Vision detection based combine track passability control method

Abstract

The present application relates to the field of intelligent control technology of agricultural machinery, in particular to a combined harvester track passability control method based on visual detection, comprising the following steps: S1: predicting a motion trajectory based on the real-time running speed and steering angle of the harvester; S2: extracting the real-time terrain slope angle of the trajectory area; S3: determining the target ground pressure distribution mode required by the track in the predicted trajectory area; S4: calculating the target hydraulic flow distribution ratio for controlling the walking motor; S5: controlling the hydraulic valve group to adjust the oil supply amount according to the target flow distribution ratio. The present application realizes the segmented adaptive adjustment of the track ground pressure through the linkage of visual perception and hydraulic regulation, effectively improving the passability and operation stability of the combined harvester under complex terrain.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for agricultural machinery, and in particular to a method for controlling the track passability of combine harvesters based on vision detection. Background Technology

[0002] Combine harvesters, as the main agricultural harvesting equipment, are widely used in the mechanized harvesting of field crops. Since they usually need to travel on unpaved and uneven farmland terrain, especially in areas with obvious slopes or localized soft soil, the ground contact status of the tracks directly affects the stability and operating efficiency of the entire machine. Existing combine harvester track systems mainly rely on static counterweights and fixed hydraulic drive parameters for control, lacking the ability to respond to dynamic terrain in real time. This often leads to problems such as track slippage, uneven ground contact, and wasted drive power when operating on slopes. In severe cases, it can even cause the machine to tip over or get stuck, posing a safety hazard.

[0003] Currently, there is a lack of a track pressure control method that combines terrain perception with an active adjustment mechanism. This makes it impossible to intelligently adjust the ground pressure distribution of different track sections according to varying terrain slopes, thus failing to achieve precise drive control. Traditional methods do not fully utilize visual sensors to acquire terrain information ahead, nor do they integrate visual information with the hydraulic system control to form a closed-loop response mechanism. Therefore, a vision-based track passability control method for combine harvesters is urgently needed to address these issues. Summary of the Invention

[0004] To achieve the above objectives, this invention provides a vision-based detection-based method for controlling the track passability of combine harvesters, in order to solve the problem of difficulty in accurately identifying complex farmland terrain.

[0005] A vision-based detection-based method for controlling the track mobility of combine harvesters includes the following steps:

[0006] S1: Based on the harvester’s real-time travel speed and turning angle, calculate the predicted movement trajectory of the track contact area within a preset time period in the future;

[0007] S2: Use the front-facing vision sensor to collect terrain images along the predicted motion trajectory, and combine the aircraft pitch angle data to extract the real-time terrain slope angle of the trajectory area;

[0008] S3: Based on the real-time terrain slope angle, query the preset slope-pressure mapping relationship table to determine the target ground pressure distribution pattern required by the track in the predicted trajectory area;

[0009] S4: Calculate the target hydraulic flow distribution ratio for controlling the walking motor based on the target ground pressure distribution pattern;

[0010] S5: Controls the hydraulic valve group to adjust the oil supply according to the target flow distribution ratio, so that when the track enters the predicted movement trajectory area, a pressure distribution adapted to the terrain slope is formed.

[0011] Optionally, S1 specifically includes:

[0012] S11: By acquiring the real-time travel speed and current steering angle of the combine harvester, establish the kinematic input parameters for the vehicle's travel trajectory;

[0013] S12: Set the prediction time period and divide the prediction time period into several equally spaced steps to form a discrete time series;

[0014] S13: Based on the uniform circular arc driving model, calculate the two-dimensional position coordinates of the vehicle's center of mass at each time step to form the vehicle's center of mass trajectory;

[0015] S14: Based on the coordinates of each centroid position and the fixed geometric offset parameter of the track center relative to the centroid, the positions of the left and right track grounding centers are deduced, ultimately forming the complete grounding trajectory of the track grounding area within the predicted time period.

[0016] Optionally, S2 specifically includes:

[0017] S21: Based on the track contact area predicted by S1, the corresponding spatial scanning area is selected within the field of view of the vision sensor, and the two-dimensional terrain image of the area is continuously captured by the front vision sensor to form a real-time image sequence.

[0018] S22: Using image stitching technology, adjacent images in a real-time image sequence are spatially overlapped and stitched together to generate a complete topographic map covering the predicted motion trajectory area;

[0019] S23: Use stereo vision technology to extract spatial three-dimensional coordinate data from a complete topographic map and establish a three-dimensional terrain model of the predicted trajectory area;

[0020] S24: Acquire real-time pitch angle data from the combine harvester's body attitude sensor, and use the pitch angle data to perform coordinate correction on the three-dimensional terrain model to eliminate the error in terrain slope measurement caused by the body tilt;

[0021] S25: Perform point-by-point slope calculation on the corrected 3D terrain model and extract the real-time terrain slope angle data of each location point within the predicted trajectory area.

[0022] Optionally, S24 specifically includes:

[0023] S241: Obtain the pitch angle data of the combine harvester in real time from the body attitude sensor to determine the tilt angle of the vehicle body relative to the horizontal reference plane;

[0024] S242: Based on the obtained pitch angle data, construct a three-dimensional rotation transformation matrix around the vehicle's lateral axis;

[0025] S243: Use the rotation transformation matrix to transform all coordinate points of the three-dimensional terrain model, so that the terrain model is restored to a coordinate system with the horizontal reference plane as the reference.

[0026] Optionally, S25 specifically includes:

[0027] S251: Based on the three-dimensional terrain model corrected by S24, determine the spatial coordinate data of each location point within the predicted trajectory area according to the vehicle's forward direction, and construct a local neighborhood centered on the location point.

[0028] S252: Perform least squares plane fitting on the local neighborhood of each location point to obtain the local fitted plane equation of the corresponding location point, which is used to represent the terrain surface of that location point;

[0029] S253: Calculate the terrain slope angle at the corresponding location point based on the fitted plane normal vector. The formula is:

[0030] ,in, The slope angle of the terrain at the corresponding location point; The coefficients are those of the fitted plane;

[0031] S254: Execute S252 and S253 point by point to calculate and extract real-time terrain slope angle data for all locations within the predicted trajectory area.

[0032] Optionally, S3 specifically includes:

[0033] S31: Based on the real-time terrain slope angle data of each location point within the predicted trajectory area obtained by S2, the slope angle values ​​are discretized one by one to obtain the slope discretization level corresponding to each location point.

[0034] S32: Based on the discrete slope level, call the pre-stored slope-pressure mapping table to extract the target ground pressure weight ratio values ​​of the front, middle and rear sections of the track corresponding to each discrete slope level;

[0035] S33: Based on the target ground pressure weight ratio values ​​of the front, middle and rear sections of the track corresponding to each location point in the predicted trajectory area, construct the target ground pressure distribution pattern curves of the three sections of the track respectively.

[0036] S34: Based on the smoothed target ground pressure distribution pattern curves of the front, middle and rear sections, determine the pressure distribution pattern of the track that is ultimately suitable for terrain slope changes within the predicted trajectory area.

[0037] Optionally, S34 specifically includes:

[0038] S341: Perform sliding window smoothing on the target ground pressure distribution curves of the front, middle, and rear sections of the track generated in S33 to remove local drastic changes and obtain three continuous pressure weight curve sequences, denoted as follows: and ;

[0039] S342: Calculate the weighted average value of the three weighted curves over the entire predicted trajectory region. , and , used to characterize the global pressure distribution trend;

[0040] S343: Yes , and Normalization is performed to ensure that the sum of the three is 1, which serves as the ground pressure distribution pattern of the track in the predicted trajectory area to adapt to changes in terrain slope.

[0041] Optionally, S4 specifically includes:

[0042] S41: Based on the target ground pressure weight values ​​of the front, middle and rear sections of the track determined in S3, and combined with the rated hydraulic flow parameters of each section of the track, calculate the initial hydraulic flow required for each section.

[0043] S42: Calculate the proportion of the initial hydraulic flow in the front, middle and rear sections to determine the proportion of each section's hydraulic flow in the total oil supply flow, and obtain the corresponding target flow distribution ratio.

[0044] S43: Based on the target hydraulic flow distribution ratio of each segment, output the corresponding travel motor control parameters for the front, middle and rear sections of the track.

[0045] Optionally, the formula for calculating the target flow allocation ratio in S42 is:

[0046] ,in, The target hydraulic flow rate for each section of the track; The target ground pressure weight values ​​corresponding to each section of the track; The hydraulic pump supplies the total hydraulic flow to the track drive system.

[0047] Optionally, S5 specifically includes:

[0048] S51: Receives the target hydraulic flow rate corresponding to the front, middle and rear travel motors of the track output from S4, and uses it as the target value for flow control of the hydraulic system.

[0049] S52: Based on the pre-calibrated flow-opening degree relationship function, the target hydraulic flow rate of each track segment is converted into the control opening degree command of the corresponding hydraulic valve group. The expression of the relationship function is:

[0050] ,in, The target opening degree for the hydraulic valve assemblies of each section of the track; The target hydraulic flow rate for each section of the track; It is the inverse function of the hydraulic flow rate-opening calibration function, used to convert the target flow rate into the valve orifice control quantity;

[0051] S53: The converted opening degree command is sent to the corresponding track section hydraulic valve group in real time to drive the valve group to perform dynamic adjustment.

[0052] The beneficial effects of this invention are:

[0053] This invention constructs a tracked predicted motion trajectory based on travel speed and steering angle, combines terrain images acquired by a front-mounted vision sensor with fuselage pitch angle data, establishes a three-dimensional terrain model, and calculates the real-time terrain slope angle. This enables accurate identification of complex farmland terrain and provides a forward-looking input basis for tracked grounding control.

[0054] This invention clarifies the target ground pressure weight ratio of the front, middle and rear sections of the track by using a slope-pressure mapping relationship, and achieves real-time fine adjustment of the hydraulic system through a mapping model of hydraulic flow and valve opening degree. This ensures that the track has a matching ground pressure distribution under different terrain conditions, significantly improving the passability, stability and operation continuity of the combine harvester. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 This is a schematic diagram of a combine harvester track passability control method according to an embodiment of the present invention;

[0057] Figure 2 This is a schematic diagram of the process for extracting real-time terrain slope angles according to an embodiment of the present invention. Detailed Implementation

[0058] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0059] It should be noted that the use of terms such as "an embodiment," "an embodiment," "an exemplary embodiment," and "some embodiments" in the specification indicates that the described embodiment may include a specific feature, structure, or characteristic, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, when a specific feature, structure, or characteristic is described in connection with an embodiment, implementing such a feature, structure, or characteristic in conjunction with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the art.

[0060] Generally, terms can be understood at least partly from their use in context. For example, depending at least partly on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in a singular sense, or a combination of features, structures, or characteristics in a plural sense. Additionally, the term "based on" can be understood not necessarily to convey an exclusive set of factors, but rather, alternatively, depending at least partly on the context, to allow for the presence of other factors that are not necessarily explicitly described.

[0061] like Figures 1-2 As shown, the vision-based detection-based method for controlling the track mobility of a combine harvester includes the following steps:

[0062] S1: Based on the harvester’s real-time travel speed and turning angle, calculate the predicted movement trajectory of the track contact area within a preset time period in the future;

[0063] S2: Use the front-facing vision sensor to collect terrain images along the predicted motion trajectory, and combine the aircraft pitch angle data to extract the real-time terrain slope angle of the trajectory area;

[0064] S3: Based on the real-time terrain slope angle, query the preset slope-pressure mapping relationship table to determine the target ground pressure distribution pattern required by the track within the predicted trajectory area;

[0065] S4: Calculate the target hydraulic flow distribution ratio for controlling the walking motor based on the target ground pressure distribution pattern;

[0066] S5: Controls the hydraulic valve group to adjust the oil supply according to the target flow distribution ratio, so that when the track enters the predicted movement trajectory area, a pressure distribution adapted to the terrain slope is formed.

[0067] S1 specifically includes:

[0068] S11: By acquiring the real-time travel speed and current steering angle of the combine harvester, establish the kinematic input parameters of the vehicle's travel trajectory to initialize the trajectory prediction model. The travel speed is the linear velocity at the vehicle's center of mass, and the steering angle is the deflection angle of the track relative to the longitudinal direction of the vehicle body.

[0069] S12: Set the prediction time period and divide it into several equally spaced steps to form a discrete time series. This facilitates subsequent point-by-point calculation of the spatial displacement trajectory of the track center. The prediction time series can be represented as follows: ;

[0070] ;in, For the first The time point corresponding to each time step; The time step interval; To predict the total duration; The number of prediction steps is a positive integer.

[0071] S13: Based on the uniform circular arc driving model, calculate the two-dimensional position coordinates of the vehicle's center of mass at each time step to form the vehicle's center of mass trajectory; the calculation formula is as follows:

[0072] ,in, For the center of mass at the th Spatial location coordinates of each time step; The initial coordinates of the vehicle's center of mass at the current moment; The speed at the vehicle's center of gravity; The radius of curvature for the vehicle's travel is calculated using the following formula: ,in, This refers to the vehicle's wheelbase. Current steering angle;

[0073] S14: Based on the coordinates of each centroid position and the fixed geometric offset parameter of the track center relative to the centroid, the positions of the left and right track ground contact centers are deduced, ultimately forming the complete ground contact trajectory of the track contact area within the future prediction time period. The above steps, by introducing a curvature motion model driven by vehicle speed and steering angle, can accurately calculate the spatial trajectory points of the track contact area within multiple time domain steps, providing a stable and continuous prediction basis for terrain slope identification and ground pressure distribution optimization, and improving the foresight and accuracy of track passability control.

[0074] S2 specifically includes:

[0075] S21: Based on the track contact area predicted by S1, the corresponding spatial scanning area is selected within the field of view of the vision sensor, and the two-dimensional terrain image of the area is continuously captured by the front vision sensor to form a real-time image sequence.

[0076] S22: Using image stitching technology, adjacent images in a real-time image sequence are spatially overlapped and stitched together to generate a high-resolution complete topographic map covering the predicted motion trajectory area;

[0077] S23: Use stereo vision technology to extract spatial three-dimensional coordinate data from a complete topographic map and establish a three-dimensional terrain model of the predicted trajectory area;

[0078] S24: Acquire real-time pitch angle data from the combine harvester's body attitude sensor, and use the pitch angle data to perform coordinate correction on the three-dimensional terrain model to eliminate the error in terrain slope measurement caused by the body tilt;

[0079] S25: Perform point-by-point slope calculation on the corrected 3D terrain model, extract the real-time terrain slope angle data of each location point in the predicted trajectory area, and use it as the input for subsequent control of track ground pressure; the above steps achieve high-precision 3D reconstruction of the terrain in the trajectory area through stereo vision and image stitching technology, and combine the real-time correction of the terrain model with the fuselage pitch angle, effectively improving the accuracy and reliability of terrain slope measurement, and providing accurate data support for the dynamic control of track passability.

[0080] S24 specifically includes:

[0081] S241: Obtain the pitch angle data of the combine harvester in real time from the body attitude sensor to determine the tilt angle of the vehicle body relative to the horizontal reference plane;

[0082] S242: Based on the obtained pitch angle data, construct a three-dimensional rotation transformation matrix around the vehicle's lateral axis to achieve coordinate correction of the three-dimensional terrain model;

[0083] S243: Use a rotation transformation matrix to transform all coordinate points of the 3D terrain model, so that the terrain model is restored to a coordinate system with the horizontal reference plane as the reference.

[0084] The expression for the three-dimensional rotation transformation matrix is ​​as follows:

[0085] ;in, Represents the rotational transformation matrix about the vehicle's lateral axis; The above steps involve real-time fuselage pitch angle correction using a three-dimensional rotation matrix, effectively eliminating measurement errors caused by vehicle tilt, ensuring consistency between the three-dimensional terrain model and the actual ground conditions, thereby improving the accuracy and reliability of terrain slope angle data, and providing high-precision data for subsequent track ground pressure distribution.

[0086] S25 specifically includes:

[0087] S251: Based on the three-dimensional terrain model corrected by S24, determine the spatial coordinate data of each location point within the predicted trajectory area according to the vehicle's forward direction, and construct a local neighborhood centered on the location point.

[0088] S252: Perform least-squares plane fitting on the local neighborhood of each location point to obtain the local fitted plane equation for that location point, which represents the terrain surface at that location point; the least-squares plane fitting calculation formula is: In the formula, each parameter satisfies the following minimization objective:

[0089] ,in, These are the th elements in the neighborhood of the location point. The three-dimensional coordinates of each point; The coefficients of the fitted plane equation; This represents the total number of points used for fitting within the neighborhood; This is the sum of squared errors for fitting the plane equation;

[0090] S253: Calculate the terrain slope angle at the corresponding location point based on the fitted plane normal vector. The formula is:

[0091] ,in, The slope angle of the terrain at the corresponding location point; The coefficients are those of the fitted plane;

[0092] S254: Execute S252 and S253 point by point to calculate and extract real-time terrain slope angle data for all locations within the predicted trajectory area. The above steps accurately calculate the real-time terrain slope angle for each location within the predicted area through local plane fitting and normal vector analysis, ensuring high accuracy and detail capture capability of terrain analysis. This provides accurate and continuous terrain input data for the track pressure control of combine harvesters, effectively enhancing adaptability to complex terrain.

[0093] S3 specifically includes:

[0094] S31: Based on the real-time terrain slope angle data of each location point within the predicted trajectory area obtained by S2, the slope angle values ​​are discretized one by one to obtain the slope discretization level corresponding to each location point.

[0095] S32: Based on the discrete slope level, call the pre-stored slope-pressure mapping table to extract the target ground pressure weight ratio values ​​of the front, middle and rear sections of the track corresponding to each discrete slope level;

[0096] Table 1. Slope-Pressure Mapping Relationship

[0097]

[0098] In Table 1 above, the slope angle interval represents the range of terrain slope angles where the track is predicted to be in contact with the ground, in degrees, and is divided into several discrete levels based on the calculation results of S2 and S25; the front pressure weight represents the proportion of target ground pressure that the front section of the track should bear under the corresponding slope angle; the middle pressure weight represents the proportion of target ground pressure that the middle section of the track should bear under the corresponding slope angle; and the rear pressure weight represents the proportion of target ground pressure that the rear section of the track should bear under the corresponding slope angle.

[0099] S33: Based on the target ground pressure weight ratio values ​​of the front, middle and rear sections of the track corresponding to each location point in the predicted trajectory area, construct the target ground pressure distribution pattern curves of the three sections of the track respectively to eliminate pressure abrupt changes between adjacent points.

[0100] S34: Based on the smoothed target ground pressure distribution pattern curves of the front, middle and rear sections, determine the pressure distribution pattern of the track that is finally suitable for terrain slope changes in the predicted trajectory area. The pressure distribution pattern is explicitly given in the form of pressure weight ratio of the front, middle and rear sections of the track, which serves as the input basis for subsequent travel motor flow control.

[0101] S34 specifically includes:

[0102] S341: Perform sliding window smoothing on the target ground pressure distribution curves of the front, middle, and rear sections of the track generated in S33 to remove local drastic changes and obtain three continuous pressure weight curve sequences, denoted as follows: and ,in Indicates the position index along the predicted trajectory;

[0103] S342: Calculate the weighted average value of the three weighted curves over the entire predicted trajectory region. , and This is used to characterize the global pressure distribution trend, and the calculation formula is as follows:

[0104] ;

[0105] ;

[0106] ,in, These are the final pressure weight values ​​for the front, middle, and rear sections of the track, respectively. The first The pressure weight value of the corresponding segment at each trajectory position; This represents the total number of location points on the trajectory.

[0107] S343: Yes Normalization is performed to ensure that the sum of the three factors is 1, which serves as the ground pressure distribution pattern of the track in the predicted trajectory area to adapt to changes in terrain slope. This pattern is then used to drive the subsequent flow ratio regulation of the travel motor. The above steps, through weighted averaging and normalization of the three pressure curves, can extract a stable pressure distribution strategy under the overall slope change trend, avoid hypersensitive responses to local abnormal slopes, and improve the stability and energy consumption control capability of the track control system during long-distance travel.

[0108] S4 specifically includes:

[0109] S41: Based on the target ground pressure weight values ​​of the front, middle and rear sections of the track determined in S3, and combined with the rated hydraulic flow parameters of each section of the track, calculate the initial hydraulic flow required for each section.

[0110] The formula for calculating the initial hydraulic flow rate is: ,in, This represents the initial hydraulic flow rate for each section of the track. The target ground pressure weight values ​​for each track section are: front section pressure weight, etc. Mid-section pressure weight Post-segment pressure weight ; These are the rated hydraulic flow parameters for each track section, i.e., the hydraulic flow required by each track section under rated operating conditions;

[0111] S42: Calculate the proportion of the initial hydraulic flow in the front, middle and rear sections to determine the proportion of each section's hydraulic flow in the total oil supply flow, and obtain the corresponding target flow distribution ratio.

[0112] S43: Based on the target hydraulic flow distribution ratio of each segment, output the corresponding travel motor control parameters for the front, middle and rear sections of the track for subsequent precise adjustment of the hydraulic valve group; the above steps accurately determine the target flow distribution ratio of each track segment through direct correlation calculation between pressure weight value and hydraulic flow parameter, ensuring effective matching between track ground pressure and actual hydraulic driving force, realizing active and precise control of track contact with terrain, and improving the combine harvester's driving stability and energy utilization efficiency in complex terrain.

[0113] The formula for calculating the target flow allocation ratio in S42 is:

[0114] ,in, The target hydraulic flow rate for each section of the track, in liters per minute; The target ground pressure weight values ​​corresponding to each section of the track; The total hydraulic flow supplied to the track drive system by the hydraulic pump, measured in liters per minute.

[0115] S5 specifically includes:

[0116] S51: Receives the target hydraulic flow rate corresponding to the front, middle and rear travel motors of the track output from S4, and uses it as the target value for flow control of the hydraulic system.

[0117] S52: Based on the pre-calibrated flow-opening degree relationship function, the target hydraulic flow rate of each track segment is converted into the control opening degree command of the corresponding hydraulic valve group. The expression of the relationship function is:

[0118] ,in, The target opening degree for the hydraulic valve assemblies of each section of the track; The target hydraulic flow rate for each section of the track; It is the inverse function of the hydraulic flow rate-opening calibration function, used to convert the target flow rate into the valve orifice control quantity;

[0119] S53: The converted opening degree command is sent to the corresponding track section hydraulic valve group in real time, driving the valve group to perform dynamic adjustment so that each track section has hydraulic driving force matching the target ground pressure when entering the predicted trajectory area; the above steps improve the real-time performance and accuracy of track hydraulic adjustment by using the obtained target hydraulic flow to perform valve group opening degree conversion control, and enhance the equipment's immediate response capability to complex terrain slope changes.

[0120] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0121] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for controlling the track passability of a combine harvester based on vision detection, characterized in that, Includes the following steps: S1: Based on the harvester’s real-time travel speed and turning angle, calculate the predicted movement trajectory of the track contact area within a preset time period in the future; S2: Use the front-facing vision sensor to collect terrain images along the predicted motion trajectory, and combine the aircraft pitch angle data to extract the real-time terrain slope angle of the predicted motion trajectory area; S3: Based on the real-time terrain slope angle, query the preset slope-pressure mapping table to determine the target ground pressure distribution pattern required by the track within the predicted motion trajectory area; S4: Calculate the target hydraulic flow distribution ratio for controlling the walking motor based on the target ground pressure distribution pattern; S5: Controls the hydraulic valve group to adjust the oil supply according to the target flow distribution ratio, so that when the track enters the predicted motion trajectory area, a pressure distribution adapted to the terrain slope is formed. S1 specifically includes: S11: By acquiring the real-time travel speed and current steering angle of the combine harvester, establish the kinematic input parameters for the vehicle's travel trajectory; S12: Set the prediction time period and divide the prediction time period into several equally spaced steps to form a discrete time series; S13: Based on the uniform circular arc driving model, calculate the two-dimensional position coordinates of the vehicle's center of mass at each time step to form the vehicle's center of mass trajectory; S14: Based on the coordinates of each centroid position and the fixed geometric offset parameter of the track center relative to the centroid, the positions of the left and right track grounding centers are deduced, and finally the complete grounding trajectory of the track grounding area in the future predicted time period is formed. S2 specifically includes: S21: Based on the track contact area predicted by S1, the corresponding spatial scanning area is selected within the field of view of the vision sensor, and the two-dimensional terrain image of the spatial scanning area is continuously captured by the front vision sensor to form a real-time image sequence. S22: Using image stitching technology, adjacent images in a real-time image sequence are spatially overlapped and stitched together to generate a complete topographic map covering the predicted motion trajectory area; S23: Use stereo vision technology to extract spatial three-dimensional coordinate data from a complete topographic map and establish a three-dimensional terrain model of the predicted movement trajectory area; S24: Acquire real-time pitch angle data from the combine harvester's body attitude sensor, and use the pitch angle data to perform coordinate correction on the three-dimensional terrain model to eliminate the error in terrain slope measurement caused by the body tilt; S25: Perform point-by-point slope calculation on the corrected 3D terrain model and extract the real-time terrain slope angle data of each location point within the predicted motion trajectory area; S24 specifically includes: S241: Obtain the pitch angle data of the combine harvester in real time from the body attitude sensor to determine the tilt angle of the vehicle body relative to the horizontal reference plane; S242: Based on the obtained pitch angle data, construct a three-dimensional rotation transformation matrix around the vehicle's lateral axis; S243: Use the three-dimensional rotation transformation matrix to transform all coordinate points of the three-dimensional terrain model, so that the terrain model is restored to a coordinate system with the horizontal reference plane as the reference. Specifically, S25 includes: S251: Based on the three-dimensional terrain model corrected by S24, determine the spatial coordinate data of each location point within the predicted motion trajectory area according to the vehicle's forward direction, and construct a local neighborhood centered on the location point. S252: Perform least squares plane fitting on the local neighborhood of each location point to obtain the local fitted plane equation of the corresponding location point, which is used to represent the terrain surface of that location point; S253: Calculate the terrain slope angle at the corresponding location point based on the fitted plane normal vector. The formula is: ,in, The slope angle of the terrain at the corresponding location point; The coefficients are those of the fitted plane; S254: Execute S252 and S253 in a loop to calculate and extract real-time terrain slope angle data for all locations within the predicted motion trajectory area; S3 specifically includes: S31: Based on the real-time terrain slope angle data of each location point within the predicted motion trajectory area obtained by S2, the slope angle values ​​are discretized one by one to obtain the slope discretization level corresponding to each location point. S32: Based on the discrete slope level, call the pre-stored slope-pressure mapping table to extract the target ground pressure weight ratio values ​​of the front, middle and rear sections of the track corresponding to each discrete slope level; S33: Based on the target ground pressure weight ratio values ​​of the front, middle and rear sections of the track corresponding to each location point in the predicted motion trajectory area, construct the target ground pressure distribution pattern curves of the three sections of the track respectively. S34: Based on the smoothed target ground pressure distribution pattern curves of the front, middle and rear sections, determine the pressure distribution pattern of the track that is ultimately suitable for terrain slope changes within the predicted motion trajectory area; S34 specifically includes: S341: Perform sliding window smoothing on the target ground pressure distribution curves of the front, middle, and rear sections of the track generated in S33 to remove local drastic changes and obtain three continuous pressure weight curve sequences, denoted as follows: ; S342: Calculate the weighted average value of the three weighted curves over the entire predicted trajectory region. , used to characterize the global pressure distribution trend; S343: Yes Normalization is performed to ensure that the sum of the three is 1, which serves as the ground pressure distribution pattern of the track in the predicted motion trajectory area to adapt to changes in terrain slope.

2. The method for controlling the track passability of a combine harvester based on vision detection according to claim 1, characterized in that, S4 specifically includes: S41: Based on the target ground pressure weight values ​​of the front, middle and rear sections of the track determined in S3, and combined with the rated hydraulic flow parameters of each section of the track, calculate the initial hydraulic flow required for each section. S42: Calculate the proportion of the initial hydraulic flow in the front, middle and rear sections to determine the proportion of each section's hydraulic flow in the total oil supply flow, and obtain the corresponding target flow distribution ratio. S43: Based on the target hydraulic flow distribution ratio of each segment, output the corresponding travel motor control parameters for the front, middle and rear sections of the track.

3. The method for controlling the track passability of a combine harvester based on vision detection according to claim 2, characterized in that, The formula for calculating the target flow allocation ratio in S42 is as follows: ,in, The target hydraulic flow rate for each section of the track; The target ground pressure weight values ​​corresponding to each section of the track; The hydraulic pump supplies the total hydraulic flow to the track drive system.

4. The method for controlling the track passability of a combine harvester based on vision detection according to claim 3, characterized in that, S5 specifically includes: S51: Receives the target hydraulic flow rate corresponding to the front, middle and rear travel motors of the track output from S4, and uses it as the target value for flow control of the hydraulic system. S52: Based on the pre-calibrated flow-opening degree relationship function, the target hydraulic flow rate of each track segment is converted into the control opening degree command of the corresponding hydraulic valve group. The expression of the flow-opening degree relationship function is: ,in, The target opening degree for the hydraulic valve assemblies of each section of the track; The target hydraulic flow rate for each section of the track; It is the inverse function of the hydraulic flow rate-opening calibration function, used to convert the target flow rate into the valve orifice control quantity; S53: The converted opening degree command is sent to the corresponding track section hydraulic valve group in real time to drive the valve group to perform dynamic adjustment.

Images