the stp model file into RecurDyn, and then carrying out parametric modeling on the crawler chassis by a RecurDyn/Track module, wherein the crawler chassis part comprises: rear tension wheel 9, driving wheel 10, bogie wheel 11 and front tension wheel 12. The constraint pairs among all parts in the model are arranged, and the number of the constraint pairs is 13, 21, 4 and 8, wherein the number of the fixed pairs is 21, the number of the moving pairs is 4, and the number of the ball pairs is 8.
The kinematic model obtained by analyzing in the step 2 is concretely as follows:
front wheel steering kinematics model: in fig. 3, the center coordinates (x, y) of the rear axle of the articulated tractor, the heading angle ψ of the articulated tractor, the front wheel rotation angle δ, and the velocity v. The components of the velocity v in the X, Y axial direction are:
Figure BDA0003793339530000061
Figure BDA0003793339530000062
the rate of change of course angle is:
Figure BDA0003793339530000071
the motion equation of the articulated steering semi-crawler tractor under the working condition of only adopting front wheel steering is as follows:
Figure BDA0003793339530000072
a bow steering kinematics model: in FIG. 4, O 1 Is the center of the rear axle, O 2 Is the center of the front axle, O is the articulated point of the articulated steering, L 1 Distance of the hinge point to the rear axle, L 2 Is the distance from the hinge point to the front axle, delta is the maximum value of the front wheel rotation angle, phi is the articulated steering angle, theta 1 Azimuth angle, θ, of rear vehicle body 2 Is the azimuth angle, v, of the front body 1 Is the rear axle center velocity, v 2 Is the front axle center speed.
Rear axle center O 1 The coordinate is (x) 1 ,y 1 ) The coordinate of the center O2 of the front axle is (x) 2 ,y 2 ). The kinematic constraints of the entire vehicle can be expressed as:
Figure BDA0003793339530000073
the relationship between the knuckle steering angle and the front and rear body azimuths is:
Figure BDA0003793339530000074
the kinematic relationship between the buckled steering angle and the front and rear body azimuth angles may be expressed as:
Figure BDA0003793339530000075
since the front and rear bodies vary in speed with respect to the rigid free joint of the vehicle, the relative velocity equation can be expressed as:
Figure BDA0003793339530000076
from equations (6) and (7), it can be derived that the rate of change of the rear body azimuth is:
Figure BDA0003793339530000077
in the proposed method, the variable is the rate of change of the knuckle steer angle, denoted by ω. The bow-turn kinematics model can be expressed as:
Figure BDA0003793339530000078
step 3, designing a path tracking controller as follows:
and (3) designing and adding an objective function capable of reflecting optimization of the system state quantity and the control quantity according to the kinematic model obtained in the step (2) in combination with model prediction control.
Figure BDA0003793339530000081
Where xi (t) is the state space expression, U (t-1) is the system input, Δ U is the system control increment, Δ η (t + i | -t) = η (t + i | -t) - η r (t + i | t), which is the difference between the actual output and the reference output, Q, R and ρ which are the weight matrices, and ε which is the relaxation factor.
Step 4, designing a time domain parameter adaptive optimization controller as follows:
the time domain parameter optimization principle is as shown in fig. 5, after a population is initialized, each individual of the population is assigned and rounded according to the time domain parameter range, the fitness corresponding to each individual is calculated through a fitness function by combining information such as the speed and the pose of a vehicle, the optimal parameter is output when the termination condition is met, and if the fitness does not meet the requirement, a new population is obtained by selection, crossing and variation, and iterative computation is performed again.
The population size of the genetic algorithm is set to be 200, the probability of crossover and mutation is set to be 0.6 and 0.1, and the evolution times are set to be 20. In order to improve the optimization efficiency, a time domain value range (0,60) is predicted, and the time domain value range (0,30) is controlled.
The fitness function is established as follows:
Figure BDA0003793339530000082
step 5, creating the joint simulation file specifically as follows:
in the Communicator module of the Recurdyn software, plant _ in is used to create system inputs, which are the speed of the drive wheels 10, the front wheel angle, and the articulated angle. Create a system output with Plant _ out, the output having: the X-axis coordinate (transverse coordinate) and the Z-axis coordinate (longitudinal coordinate) of the front axle 2, and the vehicle heading angle.
After the I/O is created, the Simulink module in the Recurdyn software Communicator module can generate the m file for the joint simulation.
Step 6, constructing a combined simulation model specifically as follows:
a joint simulation model is constructed in MATLAB/Simulink as shown in FIG. 6, and the model mainly comprises 4 modules: vehicle model, searching nearest target point, MPC controller, genetic algorithm module.
The path tracking control principle is shown in fig. 7, and during simulation, the RecurDyn vehicle model module outputs the current position and the heading angle of the vehicle. The searching nearest target point module finds the nearest tracking target point according to the current position of the vehicle. And the MPC controller controls the vehicle corner according to the nearest target and outputs the vehicle corner, the course error and the transverse error when tracking the last target point. And the genetic algorithm module solves optimal Np and Nc according to the output of the MPC controller, and the optimal Np and Nc are used for controlling the vehicle corner by the MPC controller when tracking the next target point. The calculation is repeated until the path tracking is completed.
The combined simulation model construction process comprises the following steps:
and opening the m file generated in the point step 5 in the MATLAB, running the m file, and then inputting rdlib in a command line window to generate the RecurDyn vehicle module in the simulink.
The find nearest target point Function is written and called using the MATLAB Function module.
The MPC controller Function is written according to step 3 and called using the MATLAB Function module.
And (4) establishing a corresponding genetic algorithm optimization function according to the time domain parameter optimization schematic diagram in the step (4) by using the optimal time domain parameter solved by the NSGA-II (second generation non-dominated sorting genetic algorithm), and calling the function by using an MATLAB Fcn module.
Step 7, simulation verification specifically comprises the following steps:
the traditional MPC controller is compared with the designed adaptive parameter model prediction in a simulation mode, the vehicle running speed is set to be 0.5m/s, and the adaptive parameter model prediction controller is verified according to 3 paths of U-shaped, 8-shaped and multi-concave-convex curve.
From fig. 8, when the path tracking is performed on the U-shaped curve, the maximum lateral error of the conventional MPC controller is 15.13cm, and the average error is 7.53cm; the maximum lateral error of the adaptive MPC controller is 6.21cm, and the average error is 2.10cm.
From fig. 9, when the 8-shaped curve is subjected to path tracking, the maximum transverse error of the traditional MPC controller is 22.41cm, and the average error is 21.31cm; the maximum lateral error of the adaptive MPC controller is 16.83cm, and the average error is 12.05cm.
From fig. 10, when path tracking is performed on a multi-concave-convex curve, the maximum lateral error of the conventional MPC controller is 22.34cm, and the average error is 6.99cm; the maximum lateral error of the adaptive MPC controller is 19.38cm, and the average error is 5.57cm.
In conclusion, the designed adaptive parameter model predictive controller has better path tracking effect compared with the traditional MPC controller.
Those not described in detail in this specification are within the skill of the art.

Claims (9)

1. A tracking control method for a self-adaptive parameter model prediction path of a articulated tractor is characterized by comprising the following steps:
the method comprises the following steps: according to the structure of the articulated steering semi-track tractor, a multi-body dynamic model of the articulated steering semi-track tractor is built in the RecurDyn;
step two: constructing a multi-body dynamic model of the articulated steering semi-track tractor in the step one, and performing front wheel steering and articulated steering kinematic analysis to obtain a kinematic model;
step three: designing to obtain a path tracking MPC controller according to the kinematics model obtained in the step two and by combining model prediction control, and obtaining input and output of the controller;
step four: adopting a genetic algorithm to carry out self-adaptive real-time optimization on the time domain parameters of the path tracking controller designed in the step three;
step five: creating an m file for the co-simulation;
step six: constructing a joint simulation model in Simulink;
step seven: and (3) carrying out simulation comparison on a traditional model prediction controller and the designed adaptive parameter model prediction, and verifying the effectiveness of the adaptive parameter model.
2. The adaptive parametric model predictive path tracking control method of a articulated tractor according to claim 1, wherein the first step specifically comprises:
modeling a front wheel, a front axle, an engine, a transfer case, a universal joint, a waisted oil tank, a rear case body and a suspension leg, and converting the model into an stp file which can be imported by RecurDyn;
importing the stp model file into RecurDyn, and then carrying out parametric modeling on the crawler chassis by a Track module in the RecurDyn, wherein the crawler chassis part comprises: the rear tension wheel, the driving wheel, the loading wheel and the front tension wheel;
and setting a constraint pair between each part in the model.
3. The adaptive parametric model predictive path tracking control method for a articulated tractor according to claim 2, wherein the front wheel steering kinematics model in the second step is as shown in formula (4):
Figure FDA0003793339520000011
wherein v represents the central speed of the rear axle of the articulated steering tractor,
Figure FDA0003793339520000021
representing the component of the velocity v on the X-axis,
Figure FDA0003793339520000022
indicating the component of velocity v on the Y-axis, psi represents the dog-leg tractor heading angle,
Figure FDA0003793339520000023
indicating the change rate of the course angle, and delta indicating the rotation angle of the front wheel of the articulated steering tractor;
the bending steering kinematic model in the step two is as the formula (10):
Figure FDA0003793339520000024
in the formula, v 1 The speed of the center of the rear axle is shown,
Figure FDA0003793339520000025
representing the velocity v 1 The component in the X-axis is,
Figure FDA0003793339520000026
representing the velocity v 1 Component in the Y axis, θ 1 Indicating the azimuth angle of the rear vehicle body,
Figure FDA0003793339520000027
the derivative of the azimuth angle of the rear vehicle body with respect to time is represented,
Figure FDA0003793339520000028
the angle of the knuckle steer is shown,
Figure FDA0003793339520000029
representing the derivative of the articulated steering angle with respect to time, delta representing the maximum value of the front wheel angle, L 1 Indicating the distance of the hinge point from the rear axle, L 2 Indicating the distance from the hinge point to the front axle and omega indicating the rate of change of the articulated steering angle.
4. The adaptive parametric model predictive path tracking control method for a articulated tractor according to claim 3, wherein the objective function for optimizing the system state quantities and the control quantities in step three is as shown in formula (11):
Figure FDA00037933395200000210
where xi (t) is a state space expression, U (t-1) is a system input, Δ U is a system control increment, Δ η (t + i | t) = η (t + i | t) - η r (t + i | t), which is the difference between the actual output and the reference output, Q, R and ρ are the weight matrices, and ε is the relaxation factor.
5. The adaptive parametric model predictive path tracking control method for a articulated tractor according to claim 4, wherein the fitness function of the genetic algorithm in step four is as shown in formula (12):
Figure FDA00037933395200000211
the flow of the genetic algorithm in step four: initializing a population, then assigning and rounding each individual of the population according to a time domain parameter range, calculating the fitness corresponding to each individual through a fitness function by combining the speed and pose information of a vehicle, outputting an optimal parameter when a termination condition is met, and if the terminal condition is not met, selecting, crossing and mutating to obtain a new population, and performing iterative computation again.
6. The adaptive parametric model predictive path tracking control method for a articulated tractor according to claim 5, wherein the parameter setting of the genetic algorithm in step four comprises: the population size is 200, the probability of cross and variation is 0.6 and 0.1, the evolution times are 20, the time domain value range is predicted (0,60), and the time domain value range is controlled (0,30).
7. The adaptive parametric model predictive path tracking control method of a articulated tractor according to claim 6, characterized in that step five specifically comprises:
in a Communicator module of a RecurDyn software, plant _ in is used for creating system input, the input comprises the rotating speed of a driving wheel, the rotating angle of a front wheel and a bowing steering angle, plant _ out is used for creating system output, and the output comprises a front axle horizontal coordinate, a front axle vertical coordinate and a vehicle course angle;
after the I/O is created, the Simulink module in the Recurdyn software Communicator module generates the.m file for the co-simulation.
8. The adaptive parametric model predictive path tracking control method for a articulated tractor according to claim 7, wherein the combined simulation model in step six comprises:
the system comprises a RecurDyn vehicle model module, a vehicle tracking module and a vehicle tracking module, wherein the RecurDyn vehicle model module outputs the current position and the current course angle of a vehicle;
the system comprises a nearest target point searching module, a tracking target point tracking module and a tracking target point tracking module, wherein the nearest target point searching module finds a nearest tracking target point according to the current position of a vehicle;
the MPC controller controls the vehicle corner according to the latest target and outputs the vehicle corner, the course error and the transverse error when tracking the last target point;
and the genetic algorithm module is used for solving optimal Np and Nc according to the output of the MPC controller and controlling the vehicle corner by the MPC controller when tracking the next target point.
9. The adaptive parametric model predictive path tracking control method for a articulated tractor according to claim 8, wherein the path taken by the simulation verification in the seventh step comprises: u-shaped curve, 8-shaped curve and multi-concave-convex curve.
CN202210961404.1A 2022-08-11 2022-08-11 Adaptive parameter model prediction path tracking control method for articulated tractor Pending CN115373287A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210961404.1A CN115373287A (en) 2022-08-11 2022-08-11 Adaptive parameter model prediction path tracking control method for articulated tractor

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210961404.1A CN115373287A (en) 2022-08-11 2022-08-11 Adaptive parameter model prediction path tracking control method for articulated tractor

Publications (1)

Publication Number Publication Date
CN115373287A true CN115373287A (en) 2022-11-22

Family

ID=84065085

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210961404.1A Pending CN115373287A (en) 2022-08-11 2022-08-11 Adaptive parameter model prediction path tracking control method for articulated tractor

Country Status (1)

Country Link
CN (1) CN115373287A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116176697A (en) * 2023-04-28 2023-05-30 北京市农林科学院智能装备技术研究中心 Method and device for tracking operation path of articulated steering agricultural machinery and electronic equipment

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116176697A (en) * 2023-04-28 2023-05-30 北京市农林科学院智能装备技术研究中心 Method and device for tracking operation path of articulated steering agricultural machinery and electronic equipment

Similar Documents

Publication Publication Date Title
CN110244735B (en) Heuristic dynamic programming control method for tracking preset track by mobile robot
CN108227491B (en) Intelligent vehicle track tracking control method based on sliding mode neural network
CN111158349B (en) Unmanned vehicle model prediction control method based on multi-step linearization strategy
CN111258218B (en) Intelligent vehicle path tracking method based on maximum correlation entropy criterion
CN111158264B (en) Model prediction control rapid solving method for vehicle-mounted application
CN113467470B (en) Trajectory tracking control method of unmanned autonomous trolley
CN114510063B (en) Unmanned tracked vehicle and track tracking control method and system thereof
CN113671962B (en) Transverse control method of unmanned articulated sweeper
Tang et al. Coordinated control for path following of two‐wheel independently actuated autonomous ground vehicle
Zhu et al. An adaptive path tracking method for autonomous land vehicle based on neural dynamic programming
CN115373287A (en) Adaptive parameter model prediction path tracking control method for articulated tractor
CN111240327B (en) Intelligent vehicle iteration track tracking control method based on variable rate model
CN114906128A (en) Automatic parking motion planning method based on MCTS algorithm
CN112099515A (en) Automatic driving method for lane change avoidance
Nan et al. MPC-based path tracking control with forward compensation for autonomous driving
CN116679721A (en) Control method for tracking non-linear track of automobile
CN116225004A (en) Obstacle avoidance method for six-wheel independent driving independent steering robot
Wohner et al. Hierarchical nonlinear model predictive control for an autonomous racecar
CN115447615A (en) Trajectory optimization method based on vehicle kinematics model predictive control
CN112298193B (en) Rapid real-time rear wheel active steering prediction control method
CN114967472A (en) Unmanned aerial vehicle trajectory tracking state compensation depth certainty strategy gradient control method
Wang et al. Path Tracking Method Based on Model Predictive Control and Genetic Algorithm for Autonomous Vehicle
Fényes et al. Data-driven modeling and control design in a hierarchical structure for a variable-geometry suspension test bed
Zhang et al. Nonlinear Model Predictive Control of an Omnidirectional Mobile Robot with Self-tuned Prediction Horizon
Jia et al. Trajectory Planning Controller Design for Autonomous Vehicles

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