Tracking control system and tracking control method for unmanned grain transporting vehicle

Abstract

The invention discloses an unmanned grain transport vehicle tracking control system which is additionally arranged on an unmanned grain transport vehicle and comprises a data acquisition module, a control module and an output module which are connected in sequence. The invention further discloses a tracking control method of the unmanned grain transport vehicle. The method comprises the steps that the system is initialized; the unmanned grain transport vehicle tracking control system obtains the operation states of the unmanned grain transport vehicle and the harvester, and calculates the distance error and the course angle error between the unmanned grain transport vehicle and the harvester; judging whether the unmanned grain transport vehicle enters a fast approaching control mode or not; executing a fast approach control mode; judging whether the unmanned grain transport vehicle enters an online in-place control mode or not; executing an online in-place control mode; executing a stable tracking control mode; and outputting the control variable of the unmanned grain transporting vehicle to realize grain unloading of the unmanned grain transporting vehicle. The invention belongs to the technical field of agricultural machinery cooperative control, and overcomes the defect that an existing unmanned grain transporting vehicle cannot unload grains during advancing.

Description

Technical Field

[0001] The present invention belongs to the technical field of agricultural machinery cooperative control, and relates to an unmanned grain carrier tracking control system. The present invention also relates to a tracking control method for an unmanned grain carrier. Background Art

[0002] The grain carrier is an important transportation tool in the process of modern agricultural mechanized harvesting operations, and cooperates with the harvester to complete the work of grain harvesting and transportation. The traditional grain unloading operation designates a grain unloading point. When the granary of the harvester is almost full, the grain carrier and the harvester stop at the designated grain unloading point to unload the grain. This method increases a large amount of non-working time and reduces the overall harvesting operation efficiency. With the development of intelligent and refined agricultural machinery, the unmanned operation of agricultural machinery has become a research hotspot. During the progress of the harvesting work, the unmanned grain carrier tracks the harvester to unload the grain during movement, which can effectively improve the work efficiency.

[0003] The key to the cooperative control of the grain carrier and the harvester during movement is the tracking of the harvester by the unmanned grain carrier. The existing control methods only solve the longitudinal tracking problem of the unmanned grain carrier when the harvester is on a straight working path. However, the farmland environment is complex and diverse, and there are inevitably irregular situations at the farmland boundaries. There are also some special situations, such as crop lodging. In order to avoid affecting the yield by pressing the stubble, at this time, the harvester also needs to work on a non-straight path. In order to realize the in-motion grain unloading when the harvester works on a non-straight path, the unmanned grain carrier needs to achieve the control of two degrees of freedom in the longitudinal and transverse directions, and the tracking control difficulty of the unmanned grain carrier increases. Summary of the Invention

[0004] The purpose of the present invention is to provide an unmanned grain carrier tracking control system, which solves the problem that the unmanned grain carrier and the harvester in the prior art cannot achieve in-motion docking and it is difficult to achieve in-motion grain unloading.

[0005] Another purpose of the present invention is to provide a tracking control method for an unmanned grain carrier, which solves the problem that during the work of the harvester on any path in the prior art, the unmanned grain carrier cannot quickly and stably track the harvester, and it is difficult to achieve in-motion grain unloading.

[0006] The technical solution adopted by the present invention is an unmanned grain carrier tracking control system. The unmanned grain carrier tracking control system is installed on the unmanned grain carrier, and this tracking control system includes a data acquisition module, a control module, and an output module that are connected in sequence.

[0007] Another technical solution adopted by the present invention is a tracking control method for an unmanned grain carrier. Using the above-mentioned unmanned grain carrier tracking control system, it is implemented according to the following steps: Step 1: Initialize the system; Step 2: The unmanned grain carrier tracking control system obtains the operating states of the unmanned grain carrier and the harvester; Step 3: Calculate the distance error and heading angle error between the unmanned grain transport vehicle and the harvester; Step 4: Determine whether the unmanned grain transport vehicle enters the rapid approach control mode; Step 5: The unmanned grain transport vehicle tracking control system executes a fast approach control mode; Step 6: Determine whether the unmanned grain transport vehicle has entered the online position control mode; Step 7: The unmanned grain transport vehicle tracking and control system executes the online in-place control mode; Step 8: The unmanned grain transport vehicle tracking control system executes a stable tracking control mode; Step 9: Output the control variables of the unmanned grain transport vehicle to realize the unloading of grain by the unmanned grain transport vehicle.

[0008] The beneficial effects of the present invention are: (1) it provides an unmanned grain transport vehicle tracking and control system, and proposes a multi-mode control scheme for the unmanned grain transport vehicle to quickly approach, get on line and position, and stably track, thereby ensuring that the unmanned grain transport vehicle can quickly track the harvester and realize grain unloading while moving; (2) in combination with the kinematic characteristics of the grain transport vehicle, a stable tracking control mode for the unmanned grain transport vehicle is set, thereby solving the technical problem of stably tracking the unmanned grain transport vehicle when the harvester is working on any path and realizing grain unloading while moving. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 It is a block diagram of the unmanned grain transport vehicle tracking control system in the method of the present invention; Figure 2 It is a path diagram of the multi-mode control principle of the unmanned grain transport vehicle in the method of the present invention; Figure 3 This is the control process of the unmanned grain transport vehicle multi-mode tracking harvester in the method of the present invention; Figure 4 This is the control process of the unmanned grain transport vehicle rapid approach control mode in the method of the present invention; Figure 5 It is a schematic diagram of the shortest path planning and determination of the preview point of the unmanned grain transport vehicle in the fast approach control mode in the method of the present invention; Figure 6 It is the control process of the on-line and in-place control mode of the unmanned grain transport vehicle in the method of the present invention; Figure 7 It is a schematic diagram of the on-line path planning and the determination of the preview point of the unmanned grain transport vehicle in the on-line position control mode in the method of the present invention; Figure 8 This is a schematic diagram of the tracking path of the harvester by the unmanned grain transport vehicle in Example 1 of the method of the present invention; Figure 9 This is a schematic diagram of the tracking path of the harvester by the unmanned grain transport vehicle in Example 2 of the method of the present invention; Figure 10It is a schematic diagram of the tracking path of the unmanned grain carrier to the harvester in Embodiment 3 of the method of the present invention; Figure 11 It is a schematic diagram of the tracking path of the unmanned grain carrier to the harvester in Embodiment 4 of the method of the present invention; Figure 12 It is a schematic diagram of the tracking path of the unmanned grain carrier to the harvester in Embodiment 5 of the method of the present invention; Figure 13 It is a schematic diagram of the tracking path of the unmanned grain carrier to the harvester in Embodiment 6 of the method of the present invention. Detailed implementation manners

[0010] The method of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0011] In order to improve agricultural efficiency, the harvester needs to adopt a non-stop grain unloading mode. When the grain in the harvester's grain bin reaches a certain capacity, the unmanned grain carrier quickly approaches and travels synchronously with the harvester without the harvester stopping working. At the same time, the harvester transfers the harvested grain to the unmanned grain carrier through the grain unloading port. This non-stop grain unloading mode reduces the time for stopping and starting, and significantly improves the continuity and efficiency of the harvesting operation. During the above grain unloading process, if the unmanned grain carrier cannot track the harvester in a timely and accurate manner and complete the docking, it may lead to untimely grain unloading or even grain spilling, affecting the harvesting effect. In order to ensure that the unmanned grain carrier can track the harvester quickly and stably, the method of the present invention designs an unmanned grain carrier tracking control system and a multi-mode tracking control method according to the relative position between the unmanned grain carrier and the harvester.

[0012] Refer to Figure 1 It is an overall composition block diagram of the unmanned grain carrier tracking control system. The unmanned grain carrier tracking control system is installed on the unmanned grain carrier. This tracking control system mainly includes three functional modules, namely, a data acquisition module, a control module, and an output module that are connected in sequence: The data acquisition module is used to obtain various operating states, including the RTK-inertial navigation combined sensor, the angle sensor, and the wireless communication module, and input the serial communication signal into the control module; The control module mainly includes a main processor. The main processor executes the multi-mode control algorithm for the unmanned grain carrier tracking control, obtains the control quantities of the unmanned grain carrier, including the front wheel steering angle and the speed magnitude, and inputs these control quantities into the output module; The output module outputs the front wheel steering angle and the speed magnitude of the unmanned grain carrier calculated by the control algorithm of the main processor to the movement execution mechanism of the grain carrier through the CAN bus to control the driving of the grain carrier.

[0013] The unmanned grain carrier tracking control system includes a tracking control part and various sensors installed on the unmanned grain carrier for obtaining the operating state of the unmanned grain carrier. The system hardware composition is as follows: 1) Main processor: It is used to execute various functions of the unmanned grain transport vehicle tracking control system. The main processor is of the STM32F4 model, with a high-performance ARM Cortex-M4 core, supporting complex real-time calculations, providing rich communication interfaces (CAN, UART, SPI, I2C), capable of connecting all sensors and actuators, and having low power consumption and high reliability, which is suitable for vehicle embedded applications.

[0014] 2) RTK-Inertial navigation combined sensor: (The RTK-Inertial navigation combined sensor includes an RTK-GPS module and an Inertial Measurement Unit (IMU)), which is used to measure the position, heading angle and speed of the unmanned grain transport vehicle. The RTK-Inertial navigation combined sensor sends data to the main processor through a serial port (UART); Among them, the RTK-GPS module selects the NEO-M8P chip, which supports multiple satellite navigation systems such as GPS and Beidou, and provides centimeter-level navigation and positioning accuracy; Among them, the Inertial Measurement Unit (IMU) selects the BMI088 inertial sensor, which provides high-frequency motion state data, and can improve the navigation and positioning accuracy through data fusion technologies such as Kalman filtering.

[0015] 3) Angle sensor: It is used to measure the front wheel rotation angle of the unmanned grain transport vehicle. The angle sensor communicates with the main processor through SPI. The angle sensor can select an optoelectronic coded angle sensor with high angle measurement accuracy.

[0016] 4) Wireless communication module: It is used to obtain the operating state of the harvester. The wireless communication module communicates with the main processor through a serial port (UART). The wireless communication module selects a 433MHz communication device, which has low power consumption and supports wireless communication over a distance of hundreds of meters, and is suitable for data transmission between the unmanned grain transport vehicle and the harvester during field operations.

[0017] Figure 2 It is a schematic diagram of the path for the multi-mode control algorithm of the unmanned grain transport vehicle. Figure 2In it, the thick dash-dotted line is the boundary of the field plot, the symbol ▶ is the harvester, and the solid curve is the harvester's trajectory; the symbol ▷ is the unmanned grain carrier, and the dashed curve is the unmanned grain carrier's trajectory. The initial position of the unmanned grain carrier is the waiting place C2_1 at the field head. The harvester moves along the harvesting trajectory. When the harvester issues a command to summon the unmanned grain carrier at C1_1, after the unmanned grain carrier receives the summons command, when the relative position between the unmanned grain carrier and the harvester is relatively far, the unmanned grain carrier adopts a fast approach control mode and approaches the harvester quickly at the shortest distance; when the unmanned grain carrier reaches the C2_2 position and approaches the harvester, it adopts an on-line positioning control mode, aiming to adjust the position of the unmanned grain carrier and the error between the heading angle of the unmanned grain carrier and the heading angle of the harvester's grain unloading port; when the unmanned grain carrier reaches the C2_3 position and the position error and heading angle error between it and the harvester's grain unloading port are less than the given threshold, the unmanned grain carrier then adopts a stable tracking control mode to ensure the synchronous operation of the unmanned grain carrier and the harvester, and then smoothly implement the operation of unloading grain while moving forward.

[0018] Refer to Figure 3 It is the control flow chart of the multi-mode of the unmanned grain carrier. The tracking control method of the unmanned grain carrier of the present invention is implemented according to the following steps by using the aforementioned unmanned grain carrier tracking control system: Step 1: Initialize the system, When the unmanned grain carrier tracking control system is powered on, relevant parameters are initialized, including the following parameters: 1.1) Parameters related to the unmanned grain carrier and the harvester: the wheelbase of the front and rear wheels of the harvester , the maximum safe driving speed of the unmanned grain carrier , the upper and lower bounds of the front wheel steering angle of the unmanned grain carrier , the wheelbase of the front and rear wheels of the unmanned grain carrier , the minimum turning radius of the unmanned grain carrier ; 1.2) Parameters for switching the control mode of the unmanned grain carrier: the distance error threshold , the heading angle error threshold , used to judge whether to switch to the stable tracking control mode; the distance error threshold 2 , used to judge whether to switch to the fast approach control mode; 1.3) PID controller parameters for the fast approach control of the unmanned grain carrier: the proportional coefficient , the integral coefficient and the differential coefficient , the forward viewing distance ; 1.4) PID controller parameters for the on-line positioning control of the unmanned grain carrier: the proportional coefficient , the integral coefficient and the differential coefficient , the forward viewing distance ; 1.5) MPC controller parameters for the stable tracking control of the unmanned grain carrier: upper and lower bounds of the front wheel steering angle , upper and lower bounds of the speed , control step size of the controller and prediction steps , weight matrix of the control variables of the controller and weight matrix of the state variables .

[0019] Step 2: The unmanned grain carrier tracking control system obtains the operating states of the unmanned grain carrier and the harvester. 2.1) The operating states of the unmanned grain carrier include: the current position of the unmanned grain carrier , the current heading angle of the unmanned grain carrier , the current speed magnitude of the unmanned grain carrier , the current front wheel steering angle of the unmanned grain carrier ; The unmanned grain carrier tracking control system obtains the current position, current heading angle, and current speed magnitude of the unmanned grain carrier through the RTK-inertial navigation combined sensor, and uses an angle sensor (photoelectric encoder angle sensor) to obtain the current front wheel steering angle of the unmanned grain carrier.

[0020] 2.2) The operating states of the harvester include: the current position of the unloading port of the harvester , the current heading angle of the harvester , the current speed magnitude of the harvester , the current front wheel steering angle of the harvester ; The unmanned grain carrier tracking control system obtains the current position, current heading angle, current speed magnitude, and current front wheel steering angle of the harvester through a wireless communication module (preferably a 433 wireless communication method).

[0021] Step 3: The unmanned grain carrier tracking control system calculates the distance error and heading angle error between the unmanned grain carrier and the harvester. According to the current position of the unmanned grain carrier and the current position of the unloading port of the harvester , the distance error between the unmanned grain carrier and the unloading port of the harvester is calculated. At the same time, according to the current heading angle of the unmanned grain carrier and the current heading angle of the harvester, the heading angle error between the unmanned grain carrier and the harvester is calculated. The calculation formulas are as follows: .

[0022] Step 4: The unmanned grain transport vehicle tracking control system determines whether the unmanned grain transport vehicle enters the fast approach control mode. The judgment principle is: If , go to Step 5. The unmanned grain transport vehicle tracking control system executes the fast approach control mode, and the unmanned grain transport vehicle approaches the harvester quickly; if , directly go to Step 6.

[0023] Step 5: The unmanned grain transport vehicle tracking control system executes the fast approach control mode. Figure 4 It is the control process of the fast approach control mode of the unmanned grain transport vehicle. The function of the fast approach control mode is to control the unmanned grain transport vehicle to approach the harvesting vehicle in the fastest way. In this mode, the driving path of the unmanned grain transport vehicle tracking the harvester is the shortest path M1 between the current position of the unmanned grain transport vehicle and the unloading port of the harvester.

[0024] During the path tracking process, considering that the fast approach control process has low requirements for accuracy, this step adopts a pure path tracking PID control algorithm with good robustness and low computing power requirements. At this time, the unmanned grain transport vehicle will run at the maximum safe speed allowed in the field .

[0025] The basic principle of the pure path tracking PID control algorithm is: Simplify the unmanned grain transport vehicle into a two-wheel bicycle model, and set the current position of the unmanned grain transport vehicle ( x s ，y s ) as the center position of the rear wheel of the unmanned grain transport vehicle. According to the principle of the pure path tracking PID control algorithm, with the current position of the unmanned grain transport vehicle as the origin, match a preview point forward on the planned path at a preview distance , and use the lateral error between the current position of the unmanned grain transport vehicle and the preview point as feedback to control the front wheel steering angle of the unmanned grain transport vehicle through the PID algorithm.

[0026] 5.1) Plan the shortest path. Figure 5 It is a schematic diagram of the shortest path planning and preview point determination of the unmanned grain transport vehicle in the fast approach control mode. Figure 5 In, the current speed of the harvester is the vector , the direction of the vector is the current heading of the harvester, and the magnitude of the vector is the current speed magnitude of the harvester ; the current speed of the unmanned grain transport vehicle is the vector , the vector The direction of [vector] is the current heading of the unmanned grain carrier, and the vector modulus is the current speed magnitude of the unmanned grain carrier ; In order for the unmanned grain carrier to approach the harvester as soon as possible, take the current position of the unmanned grain carrier and the current position of the harvester's grain unloading port The straight line segment between them is the shortest path ; As Figure 5 shown by the dashed line in is the planned shortest path.

[0027] 5.2) Determine the preview point , If the current position of the unmanned grain carrier ( x s ，y s ) and the current position of the harvester's grain unloading port The distance between them is less than the forward view distance , then is the preview point ; Otherwise, take the current position of the unmanned grain carrier ( x s ，y s ) as the origin, and on the planned path Match the preview point forward at the forward view distance ; As Figure 5 shown by the dashed line of the shortest path in, the point marked with " " is the current preview point ; 5.3) Calculate the lateral error , Calculate the distance between the preview point ( x s ，y s ) and the current position of the unmanned grain carrier , and the calculation formula is: , Calculate the vector pointing angle from the current position of the unmanned grain carrier to the preview point , and the calculation formula is: , Thus, the lateral error The calculation formula is: ; 5.4) Calculate the control variable of the unmanned grain transport vehicle by the PID algorithm , Through the lateral error Update the three current error components , , , and the calculation formula is: , , , where and are the historical values of the error components obtained from the previous control loop calculation; If it is the first control loop, then and are defaulted to 0, Update the historical values of the error components and and save them for the next control loop. The calculation formula is: , , Through the lateral error Combine with the proportional coefficient , integral coefficient and derivative coefficient respectively for the linear combination of the three parts to calculate the control variable of the unmanned grain transport vehicle. The calculation formula is: ; 5.5) Output the control variable, Output the control variable of the unmanned grain transport vehicle, including the front wheel angle of the unmanned grain transport vehicle being and the speed magnitude of the unmanned grain transport vehicle, and execute step 9; Step 6: The unmanned grain transport vehicle tracking control system determines whether the unmanned grain transport vehicle enters the online positioning control mode. The determination principle is: If and , execute step 8, and the unmanned grain transport vehicle tracking control system switches to the stable tracking control mode; Otherwise, execute step 7, and the unmanned grain transport vehicle tracking control system switches to the online positioning control mode.

[0028] Figure 6 This is the control flow of the online positioning control mode of the unmanned grain transport vehicle in the method of the present invention.

[0029] Step 7: The unmanned grain transport vehicle tracking control system executes the online positioning control mode The function of the on-line positioning control mode is to plan the movement path and speed of the unmanned grain carrier, so as to adjust the position and heading angle of the unmanned grain carrier, ensuring the position error and the heading angle error within a certain range, preparing for the subsequent stable tracking control of the unmanned grain carrier.

[0030] In this step, the path planning of the on-line positioning control mode adopts the Dubins curve, and the path tracking adopts the pure path tracking PID control algorithm with better robustness and lower computing power requirements. Refer to Figure 6 which is the control flow chart of the on-line positioning control mode of the unmanned grain carrier.

[0031] 7.1) Plan the on-line path. Figure 7 which is the schematic diagram of the shortest path planning and preview point determination of the unmanned grain carrier in the on-line positioning control mode. In the figure, the current speed of the harvester is the vector , the direction of the vector is the current heading of the harvester, and the magnitude of the vector is the magnitude of the current speed of the harvester ; the current speed of the unmanned grain carrier is the vector , the direction of the vector is the current heading of the unmanned grain carrier, and the magnitude of the vector is the magnitude of the current speed of the unmanned grain carrier .

[0032] The Dubins curve is a geometric line formed by the combination of arc segments and straight line segments, connecting the start and end points under the constraints of curvature constraint, start and end pose constraints, and only forward constraints. According to the constraint conditions generated by the Dubins curve, at the start and end points, the tangent direction of the Dubins curve is consistent with the given direction. In this step, the minimum turning radius of the unmanned grain carrier is the curvature constraint of the Dubins curve, and the starting point of the path is the current position of the unmanned grain carrier, and its tangent direction is the current heading of the unmanned grain carrier; considering that the curvature change of the working path of the harvester is not very large, that is, the heading of the harvester will not change violently in a short time, the harvester can continue to drive along the current heading in the future, and the heading angle of the harvester remains Remain unchanged. Select a point on this heading as the path end point, and its tangent direction is the current heading of the harvester. To plan the Dubins curve with the shortest path length, it is necessary to select the path end point. On the current heading of the harvester, take an alternative path end point at a fixed distance interval. Generate Dubins curves between the path start point and each alternative path end point respectively. Finally, select the Dubins curve with the shortest length among them as the planned up-line path. Refer to Figure 7 In , the path end point is The planned Dubins curve (i.e., Figure 7 The dashed line M2 in ) is the shortest route, which is the up-line path based on the current state of the system, and its length is 7. 2) Calculate the control variables of the unmanned grain carrier , To make the unmanned grain carrier and the harvester reach the path end point at the same time, calculate the magnitude of the target speed of the unmanned grain carrier during the up-line process. The calculation formula is: ; 7.3) Determine the preview point , If the distance between the current position ( x s ,y s ) of the unmanned grain carrier and the current position of the unloading port of the harvester is less than the forward viewing distance , then is the preview point ; Otherwise, taking the current position ( x s ,y s ) of the unmanned grain carrier as the origin, match the preview point forward on the planned path with the forward viewing distance ; Figure 7 In the up-line path shown by the dashed line , the point marked with " " is the current preview point .

[0033] 7.4) Calculate the lateral error , Calculate the preview point and the current position of the unmanned grain carrier x s ，ys The distance between is calculated as follows: , Calculate the vector pointing angle from the current position of the unmanned grain transport vehicle pointing to the preview point , The calculation formula is: , Therefore, the lateral error is calculated as follows: ; 7.5) Calculate the control variable of the unmanned grain transport vehicle by the PID algorithm , Update the three current error components through the lateral error , , , The calculation formula is: , , , where and are the historical values of the error components calculated in the previous control loop; If it is the first control loop, then and are defaulted to 0, Update the historical values of the error components and and save them for the next control loop. The calculation formula is: , , Calculate the control variable of the unmanned grain transport vehicle through the linear combination of the lateral error combined with the proportional coefficient , the integral coefficient and the differential coefficient in three parts. The calculation formula is: The calculation formula is: ; 7.6) Output the control variable. The output control variables of the unmanned grain transport vehicle include: the front wheel steering angle of the unmanned grain transport vehicle is , the speed magnitude of the unmanned grain transport vehicle , and execute step 9; Step 8: The unmanned grain transport vehicle tracking control system executes the stable tracking control mode. The function of the stable tracking control mode is to control the unmanned grain carrier to accurately track the unloading port of the harvester after the errors between the pose of the unmanned grain carrier and the position of the unloading port of the harvester and the heading angle error are both small, so that the harvester can complete the unloading operation during travel. During the process of the harvester driving along a non-linear working path, the tracking control of the unmanned grain carrier for the harvester needs to consider the lateral error, longitudinal error and heading angle error. This step designs a tracking controller based on the model predictive control (MPC) method, called the model predictive control (MPC) controller, which takes the current operating state of the harvester and the current operating state of the unmanned grain carrier as inputs and outputs the optimal control variables for the grain carrier.

[0034] The model predictive control (MPC) controller predicts the future behavior of the system through a prediction model. On this basis, it compensates for the lack of model accuracy, suppresses disturbances and improves control accuracy through rolling optimization and feedback correction; at the same time, the model predictive control (MPC) controller also allows adjusting the prediction step size and weight matrix to balance the relationship between control performance and computing resources under the condition of limited computing resources; considering the characteristics of the uncertainty of the movement of the unmanned grain carrier, the non-linearity of the model, many control variables and certain constraints, etc., using model predictive control (MPC) is the most effective solution.

[0035] The settings of the model predictive control (MPC) controller are as follows: 8.1) Construct a tracking prediction model for the unmanned grain carrier, Based on the kinematic model method of a two-wheeled bicycle, simplify the modeling of the process of the unmanned grain carrier tracking the harvester, construct a tracking prediction model for the unmanned grain carrier, take the front wheel steering angle and speed of the unmanned grain carrier as control input variables, and take the error between the pose state of the unmanned grain carrier and the pose state of the unloading port of the harvester as the state variable, the first set of control variables is the heading angle and the speed magnitude of the unmanned grain carrier, and the second set of control variables is the heading angle and the speed magnitude ， of the harvester. Then the state space equation of the tracking prediction model of the unmanned grain carrier is as follows: Linearize the above state space equation, set the , , , time as the current time, and transform to get a new linear state space equation: , ， ， ， , ， ， , wherein, 、 and are the reference state of the state variable, the reference state of the control variable of the unmanned grain carrier, and the reference state of the control variable of the harvester respectively; the reference state is the benchmark state for model linearization, and it is required to be as close as possible to the actual state of the model.

[0036] According to the performance requirements of the unmanned grain carrier tracking control system, it is required that the position error and the heading angle error between the unmanned grain carrier and the harvester are minimized. Therefore, all elements in the reference state are taken as 0; because the operating states of the unmanned grain carrier and the harvester are very close during the stable tracking process, the reference state of the front wheel steering angle of the unmanned grain carrier is set as the current front wheel steering angle of the harvester, and the reference state of the speed magnitude of the unmanned grain carrier is set as the current speed magnitude of the harvester; because the curvature of the working path of the harvester is not very large and the speed magnitude remains stable during the working process, it is assumed that the front wheel steering angle and the speed magnitude of the harvester remain unchanged in the prediction time domain. The reference state of the front wheel steering angle of the harvester is set as the current front wheel steering angle of the harvester, and the reference state of the speed magnitude of the harvester is set as the current speed magnitude of the harvester; Then, based on the above linear state space equation, the control step of the MPC controller is obtained, and the discrete state space equation of the unmanned grain carrier tracking prediction model is as follows: ， wherein, ， ， ， ， Among them, the value corresponding variable starts from the current moment k and experiences the predicted value after a control step size; Meanwhile, based on the above discrete state space equation, combined with the prediction step of the model predictive control MPC controller , the matrix equation of the unmanned grain transport vehicle tracking prediction model is obtained as follows: ， ， ， ， ， ， ， Since it is assumed that the front wheel steering angle and speed magnitude of the harvester remain unchanged within the control time domain, then there are: ， Then the aforementioned equation is simplified to: ; 8.2) Optimization solution, Based on the matrix equation of the above unmanned grain transport vehicle tracking prediction model, combined with the control variable weight matrix and the state variable weight matrix of the model predictive control MPC controller, an optimization problem is constructed, and the constructed objective function is: ， Among them, ， ， ， ， Add the control variable constraints of the model predictive control MPC controller: the upper and lower bounds of the front wheel steering angle and the upper and lower bounds of the speed , the expression of the entire multi-objective optimization problem is transformed into: , This optimization problem only has control variable increment constraints and no state variable constraints, and can be regarded as a quadratic programming (QP) problem with simple boundary constraints. Using a mature QP solver, the optimal control quantity increment sequence can be efficiently solved. ; 8.3) Calculate the control variables of the unmanned grain transport vehicle from and , Extract the control increment corresponding to the current moment from , update the control quantity at the current moment , and obtain the target front wheel angle and the target speed of the unmanned grain transport vehicle. The expressions are as follows: = , = , where the extraction calculation formulas for and are: , ; 8.4) Output the control variables. The output control variables of the unmanned grain transport vehicle include: the front wheel angle of the unmanned grain transport vehicle is , the speed magnitude of the unmanned grain transport vehicle , and execute step 9; Step 9: Output the control variables of the unmanned grain transport vehicle to achieve unloading of the unmanned grain transport vehicle. The unmanned grain transport vehicle tracking control system sends the control variables of the unmanned grain transport vehicle output by different control modes (including the front wheel angle of the unmanned grain transport vehicle and the speed magnitude ) to the execution mechanism of the unmanned grain transport vehicle, so that the unmanned grain transport vehicle adjusts its driving state according to the control information and completes a task cycle. Then return to step 2 to enter a new task cycle.

[0037] Basic settings of the embodiment: There is a harvester and an unmanned grain transport vehicle in a certain farmland field. The front and rear wheel spacing of the harvester is , the driving speed is set to , and the maximum safe driving speed of the unmanned grain transport vehicle is , the upper and lower bounds of the front wheel steering angle is , the front and rear wheel track is , the minimum turning radius is ; Install RTK-inertial combined sensors on two agricultural machines respectively to obtain the current position of the unmanned grain carrier in real time ( ), the heading angle and the speed magnitude , and at the same time, the harvester transmits the current position of its own grain unloading port ( ), the heading angle and the speed magnitude to the unmanned grain carrier through the 433 MHz wireless communication module; Install photoelectric encoder angle sensors on two agricultural machines respectively to obtain the current front wheel steering angle of the unmanned grain carrier in real time , and at the same time, the harvester transmits the current front wheel steering angle of its own to the unmanned grain carrier through the 433 MHz wireless communication module.

[0038] Set the thresholds for judging the conversion of each control mode: the distance error threshold = 3 m, the distance error threshold = 20 m, the heading angle error threshold = ; Set the PID control parameters for the fast approach control mode: the proportional coefficient , the integral coefficient , the differential coefficient , the forward viewing distance is ; Set the PID control parameters for the upper line in-place control mode: the proportional coefficient , the integral coefficient , the differential coefficient , the forward viewing distance is ; Set the parameters of the model predictive control MPC controller for the stable tracking control mode: the upper and lower bounds of the front wheel steering angle is , the upper and lower bounds of the speed is , the controller control step is 0.1 s, the prediction step is 5, the controller control variable weight matrix is , the state variable weight matrix is ; To ensure synchronous grain unloading operation, set the grain unloading position error threshold as , set the stable time required for unloading grain to 30s.

[0039] Example 1 (the working path of the harvester is Line 1) Refer to Figure 8 , the symbol ▶ represents the harvester, the symbol ▷ represents the unmanned grain carrier, and the trajectory of the grain unloading port of the harvester is the dotted line shown in the figure. The initial position of the unmanned grain carrier is A1 , and the initial heading angle is 90°. The harvester is at the initial position B1 , when the initial heading angle is 0°, a grain transportation signal is sent to the unmanned grain carrier. After receiving the signal, the unmanned grain carrier starts from the initial position A1 and performs rapid approach - on-line positioning - stable tracking on the harvester according to the method steps of the present invention.

[0040] Taking 0.1 second as the execution period, refer to Figure 8 The thin solid line in is the tracking trajectory of the unmanned grain carrier. The unmanned grain carrier receives the grain transportation signal at position A1. First, according to Step 3, calculate the position error between the harvester and the unmanned grain carrier heading angle error , according to Step 4, the position error > position error threshold = 20 meters, the control system transfers to Step 5, and the system activates the rapid approach control mode.

[0041] In the rapid approach control mode, when the unmanned grain carrier runs to position A2 in the figure, the harvester runs to position B2 in the figure. At this time, the position error , the heading angle error , according to Step 4, the position error < position error threshold = 20 meters, the control system transfers to Step 6. According to Step 6, the position error > position error threshold = 3 meters and the heading angle error > heading angle error threshold = , the control system transfers to Step 7, and the system activates the on-line positioning control mode.

[0042] In the on-line positioning control mode, the unmanned grain carrier runs to Figure 8 position A3 in, and at the same time the harvester runs to Figure 8 position B3 in; at this time, the position error and the heading angle error , according to Step 4, the position error < position error threshold = 20 meters, the control system transfers to Step 6. According to Step 6, the position error < position error threshold = 3 meters and the heading angle error <Heading angle error threshold = , transfer to step 8, the system activates the stable tracking control mode. After that, the system has been in the stable tracking control mode. During the stable tracking control process, the maximum distance error is 0.97 meters, the average distance error is 0.19 meters, and the maximum heading angle error is , fully meeting the requirements of the technical specifications.

[0043] Example 2 (The working path of the harvester is straight line 2, rough) Refer to Figure 9 , the symbol ▶ is the harvester, the symbol ▷ is the unmanned grain carrier, and the trajectory of the unloading port of the harvester is the dotted line shown in the figure. The initial position of the unmanned grain carrier is A1 , and the initial heading angle is 90º. The harvester is at the initial position B1 , when the initial heading angle is 180º, a grain transportation signal is sent to the unmanned grain carrier. After receiving the signal, the unmanned grain carrier starts from the initial position A1 and approaches the harvester quickly - gets in place on the line - and tracks stably according to the method steps of the present invention.

[0044] Taking 0.1 second as the execution period, refer to Figure 9 The thin solid line in is the tracking trajectory of the unmanned grain carrier. The unmanned grain carrier receives the grain transportation signal at position A1, and the system activates the quick approach control mode.

[0045] In the quick approach control mode, when the unmanned grain carrier runs to position A2 in the figure and the harvester runs to position B2 in the figure, the system activates the get in place on the line control mode.

[0046] In the get in place on the line control mode, the unmanned grain carrier runs to position A3 in the figure, and at the same time the harvester runs to position B3 in the figure. At this time, the system activates the stable tracking control mode. After that, the system has been in the stable tracking control mode. During the stable tracking control process, the maximum distance error is 1.02 meters, the average distance error is 0.25 meters, and the maximum heading angle error is , fully meeting the requirements of the technical specifications.

[0047] Example 3 (The working path of the harvester is arc 1, detailed) Refer to Figure 10 , the symbol ▶ is the harvester, the symbol ▷ is the unmanned grain carrier, and the trajectory of the unloading port of the harvester is the dotted line shown in the figure. The initial position of the unmanned grain carrier is A1 , and the initial heading angle is 90º. The harvester is at the initial position B1 , when the initial heading angle is 180º, a grain transportation signal is sent to the unmanned grain transport vehicle. After receiving the signal, the unmanned grain transport vehicle starts from the initial position A1 and, according to the method steps of the present invention, quickly approaches - gets on line and positions - stably tracks the harvester.

[0048] Taking 0.1 second as the execution period, referring to Figure 10 The thin solid line in is the tracking trajectory of the unmanned grain transport vehicle. The unmanned grain transport vehicle receives the grain transportation signal at position A1. First, according to step 3, calculate the position error heading angle error , according to step 4, the position error > position error threshold = 20 meters, the control system transfers to step 5, and the system activates the quick approach control mode.

[0049] In the quick approach control mode, when the unmanned grain transport vehicle runs to position A2 in the figure, the harvester runs to position B2 in the figure. At this time, the position error , the heading angle error , according to step 4, the position error < position error threshold = 20 meters, the control system transfers to step 6. According to step 6, the position error > position error threshold = 3 meters and the heading angle error > heading angle error threshold = , the control system transfers to step 7, and the system activates the getting on line and positioning control mode.

[0050] In the getting on line and positioning control mode, the unmanned grain transport vehicle runs to Figure 10 position A3 in, and at the same time the harvester runs to position B3 in the figure; at this time, the position error and the heading angle error , according to step 4, the position error < position error threshold = 20 meters, the control system transfers to step 6. According to step 6, the position error < position error threshold = 3 meters and the heading angle error < heading angle error threshold = , transfer to step 8, and the system activates the stable tracking control mode. After that, the system has been in the stable tracking control mode. During the stable tracking control process, the maximum distance error is 1.07 meters, the average distance error is 0.33 meters, and the maximum heading angle error is , fully meeting the requirements of the technical specifications.

[0051] Example 4 (The working path of the harvester is arc 2, rough) Refer to Figure 11 , the symbol ▶ is the harvester, the symbol ▷ is the unmanned grain carrier, and the trajectory of the grain unloading port of the harvester is Figure 11 the dotted line shown in . The initial position of the unmanned grain carrier is A1 , and the initial heading angle is 90°. When the harvester is at the initial position B1

[0052] and the initial heading angle is 180°, a grain transportation signal is sent to the unmanned grain carrier. After receiving the signal, the unmanned grain carrier starts from the initial position A1 and performs rapid approach - on - line positioning - stable tracking on the harvester according to the method steps of the present invention. Figure 11 Taking 0.1 second as the execution period, refer to

[0053] the thin solid line in Figure 11 as the tracking trajectory of the unmanned grain carrier. The unmanned grain carrier receives the grain transportation signal at position A1, and the system activates the rapid approach control mode. Figure 11 When the unmanned grain carrier runs to

[0054] position A2 in Figure 11 in the rapid approach control mode, the harvester runs to Figure 11 position B2 in . At this time, the system activates the on - line positioning control mode.

[0055] When the unmanned grain carrier runs to Example 5 (The working path of the harvester is curve 1, detailed) Figure 12 Refer to Figure 12 , the symbol ▶ is the harvester, the symbol ▷ is the unmanned grain carrier, and the trajectory of the grain unloading port of the harvester is the dotted line shown in . The initial position of the unmanned grain carrier is A1

[0056] Taking 0.1 second as the execution period, refer to Figure 12The thin solid line in it is the tracking trajectory of the unmanned grain carrier. The unmanned grain carrier receives the grain transportation signal at position A1. First, according to step 3, calculate the position error between the harvester and the unmanned grain carrier Course angle error , according to step 4, the position error > Position error threshold = 20 meters, the control system transfers to step 5, and the system activates the fast approach control mode.

[0057] In the fast approach control mode, when the unmanned grain carrier runs to Figure 12 position A2 in it, the harvester runs to Figure 12 position B2 in it. At this time, the position error , the course angle error , according to step 4, the position error < Position error threshold = 20 meters, the control system transfers to step 6. According to step 6, the position error > Position error threshold = 3 meters and the course angle error > Course angle error threshold = , the control system transfers to step 7, and the system activates the on-line positioning control mode.

[0058] In the on-line positioning control mode, the unmanned grain carrier runs to Figure 12 position A3 in it, and at the same time the harvester runs to Figure 12 position B3 in it; at this time, the position error and the course angle error , according to step 4, the position error < Position error threshold = 20 meters, the control system transfers to step 6. According to step 6, the position error < Position error threshold = 3 meters and the course angle error < Course angle error threshold = , transfer to step 8, and the system activates the stable tracking control mode. After that, the system has been in the stable tracking control mode. During the stable tracking process, the maximum distance error is 1.20 meters, the average distance error is 0.65 meters, and the maximum course angle error is , fully meeting the requirements of the technical specifications.

[0059] Example 6 (The working path of the harvester is curve 2, rough) Refer to Figure 13 , the symbol ▶ is the harvester, the symbol ▷ is the unmanned grain carrier, and the trajectory of the unloading port of the harvester is Figure 13The dashed line shown in the figure. The initial position A1 of the unmanned grain carrier , and the initial heading angle is 90°. The harvester is at the initial position B1 . When the initial heading angle is 180°, a grain transportation signal is sent to the unmanned grain carrier. After receiving the signal, the unmanned grain carrier starts from the initial position A1 and approaches the harvester quickly - gets in place on the line - and tracks stably according to the method steps of the present invention.

[0060] Taking 0.1 second as the execution cycle, referring to Figure 13 the thin solid line in it as the tracking trajectory of the unmanned grain carrier. The unmanned grain carrier receives the grain transportation signal at position A1, and the system activates the fast approach control mode.

[0061] In the fast approach control mode, when the unmanned grain carrier runs to Figure 13 position A2 in it, the harvester runs to Figure 13 position B2 in it. At this time, the system activates the getting in place on the line control mode.

[0062] In the getting in place on the line control mode, the unmanned grain carrier runs to Figure 13 position A3 in it, and at the same time the harvester runs to Figure 13 position B3 in it. At this time, the system activates the stable tracking control mode. After that, the system is always in the stable tracking control mode. During the stable tracking control process, the maximum distance error is 1.21 meters, the average distance error is 0.49 meters, and the maximum heading angle error is , fully meeting the requirements of the technical specifications.

Claims

1. Unmanned grain transport vehicle tracking and control system, characterized by: The unmanned grain transport vehicle tracking and control system is installed on the unmanned grain transport vehicle, and the tracking and control system includes a data acquisition module, a control module, and an output module which are connected in sequence.

2. The unmanned grain transport vehicle tracking and control system according to claim 1 is characterized in that: The data acquisition module is used to obtain various operating states and input signals from the RTK-inertial navigation combined sensor, angle sensor and wireless communication module into the control module; The control module mainly includes a main processor, which executes a multi-mode control algorithm for tracking and controlling the unmanned grain transport vehicle, obtains the control quantity of the unmanned grain transport vehicle, including the front wheel turning angle and speed, and inputs these control quantities into the output module; The output module outputs the front wheel steering angle and speed of the unmanned grain truck calculated by the control algorithm of the main processor to the grain truck motion actuator through the CAN bus to control the driving of the grain truck.

3. A tracking and control method for an unmanned grain transport vehicle, using the unmanned grain transport vehicle tracking and control system according to claim 1 or 2, characterized in that: Follow these steps to implement: Step 1: Initialize the system; Step 2: The unmanned grain transport vehicle tracking control system obtains the operating status of the unmanned grain transport vehicle and the harvester; Step 3: Calculate the distance error and heading angle error between the unmanned grain transport vehicle and the harvester; Step 4: Determine whether the unmanned grain transport vehicle enters the rapid approach control mode; Step 5: The unmanned grain transport vehicle tracking control system executes a fast approach control mode; Step 6: Determine whether the unmanned grain transport vehicle has entered the online position control mode; Step 7: The unmanned grain transport vehicle tracking and control system executes the online in-place control mode; Step 8: The unmanned grain transport vehicle tracking control system executes a stable tracking control mode; Step 9: Output the control variables of the unmanned grain transport vehicle to realize the unloading of grain by the unmanned grain transport vehicle.

4. The tracking control method of the unmanned grain transport vehicle according to claim 3 is characterized in that: In step 1, the specific process is: the unmanned grain transport vehicle tracking control system is turned on and the relevant parameters are initialized, including the following parameters: 1.1) Parameters of unmanned grain transport vehicles and harvesters: Front and rear wheel width of harvester , the maximum safe driving speed of unmanned grain transport vehicles , the upper and lower limits of the front wheel turning angle of the unmanned grain transport vehicle , front and rear wheel width of unmanned grain transport vehicle , the minimum turning radius of unmanned grain transport vehicles ; 1.2) Unmanned grain transport vehicle control mode switching parameters: distance error threshold , heading angle error threshold , used to determine whether to switch to stable tracking control mode; distance error threshold 2 , used to determine whether to switch to the fast approach control mode; 1.3) PID controller parameters for unmanned grain truck rapid approach control: proportional coefficient , integral coefficient and the differential coefficient , forward viewing distance ; 1.4) PID controller parameters for the on-line control of unmanned grain transport vehicles: proportional coefficient , integral coefficient and the differential coefficient , forward viewing distance ; 1.5) MPC controller parameters for stable tracking control of unmanned grain transport vehicles: upper and lower bounds of the front wheel steering angle , speed upper and lower bounds , controller control step size And the number of prediction steps , controller control variable weight matrix and the state variable weight matrix .

5. The tracking control method of the unmanned grain transport vehicle according to claim 3 is characterized in that: In step 2, the specific process is: 2.1) Operation status of the unmanned grain transport vehicle: Current location of the unmanned grain transport vehicle , the current heading angle of the unmanned grain transport vehicle , the current speed of the unmanned grain transport vehicle , the current front wheel turning angle of the unmanned grain transport vehicle ; The current position, heading angle and speed of the unmanned grain transport vehicle are acquired through the RTK-inertial navigation combined sensor; the current front wheel turning angle of the unmanned grain transport vehicle is acquired through the angle sensor; 2.2) Harvester operation status: Current position of the harvester unloading port , harvester current heading angle , the current speed of the harvester , harvester's current front wheel angle ; The current position, heading angle, speed and front wheel turning angle of the harvester are obtained through the wireless communication module.

6. The tracking control method of the unmanned grain transport vehicle according to claim 3 is characterized in that: In step 5, the specific process is: 5.1) Plan the shortest path, The current speed of the harvester is a vector , vector The direction of the harvester is the current heading of the vector The model is the current speed of the harvester. ; The current speed of the unmanned grain transport vehicle is vector , vector The direction is the current heading of the unmanned grain transport vehicle, and the vector The model is the current speed of the unmanned grain transport vehicle. ; In order for the unmanned grain transporter to get close to the harvester as soon as possible, take the current position of the unmanned grain transporter Current location of harvester unloading port The straight line segment between them is the shortest path ; 5.2) Determine the preview point , If the current location of the unmanned grain truck ( x s ，y s ) and the current position of the harvester unloading port The distance between them is less than the foresight distance ,but Preview Point ; Otherwise, the current position of the unmanned grain transport vehicle ( x s ，y s ) is the origin, and the planned path Up front sight distance Match the preview point forward ; 5.3) Calculation of lateral error , Calculate preview point Current location of unmanned grain transport vehicles ( x s ，y s ) , the calculation formula is: , Calculate the current position of the unmanned grain transport vehicle Point to preview point The vector pointing angle , the calculation formula is: ， Therefore, the lateral error The calculation formula is: ; 5.4) Calculation of control variables of unmanned grain transport vehicle using PID algorithm , Through the lateral error Update the three current error components , , , the calculation formula is: ， ， ， in and The historical value of the error component calculated for the last control cycle; If this is the first control cycle, then and The default value is 0. Update error component history value and And save it for the next control cycle. The calculation formula is: , , Through the lateral error Combined with the proportional coefficient , integral coefficient and the differential coefficient Linear combination of three parts to calculate the control variables of the unmanned grain transport vehicle , the calculation formula is: ； 5.5) Output control variables: Output the control variables of the unmanned grain transport vehicle, including the front wheel turning angle of the unmanned grain transport vehicle. , Speed ​​of unmanned grain transport vehicles , proceed to step 9.

7. The tracking and control method of the unmanned grain transport vehicle according to claim 3 is characterized in that: In step 6, the specific process is: The unmanned grain transport vehicle tracking control system determines whether the unmanned grain transport vehicle enters the online position control mode if and , the unmanned grain transport vehicle tracking and control system executes step 8 and switches to the stable tracking control mode; otherwise, the unmanned grain transport vehicle tracking and control system executes step 7 and switches to the online in-place control mode.

8. The tracking control method of the unmanned grain transport vehicle according to claim 3 is characterized in that: In step 7, the specific process is: 7.1) Plan the online path: The current speed of the harvester is a vector , vector The direction of the harvester is the current heading of the vector The model is the current speed of the harvester. ; The current speed of the unmanned grain transport vehicle is vector , vector The direction is the current heading of the unmanned grain transport vehicle, and the vector The model is the current speed of the unmanned grain transport vehicle. ; 7. 2) Calculate the control variables of the unmanned grain transport vehicle , In order to make the unmanned grain transport vehicle and the harvester reach the end of the path at the same time, calculate the target speed of the unmanned grain transport vehicle during the process of going online , the calculation formula is: ， 7.3) Determine the preview point , If the unmanned grain truck is currently located ( x s ,y s ) and the current position of the harvester unloading port The distance between them is less than the foresight distance ,but Preview Point ; Otherwise, the current position of the unmanned grain transport vehicle ( x s ,y s ) is the origin, and the planned path Up front sight distance Match the preview point forward ; 7.4) Calculation of lateral error , Calculate preview point Current location with unmanned grain transport vehicles ( x s ，y s ) , the calculation formula is: , Calculate the current position of the unmanned grain transport vehicle Point to preview point The vector pointing angle , the calculation formula is: ， Therefore, the lateral error The calculation formula is: , in is the current heading angle of the unmanned grain transport vehicle; 7.5) Calculation of control variables of unmanned grain transport vehicle using PID algorithm , Through the lateral error Update the three current error components , , , the calculation formula is: ， ， ， in and The historical value of the error component calculated for the last control cycle; If this is the first control cycle, then and The default value is 0. Update error component history value and And save it for the next control cycle. The calculation formula is: , , Through the lateral error Combined with the proportional coefficient , integral coefficient and the differential coefficient Linear combination of three parts to calculate the control variables of the unmanned grain transport vehicle , the calculation formula is: ， 7.6) Output control variables: The output control variables of the unmanned grain transport vehicle include: the front wheel turning angle of the unmanned grain transport vehicle is , speed of unmanned grain transport vehicle , proceed to step 9.

9. The tracking control method of the unmanned grain transport vehicle according to claim 3, characterized in that: In step 8, the specific process is: This step designs a tracking controller based on the model predictive control MPC method, which is called a model predictive control MPC controller. The settings of the model predictive control MPC controller are as follows: 8.1) Build a tracking prediction model for unmanned grain transport vehicles. The front wheel angle and speed of the unmanned grain transport vehicle are used as the control input, and the error between the posture state of the unmanned grain transport vehicle and the posture state of the harvester unloading port is used as the control input. is the state variable, the first set of control variables The heading angle of the unmanned grain transport vehicle With speed , the second set of control variables is the heading angle of the harvester With speed , then the state space equation of the unmanned grain transport vehicle tracking prediction model is as follows: ， Linearize the above state space equations and The time is set to the current time, and the transformation is to obtain the new linear state space equation: , , , , ， ， ， , ， ， ， in, , and They are the reference state of state variables, the reference state of control variables of unmanned grain transport vehicles, and the reference state of control variables of harvesters; Setting the reference state All elements in are set to 0; the reference state of the front wheel angle of the unmanned grain transport vehicle Set to the current front wheel angle of the harvester , unmanned grain transport vehicle speed reference state Set to the current speed of the harvester ; Reference state of the harvester's front wheel angle Set to the current front wheel angle of the harvester , Harvester speed reference state Set to the current speed of the harvester ; Then, based on the above linear state space equation, the control step size of the MPC controller is , the discrete state space equation of the unmanned grain transport vehicle tracking prediction model is obtained as follows: ， in, ， ， ， ， in, The value corresponding to the variable from the current moment k Start Experience At the same time, based on the above discrete state space equation, combined with the model predictive control MPC controller prediction step number , the matrix equation of the unmanned grain transport vehicle tracking prediction model is as follows: ， ， ， ， ， ， ， Assuming that the front wheel turning angle and speed of the harvester remain unchanged in the control time domain, we have: ， Then the above equation is simplified to: ; 8.2) Optimization solution, Combined with the model predictive control MPC controller control variable weight matrix and the state variable weight matrix , construct an optimization problem, and the constructed objective function is: ， in, ， ， ， ， Add control variable constraints of the model predictive control MPC controller: upper and lower bounds of the front wheel angle , speed upper and lower bounds , then the expression of the entire multi-objective optimization problem is transformed into: ， This optimization problem only has control variable increment constraints, but no state variable constraints. It is solved efficiently using a mature QP solver to obtain the optimal control variable increment sequence. ; 8.3) By Calculating Control Variables for Unmanned Grain Transport Vehicles and , from Extract the current time Corresponding control increment , update the current control amount , get the target front wheel turning angle of the unmanned grain transport vehicle With target speed , the expression is as follows: = ， = ， in and The extraction formula is: ， ； 8.4) Output control variables: The output control variables of the unmanned grain transport vehicle include: the front wheel turning angle of the unmanned grain transport vehicle is , speed of unmanned grain transport vehicle , proceed to step 9.

10. The tracking control method of the unmanned grain transport vehicle according to claim 3, characterized in that: In step 9, the specific process is: the unmanned grain transport vehicle tracking and control system sends the control variables of the unmanned grain transport vehicle output by different control modes to the unmanned grain transport vehicle actuator, so that the unmanned grain transport vehicle adjusts the driving state according to the control information and completes a task cycle.

Images