Unmanned agricultural machine second-order sliding mode path tracking control method and system considering input time delay

Abstract

The application discloses an unmanned agricultural machine second-order sliding mode path tracking control method and system considering input time delay, constructs a deviation dynamic model containing input time delay characteristics of lateral deviation and heading deviation, and defines state variables of an unmanned agricultural machine path tracking process; an input time delay estimator is designed to online estimate unknown input time delay; based on the input time delay estimator, a compensation system suitable for unknown input time delay is constructed; a new second-order sliding mode system without input time delay is constructed from the compensation system, a new composite second-order sliding mode controller of the unmanned agricultural machine path tracking is designed in combination with a compensation amount of the second-order sliding mode controller, and the unmanned agricultural machine in actual operation tracks an expected path. The application solves the problem of control deviation caused by unknown input signal delay, maintains rapid response and high-precision tracking performance, enables the agricultural machine to track the operation path with high precision in the case of complex disturbance, and improves operation quality.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of agricultural machinery control, and relates to a second-order sliding mode path tracking control method and system for unmanned agricultural machinery considering input time delay. BACKGROUND

[0002] In the global range, many countries are committed to developing unmanned agricultural machinery to cope with the challenges of food security and labor shortage. These machines replace a large amount of manual labor required in traditional agricultural production by integrating autonomous navigation technology. Autonomous navigation technology of unmanned agricultural machinery has been widely used in various agricultural production activities and has become one of the core technologies to realize precision agriculture. In this field, path tracking control is a key link to improve the precision and stability of the autonomous navigation system of agricultural machinery. Therefore, developing an efficient path tracking control strategy is crucial to enhance the tracking precision and robustness of agricultural machinery.

[0003] As a widely used robust control strategy for nonlinear uncertain systems, sliding mode control is favored for its ease of implementation, rapid dynamic response, and strong anti-interference ability. Many researchers choose sliding mode control to solve the path tracking problem of agricultural machinery.

[0004] However, most sliding mode control algorithms are usually first-order sliding mode algorithms. For traditional first-order sliding mode algorithms, the inevitable chattering phenomenon is a major problem. This phenomenon is not only harmful to actuators in practical applications, but also may challenge the stability of the control system. Second-order sliding mode control can effectively reduce chattering, achieve finite-time convergence, improve control accuracy and reliability, and support agricultural machinery to adapt to complex environments faster. Currently, second-order sliding mode control algorithms have been applied to improve the path tracking precision of agricultural machinery, but due to the influence of input time delay, the tracking precision is significantly reduced in the case of large input signal delay. And due to the difficulty of measuring unknown time delay, the precise control of unmanned agricultural machinery is difficult to achieve. SUMMARY

[0005] In view of the deficiencies in the prior art, the present application provides a second-order sliding mode path tracking control method and system for unmanned agricultural machinery considering input time delay, aiming to improve the path tracking accuracy of agricultural machinery under the influence of input time delay, and ensure that the agricultural machinery can achieve high-precision path tracking under complex actual working conditions, thereby improving the working efficiency and making it more in line with the actual application requirements.

[0006] The present application achieves the above technical purpose by the following technical means.

[0007] A second-order sliding mode path tracking control method for unmanned agricultural machinery considering input time delay:

[0008] An input time delay estimator is designed to estimate the unknown input time delay online:

[0009]

[0010] wherein, is the estimation value of the input time delay estimator, τ is the actual value of the input time delay, is the error of the control input under the predicted delay and the real delay, is the error of the control input under the predicted delay and the artificially given delay, and ρ is a normal number, and ξ is a time constant, i.e., the artificially given delay;

[0011] Based on the input time delay estimator, a compensation system suitable for unknown input time delay is constructed:

[0012]

[0013] wherein, λ represents the compensation system, u(t) is the control input without input time delay, p1 and b0 are normal numbers, sign(λ) is a sign function, is the control input under the predicted delay;

[0014] A new type of second-order sliding mode system without input time delay is constructed from the compensation system, a new type of compound second-order sliding mode controller for unmanned agricultural machinery path tracking is designed by combining the compensation amount v of the second-order sliding mode controller, and the unmanned agricultural machinery in actual operation is controlled to track the expected path;

[0015] The new type of compound second-order sliding mode controller is:

[0016] u(t) = -β2sign(|s2| 2α ·sign(s2) + β1|s1| α ·sign(s1)) + v

[0017] wherein, α, β1 and β2 are normal numbers, s1 represents a lateral deviation, and s2 represents a lateral deviation derivative compensated by the compensation system.

[0018] Further, it also includes: constructing a deviation dynamic model containing input time delay characteristics of lateral deviation and heading deviation according to the motion characteristics of unmanned agricultural machinery, and defining the state variables of the unmanned agricultural machinery path tracking process.

[0019] Further, the deviation dynamic model containing input time delay characteristics of lateral deviation and heading deviation is:

[0020]

[0021] wherein, L e , θe respectively represent the lateral deviation and the heading deviation generated when the unmanned agricultural machine tracks the desired path, V is the longitudinal speed of the unmanned agricultural machine, δ is the direction parameter of the unmanned agricultural machine, L t is the wheelbase between the front and rear wheels of the unmanned agricultural machine, c t is the radius of curvature of the position on the desired path at time t, σ τ is the delayed control steering angle with input delay, u(t-τ) is the control input delayed by time τ based on the current time t.

[0022] Further, the cost function of the input delay estimator is:

[0023] Further, the new second-order sliding mode system without input delay is:

[0024]

[0025] where the lateral deviation s1 = s = L e , the derivative of the lateral deviation compensated by the compensation system λ the error of the second-order sliding mode controller under the actual delay and the estimated delay a(t, x) is a bounded function, which is specifically represented as where ω is the uncertainty in system modeling, which is regarded as an unknown disturbance.

[0026] Further, the controller compensation quantity is:

[0027]

[0028] where k is a normal number, and arctan(kλ) represents the inverse tangent function.

[0029] An unmanned agricultural machine second-order sliding mode path tracking control system considering input delay, comprising:

[0030] A host computer, which implements the control method and writes into a processor unit;

[0031] A path planning unit, which is responsible for planning the ideal driving path of the unmanned agricultural machine;

[0032] A navigation positioning system, which is used to continuously obtain the real-time position information of the unmanned agricultural machine;

[0033] A communication unit, which is responsible for coordinating the data communication between various components inside the agricultural machine;

[0034] A processor unit, which is responsible for collecting the state variables required by the control method and outputting the corresponding control instructions through the control method;

[0035] A vehicle control unit is responsible for executing the instructions of the agricultural machine.

[0036] In the technical solution, the upper computer comprises:

[0037] A second-order sliding mode control module is used to design a new type of composite second-order sliding mode controller, which ensures that the lateral deviation and the heading deviation of the agricultural machine converge rapidly in a limited time in the presence of input delay.

[0038] An input delay estimator module is used to estimate the unknown input delay online.

[0039] An auxiliary compensation system module is used to compensate for the system control deviation caused by the input delay and to adjust the second-order sliding mode control module in real time under the condition of input delay.

[0040] A deviation conversion module converts the difference between the current position coordinates of the agricultural machine and the nearest predetermined path coordinate point into lateral deviation and heading deviation.

[0041] A signal delay processing module buffers and delays the collected signals to obtain a control signal with a given delay time.

[0042] The present application has the following beneficial effects:

[0043] 1. The input delay estimator is used to solve the problem of unknown input delay that is difficult to measure. The input delay estimator takes into account the discontinuity of the second-order controller, thereby avoiding the need for differentiation of the controller and effectively solving the challenge of estimating unknown delay in the second-order sliding mode system.

[0044] 2. The input delay estimator is used to construct an auxiliary compensation system suitable for unknown input delay. A new type of second-order sliding mode model is constructed through the auxiliary compensation system to adapt to the inherent input delay and uncertainty of the system, which helps to solve the path tracking problem of agricultural machinery with unknown input delay and uncertainty.

[0045] 3. The second-order sliding mode control considering input delay not only offsets the adverse effects of unknown input delay, but also ensures the hardware safety of the agricultural machine and improves the tracking accuracy of the unmanned agricultural machine considering input delay. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 The present application is a schematic diagram of the second-order sliding mode path tracking control of the unmanned agricultural machine considering input delay.

[0047] Figure 2 The present application is a schematic diagram of the second-order sliding mode path tracking control device of the unmanned agricultural machine considering input delay.

[0048] Figure 3 Structure diagram of the unmanned agricultural machine second-order sliding mode path tracking control system considering input delay of the application;

[0049] Figure 4 Path tracking curve diagram of the agricultural machine of the application;

[0050] Figure 5 Lateral deviation diagram of the agricultural machine path tracking of the application;

[0051] Figure 6 Course deviation diagram of the agricultural machine path tracking of the application;

[0052] Figure 7 Unknown input delay estimation diagram of the agricultural machine of the application. DETAILED DESCRIPTION

[0053] The application provides a second-order sliding mode path tracking control method and system for unmanned agricultural machines considering input delay, and the overall control scheme diagram is as shown in Figure 1 To make the purpose, technical scheme and effects of the application clearer and more explicit, the technical scheme in the example of the application will be described clearly and completely below in combination with the drawings. It should be noted that the described embodiments are only part of the embodiments of the application, not all. They are only used to explain the application, and not to limit the application.

[0054] The second-order sliding mode path tracking control system for unmanned agricultural machines considering input delay is as shown in Figure 2 It will be used to specifically illustrate the application, including the following parts:

[0055] The host computer realizes the writing and debugging of the control program of the application, writes the control algorithm into the processor unit through CAN communication, and records the working information of the agricultural machine through real-time communication;

[0056] The path planning unit is responsible for planning the ideal driving path of the agricultural machine, and real-time monitoring and displaying the current position of the agricultural machine;

[0057] The navigation positioning system is used to continuously obtain the real-time position information of the agricultural machine;

[0058] The communication unit is responsible for coordinating the data communication between various components inside the agricultural machine, including receiving and sending key data;

[0059] The processor unit is responsible for collecting relevant state variable data required by the control program, calculating and outputting the corresponding control instructions through the control program;

[0060] The vehicle control unit is responsible for executing the instructions of the agricultural machine, managing the forward and steering actions of the machine.

[0061] For the above embodiment of the system, the host computer comprises the following modules, as shown in Figure 3 :

[0062] The second-order sliding mode control module is responsible for ensuring that the lateral deviation and the heading deviation of the agricultural machinery can converge rapidly in a limited time in the presence of input delay, so as to realize efficient and rapid tracking of the unmanned agricultural machinery to the predetermined path.

[0063] The input delay estimator module is used to solve the problem of high precision requirement for input delay estimation due to the difficulty in measuring unknown input delay.

[0064] The auxiliary compensation system module is used to compensate for the system control deviation caused by input delay, and to ensure real-time compensation and adjustment of the second-order sliding mode controller under the condition of input delay through an adaptive mechanism.

[0065] The deviation conversion module is responsible for converting the difference between the current position coordinates of the agricultural machinery and the nearest predetermined path coordinates into lateral deviation and heading deviation.

[0066] The signal delay processing module buffers and delays the collected signals to obtain control signals with a given delay time.

[0067] For the above embodiment of the system, an embodiment of the second-order sliding mode path tracking control method for unmanned agricultural machinery considering input delay comprises the following steps:

[0068] S1, according to the motion characteristics of the unmanned agricultural machinery, a deviation dynamic model of lateral deviation and heading deviation containing input delay characteristics is constructed, and state variables related to the path tracking process of the agricultural machinery are defined.

[0069] Specifically, in this example, the tracking deviation dynamic model of the unmanned agricultural machinery with input signal delay is:

[0070]

[0071] Where V is the longitudinal speed of the unmanned agricultural machinery, δ is the direction parameter of the unmanned agricultural machinery, L t is the wheelbase between the front and rear wheels of the unmanned agricultural machinery, c t is the curvature radius of the position on the desired path at time t, L e , θ e respectively represent the lateral deviation and the heading deviation generated when the unmanned agricultural machinery tracks the desired path, σ τ represents the delayed control steering angle with input delay, and u(t-τ) is the control input delayed by time τ based on the current time t.

[0072] S2, due to the actual control input is affected by unknown input delay, using negative gradient optimization technique to design an input delay estimator, quickly and accurately to unknown input delay online estimation, real-time input delay estimation can effectively deal with the unknown input delay problem of unmanned agricultural machinery in the variable environment, to ensure the effectiveness of the controller;

[0073] The unknown input delay is estimated by using negative gradient optimization, and the cost function uses the controller bias as an index, and has the following form:

[0074]

[0075] Wherein, is the cost function, is the estimated value of the input delay estimator, τ is the actual value of the input delay, represents the square error of the control input under the predicted delay and the real delay.

[0076] The innovation of the cost function design lies in: it is not feasible to directly measure the input delay in the actual system, and the traditional cost function based on input delay estimation error cannot be constructed. Therefore, the input delay estimation problem becomes difficult to handle, and the unmeasurable error Compared with the error is easier to obtain.

[0077] The input delay estimator is specifically:

[0078]

[0079] Wherein, is the error of the control input under the predicted delay and the real delay, and is specifically represented as represents the error of the control input under the predicted delay and the artificially given delay, and is specifically represented as ρ is a normal number selected empirically, and ξ is a time constant not less than the control period, that is, the artificially given delay.

[0080] The innovation of the input delay estimator lies in: it can accurately estimate the unknown input delay, the convergence speed can be adjusted by the parameter ρ, and it can be applied to discontinuous controllers such as second-order sliding mode, and has more applicability in actual engineering.

[0081] S3, based on the input delay estimator, a compensation system suitable for unknown input delay is constructed, which can realize adaptive adjustment and real-time offset the control bias caused by the controller input delay; the compensation system is specifically:

[0082]

[0083] Wherein, λ represents the compensation system, u(t) is the control input without input delay, p1 and b0 are normal numbers, and the sign function sign(λ) is specifically represented as:

[0084]

[0085] The innovation of the compensation system is that the input delay estimation value is used for construction, which avoids the problem of unknown input delay. The compensation system can be dynamically adjusted according to the controller deviation, compensates the control error of the system, and shortens the convergence time.

[0086] S4, a new type of second-order sliding mode system without input delay is constructed through the compensation system, which has the following form:

[0087]

[0088] Wherein, s1 = s = L e represents the lateral deviation; represents the derivative of the lateral deviation compensated by the compensation system; represents the error of the second-order sliding mode controller under the actual delay and the estimated delay; a(t, x) is a bounded function, which is specifically represented as Wherein ω is the uncertainty in system modeling, which is regarded as an unknown disturbance.

[0089] The innovation of the design of the new type of second-order sliding mode system is that the non-delay expression of the system is obtained by introducing the compensation system, which solves the problem of difficult real-time controller design. The design of the second-order sliding mode system also simplifies the conditions of system uncertainty, which is more advantageous in verification conditions and parameter selection.

[0090] S5, the compensation amount v of the second-order sliding mode controller is designed, which has the following form:

[0091]

[0092] Wherein, k is a normal number selected empirically, and the function arctan(kλ) represents the inverse tangent function.

[0093] The innovation of the controller compensation amount is that by introducing the auxiliary system, the control law with input delay can be approximated to the actual control law. On the one hand, for agricultural machinery, it can effectively alleviate the deviation problem caused by the input signal delay of agricultural machinery. On the other hand, for unknown input delay, the designed auxiliary system solves the problem of difficult effective compensation of the controller under unknown input.

[0094] S6, based on S4 and S5, a new type of composite second-order sliding mode controller for unmanned agricultural machinery path tracking is designed by power integration method, so that the agricultural machinery can quickly and accurately track the expected path, the new type of composite second-order sliding mode controller is a real-time controller without time delay, and has the following form:

[0095] u(t) = -β2sign(|s2| 2α ·sign(s2) + β1|s1| α ·sign(s1)) + v (7)

[0096] Wherein, β1 and β2 are normal numbers, and the normal number α satisfies the condition α≥1.

[0097] The new type of composite second-order sliding mode controller can ensure the stability of the agricultural machinery in a limited time. In the case of input delay, high-precision tracking effect can be ensured, and in the case of no input delay, ideal second-order sliding mode tracking effect can be further ensured.

[0098] One embodiment of the unmanned agricultural machinery second-order sliding mode path tracking control method and system considering input delay has been specifically explained and described, so that the technical solution is more clear and explicit, and easy for engineers to implement. Based on the example, the effect of the present application is further demonstrated through experiments.

[0099] In order to verify that the present application has excellent tracking precision, when the initial lateral deviation of the unmanned agricultural machinery is 100 cm and the unknown input delay is 50 ms, the tracking precision of the embodiment is tested according to the path tracking performance. The experimental results are shown in Figures 4-7 It is not difficult to see that the control precision of the embodiment in the experiment is higher, and the control method can more effectively track the reference path in the presence of input delay, and the estimation performance of the input delay estimator is also very excellent, so it can be demonstrated that the present application has excellent tracking precision. In summary, the present application greatly improves the dynamic performance and tracking precision while ensuring the hardware safety of the agricultural machinery.

[0100] In the description of the present application, the description of the terms "one embodiment", "some embodiments", "exemplary embodiment", "example", "specific example" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present application, the exemplary description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0101] While the application has been described in terms of various specific embodiments, those skilled in the art will recognize that the application can be practiced with modifications within the spirit and scope of the claims, which are appended hereto. Accordingly, any obvious modifications, equivalents or variations are intended to be covered by the present application.

Claims

1. A second-order sliding mode path tracking control method for unmanned agricultural machinery considering input time delay, characterized in that: an input time delay estimator is designed to estimate the unknown input time delay online; a compensation system suitable for unknown input time delay is constructed based on the input time delay estimator; a new second-order sliding mode system without input time delay is constructed from the compensation system, and a new composite second-order sliding mode controller for unmanned agricultural machinery path tracking is designed by combining the compensation amount v of the second-order sliding mode controller to control the unmanned agricultural machinery to track the expected path in actual operation; the new composite second-order sliding mode controller is: wherein a, β1 and β2 are normal numbers, s1 represents the lateral deviation, and s2 represents the derivative of the lateral deviation compensated by the compensation system. Further comprising: constructing a deviation dynamic model of lateral deviation and heading deviation containing input time delay characteristics based on the motion characteristics of unmanned agricultural machinery, and defining the state variables of the unmanned agricultural machinery path tracking process. wherein is the estimate of the input delay estimator, τ is the actual value of the input delay, is the error of the control input under the predicted delay and the real delay, is the error of the control input under the predicted delay and the artificially given delay, and ρ is a normal number, and ξ is a time constant, i.e., the artificially given delay; The deviation dynamic model of lateral deviation and heading deviation containing input time delay characteristics is: where λ represents a compensation system, u(t) is a control input without input delay, p1 and b0 are normal numbers, sign(λ) is a sign function, a control input for predicting a delay; The new second-order sliding mode system without input time delay is: The controller compensation amount is: u(t) = -β2sign(|s2| 2α • sign(s2) + β1|s1| α • sign(s1)) + v wherein k is a normal number, and arctan(kλ) represents the inverse tangent function.

2. The unmanned farm machine second-order sliding mode path following control method considering input time delay according to claim 1, characterized in that, It comprises:

3. The unmanned farm machine second-order sliding mode path following control method considering input time delay according to claim 2, characterized in that, a host computer that implements the control method and writes into a processor unit; wherein L e , θ e respectively represent lateral deviation and heading deviation generated when the unmanned agricultural machine tracks the desired path, V is the longitudinal speed of the unmanned agricultural machine, δ is the direction parameter of the unmanned agricultural machine, L t is the wheelbase between the front and rear wheels of the unmanned agricultural machine, c t is the radius of curvature of the position on the desired path at time t, σ τ is the delayed control steering angle with input delay, and u(t-τ) is the control input delayed by time τ on the basis of the current time t.

4. The unmanned farm machine second-order sliding mode path following control method considering input time delay according to claim 3, characterized in that, The cost function for the input delay estimator is:

5. The unmanned farm machine second-order sliding mode path following control method considering input time delay according to claim 3, characterized in that, a path planning unit responsible for planning the ideal driving path of unmanned agricultural machinery; where the lateral deviation s1 = s = L e derivative of the lateral deviation compensated by the compensation system error of the second order sliding mode controller under the actual delay and the estimated delay a(t, x) is a bounded function, which is specifically represented as where ω is the uncertainty in the system modeling, which is regarded as an unknown disturbance.

6. The unmanned farm machine second-order sliding mode path following control method considering input time delay according to claim 1, characterized in that, a navigation positioning system for continuously obtaining the real-time position information of unmanned agricultural machinery; a communication unit responsible for coordinating data communication between various components inside the agricultural machinery; 7. A system for implementing the unmanned agricultural machine second-order sliding mode path following control method considering input time delay according to any one of claims 1-6, characterized in that, a processor unit responsible for collecting state variables required by the control method and outputting corresponding control instructions through the control method; a vehicle control unit responsible for executing the instructions of the agricultural machinery. The host computer comprises: a second-order sliding mode control module for designing a new composite second-order sliding mode controller to ensure that the lateral deviation and heading deviation of the agricultural machinery converge rapidly in a limited time in the presence of input delay; an input time delay estimator module for online estimation of unknown input time delay; an auxiliary compensation system module for compensating for system control deviation caused by input time delay and performing real-time compensation adjustment on the second-order sliding mode control module under input delay conditions; a deviation conversion module for converting the difference between the current position coordinates of the agricultural machinery and the nearest predetermined path coordinate point into lateral deviation and heading deviation; 8. The system of claim 7, wherein, a signal delay processing module for buffering and delaying the collected signals to obtain control signals with a given delay time. ​ ​ ​ ​ ​

Images