Agricultural machinery navigation wheel angle controller based on model prediction optimization
By constructing a second-order model and combining it with a model prediction optimization algorithm, precise control of the agricultural machinery navigation system was achieved, solving the problems of navigation accuracy and operation quality in complex farmland environments and improving the navigation performance of agricultural machinery.
Patent Information
- Application Number
- CN202411157740.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-22
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-08-22
AI Technical Summary
Traditional agricultural machinery navigation systems struggle to effectively cope with nonlinear and uncertain factors in complex farmland environments, leading to a decline in navigation accuracy and operational quality.
A model prediction-based optimization-based navigation wheel angle controller for agricultural machinery is adopted. By constructing a second-order model of the agricultural machinery steering system, combining satellite navigation and attitude sensors to obtain position information, and using MATLAB for data processing and model prediction, the navigation wheel angle is calculated and optimized to achieve precise control.
It has improved the navigation accuracy and operational quality of agricultural machinery in complex farmland environments, and promoted the modernization and intelligent development of agriculture.
Smart Images

Figure CN119045328A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a navigation wheel angle controller for agricultural machinery based on model prediction optimization. The controller is mainly used in an electronically controlled steering system of a navigation system for agricultural machinery. By compensating and controlling the steering wheel angle of the agricultural machinery, the stable driving and precise operation of the agricultural machinery in a complex farmland environment can be achieved, thereby improving the automation level and operation efficiency of agricultural production. Background Art
[0002] With the continuous advancement of agricultural modernization in my country, agricultural machinery navigation systems are playing an increasingly important role in agricultural production. Agricultural machinery navigation systems mainly include key technologies such as positioning, path planning, and controller design. Among them, controller design is the core part of achieving precise navigation, which directly affects the driving stability and operation quality of agricultural machinery.
[0003] Traditional agricultural machinery navigation wheel angle controllers mainly use PID control, sliding mode control and other methods, which can meet the needs of agricultural machinery navigation to a certain extent. However, in practical applications, due to the complex and changeable operating environment of agricultural machinery, these traditional methods have certain limitations when dealing with nonlinear and uncertain factors.
[0004] Therefore, it is of great practical significance to study an agricultural machinery navigation wheel angle controller suitable for complex environments. Summary of the invention
[0005] The purpose of the present invention is to solve the problems existing in the existing agricultural machinery navigation system, especially the limitations of traditional controllers in dealing with nonlinear and uncertain factors in complex farmland environments. Normally, the agricultural machinery navigation system obtains the lateral deviation and heading deviation of the vehicle by comparing the actual path with the preset path, and then controls the steering of the vehicle by calculating the desired wheel angle. However, during the vehicle steering process, the response time and error of the vehicle's own electronic steering hydraulic control system are usually ignored. This design is based on the second-order model of most vehicle steering systems, adds predictive control and combines it with an optimization algorithm to achieve precise control of the wheel angle of agricultural machinery, thereby improving the navigation performance of agricultural machinery.
[0006] In order to achieve the above object, the present invention adopts the following technical solutions:
[0007] The present invention provides an agricultural machinery navigation wheel angle controller based on model prediction optimization, which is used to be installed on agricultural machinery and to perform an agricultural machinery navigation wheel angle control method based on model prediction optimization. The control method includes:
[0008] The agricultural machinery navigation wheel angle controller obtains the position information and motion parameters of the agricultural machinery in real time through the satellite navigation positioning system and attitude sensor installed on the agricultural machinery, and calculates the desired front wheel steering angle through the path tracking control algorithm in the navigation controller;
[0009] Inputting the desired wheel angle into a pre-built agricultural machinery steering system model to predict a compensated wheel angle;
[0010] Based on the desired wheel angle and the compensation wheel angle, an optimized navigation wheel angle is calculated;
[0011] Wherein, the method for constructing the agricultural machinery steering system model includes:
[0012] Step S1: construct a standard second-order transfer function G c (s) Mathematical model of the electronically controlled steering system for agricultural machinery;
[0013] Step S2: Based on the pre-built operation data set of the current plot, use MATLAB to obtain the transfer function G of the agricultural machinery operating on this plot. w (s);
[0014] Step S3: According to the transfer function G w (s) and the transfer function G of the current electronic steering system v (s) Calculate the plot operation system K of the current plot;
[0015] Step S4: according to the transfer function G c (s) and the plot operation system K construct a basic model G(s);
[0016] Step S5, calculating the system response time t of the basic model G(s);
[0017] Step S6: converting the basic model G(s) into a discrete time state space model to serve as the agricultural machinery steering system model.
[0018] Optionally, in step S1, G c The expression of (s) is:
[0019]
[0020] The input of the second-order transfer function is the desired steering wheel deflection angle δ u (s), the output is the actual steering wheel deflection angle δ(s), and a, b, and c are the coefficients of the electronic steering system.
[0021] Step S2 includes:
[0022] Step S21, performing actual operation in an operating environment to collect input and output data of the electronically controlled steering system operation to obtain an input and output data set;
[0023] Step S22, using MATLAB data matrix construction, constructing the input and output data sets into a data matrix, where the input matrix is represented as A and the output matrix is represented as B;
[0024] Step S23: According to the input-output matrix, use the system identification toolbox of MATLAB to obtain the transfer function G w (s);
[0025] The expression of the plot operation coefficient K in step S3 is:
[0026]
[0027] The basic model G(s) in step S4 is expressed as:
[0028]
[0029] Optionally, the step S21 further includes filtering and denoising the collected input and output data.
[0030] Optionally, the system response time t in step S5 is defined as the time for the response to enter and remain at ±2% of the final value; the expression is:
[0031]
[0032] Where W n represents the natural frequency of the second-order electronic steering system, and ζ represents the damping ratio of the second-order electronic steering system.
[0033] Optionally, the discrete time state space model in step S6 is expressed as:
[0034] x[k+1]=A d x[k]+B d u[k] Formula (7)
[0035] y[k]=C d x[k] Formula (8)
[0036] Among them, A d , B d , C d is the discretized state space matrix.
[0037] Optionally, the step of inputting the expected wheel angle into a pre-built agricultural machinery steering system model to predict the compensated wheel angle comprises:
[0038] Set Np is the length of the prediction time domain, set N c The length of the control time domain is used to adjust the parameters of the navigation wheel angle controller. The compensation wheel angle y is obtained through simulation experiments and actual tests, which is expressed as:
[0039] N p =t Formula (9)
[0040] N c =t z -t Formula (10)
[0041]
[0042] Where J is the performance index function, y[k+i] is the predicted system output, r[k+i] is the preset path deviation of the agricultural machinery during operation, Δu[k+i]=u[k+i]-u[k+i-1] is the increment of the control input, Q is the control input increment weight matrix, and R is the control increment weight matrix, which is used to balance the state error and the weight of the control input; t z It is the time required for the vehicle to reach the specified location; the symbol T represents transposition.
[0043] Optionally, the method further includes step S7, constructing constraint conditions of the model, taking into account the dynamic constraints and input constraints of the system, and these constraints are expressed as follows:
[0044] u_min ≤u[k+i]≤ u_max formula (12)
[0045] Δu_min ≤ Δu[k+i] ≤ Δu_max Formula (13)
[0046] u_min and u_max are the minimum and maximum values of the control input vector, which define the feasible range of the control input; Δu_min and Δu_max are the minimum and maximum values of the system control increment.
[0047] Compared with the prior art, the present invention has the following beneficial effects:
[0048] 1. The present invention provides an agricultural machinery navigation wheel angle control method based on model prediction optimization, which optimizes the control process of agricultural machinery in complex farmland environments, improves navigation accuracy and operation quality, and promotes the development of agricultural modernization and intelligence.
[0049] 2. The present invention obtains the wheel angle compensation angle by the turning angle obtained in the navigation controller, which has high accuracy. In addition, the present invention constructs a transfer function model of the land parcel and performs model optimization calculation by comparing the operation path in real time, so that the prediction result of the model is more accurate.
[0050] 3. The agricultural machinery equipped with the agricultural machinery navigation wheel angle controller based on model prediction optimization in the present invention has more accurate tracking accuracy than the agricultural machinery without the controller installed. Therefore, the agricultural machinery equipped with the agricultural machinery navigation wheel angle controller based on model prediction optimization can achieve precise control of the wheel angle of the agricultural machinery, improve tracking accuracy, and thus enhance the navigation performance of the agricultural machinery. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] Figure 1 is a control principle diagram of a method for controlling the navigation wheel angle of agricultural machinery based on model prediction optimization described in Example 1;
[0052] Figure 2 is a construction diagram of the agricultural machinery steering system model in Example 1;
[0053] Figure 3 is a flow chart of the prediction algorithm described in Example 1;
[0054] Figure 4 is a flow chart of wheel angle calculation compensation in embodiment 1;
[0055] Figure 5 is a schematic diagram of an Ω-shaped turning curve of a vehicle in Example 2;
[0056] Figure 6 This is a comparison chart of tracking accuracy when a navigation wheel angle controller is installed in Example 2. DETAILED DESCRIPTION
[0057] Combination Figure 1 This embodiment provides an agricultural machinery navigation wheel angle controller based on model prediction optimization, which is used to be installed on agricultural machinery and is used to execute an agricultural machinery navigation wheel angle control method based on model prediction optimization. The control method includes:
[0058] Step 1: The agricultural machinery navigation wheel angle controller obtains the position information and motion parameters of the agricultural machinery in real time through the satellite navigation positioning system and attitude sensor installed on the agricultural machinery, and calculates the desired front wheel steering angle through the path tracking control algorithm in the navigation controller;
[0059] Step 2, inputting the expected wheel angle into a pre-built agricultural machinery steering system model to predict and obtain a compensation wheel angle;
[0060] Among them, combined Figure 2 , the method for constructing the agricultural machinery steering system model includes:
[0061] Step S1: construct a standard second-order transfer function G c (s) Mathematical model of the electronically controlled steering system for agricultural machinery;
[0062] Since most electronically controlled hydraulic steering systems can be regarded as closed-loop control loops of second-order systems, we establish a standard second-order transfer function as the mathematical model of the electronically controlled steering system of agricultural machinery. The expression of this mathematical model is:
[0063]
[0064] The input of the second-order transfer function is the desired steering wheel deflection angle δ u (s), the output is the actual steering wheel deflection angle δ(s), and a, b, and c are the coefficients of the electronic steering system.
[0065] Step S2: Based on the pre-built operation data set of the current plot, use MATLAB to obtain the transfer function G of the agricultural machinery operating on this plot. w (s);
[0066] Step S2 includes:
[0067] Step S21, performing actual operation in an operating environment, collecting input and output data of the electronically controlled steering system operation, filtering and denoising the collected input and output data to obtain an input and output data set;
[0068] Step S22, using MATLAB data matrix construction, constructing the input and output data sets into a data matrix, where the input matrix is represented as A and the output matrix is represented as B;
[0069] Step S23: According to the input-output matrix, use the system identification toolbox of MATLAB to obtain the transfer function G w (s);
[0070] The expression of the plot operation coefficient K in step S3 is:
[0071]
[0072] Step S3: According to the transfer function G w (s) and the transfer function G of the current electronic steering system v (s) Calculate the plot operation system K of the current plot;
[0073] Step S4: according to the transfer function G c (s) and the plot operation system K construct a basic model G(s);
[0074] The basic model G(s) in step S4 is expressed as:
[0075]
[0076] Step S5, calculating the system response time t of the basic model G(s);
[0077] The system response time t is defined as the response to enter and remain within ±2% of the final value; the expression is:
[0078]
[0079] Where W n represents the natural frequency of the second-order electronic steering system, and ζ represents the damping ratio of the second-order electronic steering system.
[0080] Step S6: Perform predictive control design based on the model, use the model predictive control algorithm to design the navigation wheel angle controller, and combine the flowchart of the predictive algorithm with Figure 3 , converting the basic model G(s) into a discrete time state space model to serve as the agricultural machinery steering system model.
[0081] The discrete time state space model in step S6 is expressed as:
[0082] x[k+1]=A d x[k]+B d u[k] Formula (7)
[0083] y[k]=C d x[k] Formula (8)
[0084] Among them, A d , B d , C d is the discretized state space matrix.
[0085] The step of inputting the desired wheel angle into a pre-built agricultural machinery steering system model to predict the compensated wheel angle comprises:
[0086] Set N p is the length of the prediction time domain, set N c The length of the control time domain is used to adjust the parameters of the navigation wheel angle controller. The compensation wheel angle y is obtained through simulation experiments and actual tests, which is expressed as:
[0087] N p =t Formula (9)
[0088] N c =t z -t Formula (10)
[0089]
[0090] Where J is the performance index function, y[k+i] is the predicted system output, r[k+i] is the preset path deviation of the agricultural machinery during operation, Δu[k+i]=u[k+i]-u[k+i-1] is the increment of the control input, Q is the control input increment weight matrix, and R is the control increment weight matrix, which is used to balance the state error and the weight of the control input; t z It is the time required for the vehicle to reach the specified location; the symbol T represents transposition.
[0091] Step S7: construct the constraints of the model, taking into account the dynamic constraints and input constraints of the system. These constraints are expressed as follows:
[0092] u_min ≤u[k+i]≤ u_max formula (12)
[0093] Δu_min ≤ Δu[k+i] ≤ Δu_max Formula (13)
[0094] Among them, u_min and u_max are the minimum and maximum values of the control input vector, which define the feasible range of the control input; Δu_min and Δu_max are the minimum and maximum values of the system control increment.
[0095] Step 3: Combine Figure 4 , based on the desired wheel angle and the compensation wheel angle, calculating an optimized navigation wheel angle;
[0096] The compensation wheel angle y is solved by the above steps, and combined with the desired wheel angle θ given by the navigation controller c , it can be obtained that the formula for the navigation wheel angle of agricultural machinery after model prediction optimization is:
[0097] θ=θ c ±y formula (14)
[0098] Through the above steps, the present invention provides an agricultural machinery navigation wheel angle controller based on model prediction optimization. The controller optimizes the control process of agricultural machinery in complex farmland environments, improves navigation accuracy and operation quality, and promotes the development of agricultural modernization and intelligence.
[0099] Example 2
[0100] This embodiment uses the navigation wheel angle controller described in Embodiment 1 to conduct experiments, specifically:
[0101] In order to verify the tracking accuracy of the agricultural machinery navigation wheel angle controller based on model prediction optimization, the modified Dongfanghong tractor was used as the experimental platform, and the tractor was tracked at low speed and first gear. Figure 5As shown in the figure, the vehicle Ω-shaped turning curve is used to obtain the tracking accuracy comparison diagram of whether the navigation wheel angle controller is installed, as shown in Figure 5 shown.
[0102] Combination Figure 6 ,According to the tracking accuracy comparison chart of whether or not the navigation wheel angle controller is installed, it can be seen that the agricultural machinery installed with the agricultural machinery navigation wheel angle controller based on model prediction optimization has a more accurate tracking accuracy than the one without installation. Therefore, agricultural machinery installed with the agricultural machinery navigation wheel angle controller based on model prediction optimization can achieve precise control of the wheel angle of agricultural machinery, improve tracking accuracy, and thus improve the navigation performance of agricultural machinery.
[0103] It is easy to understand that those skilled in the art can combine, split, reorganize, etc. the embodiments of the present application to obtain other embodiments based on one or several embodiments provided in the present application, and these embodiments do not exceed the protection scope of the present application.
[0104] The present invention and its implementation methods are described schematically above, and the description is not restrictive. The embodiments shown in the embodiments are only part of the implementation methods of the present invention, and the actual structure is not limited thereto. Therefore, if a person skilled in the art is inspired by the embodiments and designs a structure and an implementation method similar to the technical solution without creativity without departing from the purpose of the invention, they should all fall within the protection scope of the present invention.
Claims
1. An agricultural machinery navigation wheel angle controller based on model prediction optimization, used for installation on agricultural machinery, characterized in that: Used to execute a navigation wheel angle control method for agricultural machinery based on model prediction optimization, the control method comprising: The agricultural machinery navigation wheel angle controller obtains the position information and motion parameters of the agricultural machinery in real time through the satellite navigation positioning system and attitude sensor installed on the agricultural machinery, and calculates the desired front wheel steering angle through the path tracking control algorithm in the navigation controller; Inputting the desired wheel angle into a pre-built agricultural machinery steering system model to predict a compensated wheel angle; Based on the desired wheel angle and the compensation wheel angle, an optimized navigation wheel angle is calculated; Wherein, the method for constructing the agricultural machinery steering system model includes: Step S1: construct a standard second-order transfer function G c (s) Mathematical model of the electronically controlled steering system for agricultural machinery; Step S2: Based on the pre-built operation data set of the current plot, use MATLAB to obtain the transfer function G of the agricultural machinery operating on this plot. w (s); Step S3: According to the transfer function G w (s) and the transfer function G of the current electronic steering system v (s) Calculate the plot operation system K of the current plot; Step S4: according to the transfer function G c (s) and the plot operation system K construct a basic model G(s); Step S5, calculating the system response time t of the basic model G(s); Step S6: converting the basic model G(s) into a discrete time state space model to serve as the agricultural machinery steering system model.
2. The agricultural machinery navigation wheel angle controller based on model prediction optimization according to claim 1 is characterized in that: Step S1 c The expression of (s) is: The input of the second-order transfer function is the desired steering wheel deflection angle δ u (s), the output is the actual steering wheel deflection angle δ(s), a, b, c are the coefficients of the electronic steering system; Step S2 includes: Step S21, performing actual operation in an operating environment to collect input and output data of the electronically controlled steering system operation to obtain an input and output data set; Step S22, using MATLAB data matrix construction, constructing the input and output data sets into a data matrix, where the input matrix is represented as A and the output matrix is represented as B; Step S23: According to the input-output matrix, use the system identification toolbox of MATLAB to obtain the transfer function G w (s); The expression of the plot operation coefficient K in step S3 is: The basic model G(s) in step S4 is expressed as:
3. The agricultural machinery navigation wheel angle controller based on model prediction optimization according to claim 2 is characterized in that: The step S21 also includes filtering and denoising the collected input and output data.
4. The agricultural machinery navigation wheel angle controller based on model prediction optimization according to claim 2, characterized in that: The system response time t in step S5 is defined as the time for the response to enter and remain at ±2% of the final value; the expression is: Where W n represents the natural frequency of the second-order electronic steering system, and ζ represents the damping ratio of the second-order electronic steering system.
5. The agricultural machinery navigation wheel angle controller based on model prediction optimization according to claim 2, characterized in that: The discrete time state space model in step S6 is expressed as: x[k+1]=A d x[k]+B d u[k] Formula (7) y[k]=C d x[k] Formula (8) Among them, A d , B d , C d is the discretized state space matrix.
6. The agricultural machinery navigation wheel angle controller based on model prediction optimization according to claim 5, characterized in that: The step of inputting the desired wheel angle into a pre-built agricultural machinery steering system model to predict the compensated wheel angle comprises: Set N p is the length of the prediction time domain, set N c The length of the control time domain is used to adjust the parameters of the navigation wheel angle controller. The compensation wheel angle y is obtained through simulation experiments and actual tests, which is expressed as: N p =t Formula (9) N c =t z -t Formula (10) Where J is the performance index function, y[k+i] is the predicted system output, r[k+i] is the preset path deviation of the agricultural machinery during operation, Δu[k+i]=u[k+i]-u[k+i-1] is the increment of the control input, Q is the control input increment weight matrix, and R is the control increment weight matrix, which is used to balance the state error and the weight of the control input; t z It is the time required for the vehicle to reach the specified location; the symbol T represents transposition.
7. The agricultural machinery navigation wheel angle controller based on model prediction optimization according to claim 6, characterized in that: The method also includes step S7, constructing the constraint conditions of the model, taking into account the dynamic constraints and input constraints of the system, and these constraints are expressed as follows: u_min ≤u[k+i]≤ u_max formula (12) Δu_min ≤ Δu[k+i] ≤ Δu_max Formula (13) u_min and u_max are the minimum and maximum values of the control input vector, which define the feasible range of the control input; Δu_min and Δu_max are the minimum and maximum values of the system control increment.
Citation Information
Patent Citations
Adaptive model prediction control-based harvester path tracking controller design method
CN109884900A
Self-adaptive feed-forward model prediction control method and system for agricultural machinery automatic driving
CN114047748A
Automatic driving navigation equipment for agricultural machinery
CN117075485A
Automatic driving vehicle transverse control method based on DRL-MPC
CN117360544A
Steering control for vehicles
US10829149B1
