All-terrain driving mode automatic switching method and device, equipment and storage medium

Abstract

The invention discloses an all-terrain driving mode automatic switching method and device, equipment and a storage medium, and the method comprises the steps: obtaining a monitoring image of an environment where a current vehicle is located through a camera, carrying out the image segmentation of the monitoring image, and obtaining ground edge information, forward road material information and current road surface material information; according to the ground edge information, the forward road material information and the current road surface material information, a material change edge distance and material change estimation time are obtained through measurement and calculation; according to the material change edge distance and the material change estimation time, the estimation response adjustment amount is determined, the current vehicle is controlled to automatically switch the all-terrain driving mode according to the estimation response adjustment amount, the material change edge distance and the estimation time can be accurately calculated, millisecond-level smooth switching of the all-terrain driving mode is achieved, and the driving efficiency is improved. Chassis jerking and mode false triggering at the junction of road surface materials are effectively avoided, and the control stability and driving comfort of the vehicle under the complex road condition are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent driving of automobiles, and in particular to a full-terrain driving mode automatic switching method, device, equipment and storage medium. BACKGROUND

[0002] The existing patent CN115534967A - a full-terrain driving mode switching method and system discloses that the driving mode is switched in the same type by two control units respectively, which is clear and does not need to be selected one by one in the complex full-terrain driving mode.

[0003] However, the acquisition of the actual road condition information in the technical solution of the existing patent CN115534967A relies on the front visual sensor of the vehicle; the acquired road condition information is the road condition in front of the vehicle rather than the current road condition of the vehicle; when the vehicle drives on a boundary road where the road surface material changes, the driving mode may be incorrectly switched to adapt to the material of the road in front, affecting the driving experience.

[0004] The existing patent CN115534967A only relies on the front visual sensor to acquire the front road condition information for material identification and mode switching, which leads to the technical problems of incorrect early switching or delayed switching of the driving mode, chassis parameter mutation, driving jerk and decreased control stability in the road surface material boundary area (such as asphalt-gravel transition zone) due to the inability to accurately determine the actual road surface material where the current wheel is located.

[0005] In the intelligent selection mode of the existing scheme, only the ground material identification result of the front camera image is relied on, and due to the limitation of the image recognition technology itself, there is a fixed misrecognition probability, which may lead to incorrect information received by the chassis and affect the driving experience due to the incorrect adjustment of the chassis; in the boundary area of the change of the road material, the chassis adjustment is prior to the actual ground material change due to the reliance on only the front camera recognition, which causes the driving experience to be reduced. SUMMARY

[0006] The main purpose of the present application is to provide a full-terrain driving mode automatic switching method, device, equipment and storage medium, which aims to solve the poor performance of the intelligent selection mode in the full-terrain driving mode in the prior art in the boundary road where the road surface material changes, the inability to accurately determine the actual road surface material where the current wheel is located, the incorrect early switching or delayed switching of the driving mode, and the technical problems of chassis parameter mutation, driving jerk and decreased control stability.

[0007] In a first aspect, the present application provides a full-terrain driving mode automatic switching method, which comprises the following steps: The monitoring image of the environment where the current vehicle is located is acquired through a camera, the monitoring image is image segmented to obtain ground edge information, forward road material information and current road surface material information; The material change edge distance and the material change estimated time are obtained according to the ground edge information, the forward road material information and the current road surface material information; The estimated response adjustment amount is determined according to the material change edge distance and the material change estimated time, and the current vehicle is automatically switched to the full terrain driving mode according to the estimated response adjustment amount.

[0008] Optionally, the monitoring image of the environment where the current vehicle is located is acquired through a camera, the monitoring image is image segmented to obtain ground edge information, forward road material information and current road surface material information, comprising: The monitoring image of the environment where the current vehicle is located is acquired through a forward medium focus camera and a long focus camera of the current vehicle; The monitoring image is image segmented by using a preset semantic segmentation algorithm to obtain ground edge information of a ground edge contour line of a forward road and forward road material information; The road surface material information of the current driving road is obtained by monitoring the road surface material information of the front and rear wheels of the current vehicle on the current driving road through a side view camera of the current vehicle.

[0009] Optionally, the material change edge distance and the material change estimated time are obtained according to the ground edge information, the forward road material information and the current road surface material information, comprising: The material change edge point is located according to the ground edge information, the forward road material information and the current road surface material information; The straight line spatial distance between the material change edge point and the center of the current vehicle is calculated in the vehicle coordinate system, and the straight line spatial distance is taken as the material change edge distance; The real-time vehicle speed of the current vehicle is acquired, and the material change estimated time is determined according to the real-time vehicle speed and the material change edge distance.

[0010] Optionally, the straight line spatial distance between the material change edge point and the center of the current vehicle is calculated in the vehicle coordinate system, and the straight line spatial distance is taken as the material change edge distance, comprising: The pixel coordinates of the material edge point are mapped to the world coordinate system through a preset camera calibration parameter to obtain a three-dimensional space point; The real-time pose information of the current vehicle is acquired, and the three-dimensional space point is converted to the vehicle coordinate system according to the real-time pose information to obtain the three-dimensional coordinates of the material edge point in the vehicle coordinate system; Solving a straight line space distance according to the three-dimensional coordinates, and taking the straight line space distance as the material change edge distance.

[0011] Optionally, the real-time vehicle speed of the current vehicle is acquired, and a material change estimated time is determined according to the real-time vehicle speed and the material change edge distance, including: The real-time vehicle speed of the current vehicle is acquired, and a material change estimated time is determined according to the real-time vehicle speed and the material change edge distance by the following formula:

[0012] wherein, is the material change estimated time, is the material change edge distance, is the real-time vehicle speed.

[0013] Optionally, the estimated response adjustment amount of the chassis system of the current vehicle is determined according to the material change edge distance and the material change estimated time, and the current vehicle is automatically switched to the full terrain driving mode according to the estimated response adjustment amount, including: The estimated response adjustment amount of the chassis system of the current vehicle is estimated according to the material change edge distance and the material change estimated time by a preset dynamic control algorithm; Before the material change estimated time arrives, the chassis system is controlled to execute the estimated response adjustment amount, so that the current vehicle is automatically switched from the current mode to the full terrain driving mode.

[0014] Optionally, the estimated response adjustment amount of the chassis system of the current vehicle is estimated according to the material change edge distance and the material change estimated time by a preset dynamic control algorithm, including: The material change edge distance is taken as a core error input by an adaptive PID controller, and an output value of a current core error weighted by a proportional gain is dynamically calculated; The historical error is accumulated by an integral term to eliminate a steady-state deviation, and the change trend of the output value is predicted by a differential term; According to the change trend and the material change estimated time, reinforcement learning is performed to output a smooth sequence, the smooth sequence is weighted and fused according to a preset road condition mutation level weight, and the estimated response adjustment amount of the chassis system of the current vehicle is obtained.

[0015] In a second aspect, to achieve the above object, the present application further provides a full terrain driving mode automatic switching device, which comprises: An image segmentation module is configured to acquire a monitoring image of an environment where the current vehicle is located through a camera, perform image segmentation on the monitoring image, and obtain ground edge information, forward road material information, and current road surface material information. A data measurement module is configured to measure and obtain material change edge distance and material change estimated time according to the ground edge information, the forward road material information, and the current road surface material information. A mode automatic switching module is configured to determine an estimated response adjustment amount according to the material change edge distance and the material change estimated time, and control the current vehicle to automatically switch the all-terrain driving mode according to the estimated response adjustment amount.

[0016] In a third aspect, the present application also provides an all-terrain driving mode automatic switching device, which comprises a memory, a processor, and an all-terrain driving mode automatic switching program stored in the memory and executable on the processor, and the all-terrain driving mode automatic switching program is configured to implement the steps of the all-terrain driving mode automatic switching method.

[0017] In a fourth aspect, the present application also provides a storage medium, which stores an all-terrain driving mode automatic switching program, and the all-terrain driving mode automatic switching program is executable on a processor to implement the steps of the all-terrain driving mode automatic switching method.

[0018] The all-terrain driving mode automatic switching method provided by the present application can detect the road surface material in front of the vehicle and the current road surface material through a visual sensor in front of the vehicle, make the all-terrain driving mode switching more intelligent, and be more suitable for the current road surface material, eliminate the influence of misrecognition, avoid the decline of driving experience caused by chassis error adjustment, accurately calculate the material change edge distance and the estimated time, realize millisecond-level smooth switching of the all-terrain driving mode, effectively avoid chassis jerk at the junction of the road surface material and mode mis-triggering, significantly improve the control stability and driving comfort of the vehicle in complex road conditions, and improve the speed and efficiency of the all-terrain driving mode automatic switching. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1The device structure schematic diagram of the hardware running environment involved in the embodiment of the present application; Figure 2 The flowchart schematic diagram of the first embodiment of the all-terrain driving mode automatic switching method of the present application; Figure 3 The flowchart schematic diagram of the second embodiment of the all-terrain driving mode automatic switching method of the present application; Figure 4 The multi-sensor identification range schematic diagram in the all-terrain driving mode automatic switching method of the present application; Figure 5 The flowchart schematic diagram of the third embodiment of the all-terrain driving mode automatic switching method of the present application; Figure 6 The flowchart schematic diagram of the fourth embodiment of the all-terrain driving mode automatic switching method of the present application; Figure 7 The chassis adjustment flowchart schematic diagram of the all-terrain driving mode automatic switching method of the present application; Figure 8 The functional module diagram of the first embodiment of the all-terrain driving mode automatic switching device of the present application.

[0020] The object realization, functional features and advantages of the present application will be further explained with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein merely set forth instead of limiting the present application.

[0022] The solution of the embodiment of the present application is mainly: obtaining a monitoring image of an environment where a current vehicle is located through a camera, performing image segmentation on the monitoring image to obtain ground edge information, forward road material information and current road surface material information; calculating a material change edge distance and a material change estimation time according to the ground edge information, the forward road material information and the current road surface material information; determining an estimated response adjustment amount according to the material change edge distance and the material change estimation time, and controlling the current vehicle to automatically switch the all-terrain driving mode according to the estimated response adjustment amount. The road surface material in front of the vehicle and the current road can be detected by using the visual sensor in front of the vehicle, the all-terrain driving mode switching is more intelligent, and is more suitable for the current road material. The influence of misrecognition can be eliminated, so as to avoid the decline of driving experience caused by chassis error adjustment. The material change edge distance and the estimation time are accurately calculated, the millisecond-level smooth switching of the all-terrain driving mode is realized, the chassis hesitation at the junction of the road surface material and the mode mis-triggering are effectively avoided, the control stability and the driving comfort of the vehicle in complex road conditions are significantly improved, the speed and the efficiency of the automatic switching of the all-terrain driving mode are improved, and the technical problems that the intelligent selection mode in the all-terrain driving mode performs poorly on the junction road where the road surface material changes, cannot accurately judge the actual road surface material where the current wheel is located, causes the driving mode to be switched in advance or delayed, and causes the chassis parameter to suddenly change, the driving to be hesitant, and the control stability to be reduced are solved.

[0023] Reference Figure 1 , Figure 1 The device structure schematic diagram of a hardware running environment involved in the embodiment of the present application.

[0024] As Figure 1 shown, the device can include a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to realize the connection and communication between the components. The user interface 1003 can include a display screen (Display) and an input unit such as a keyboard (Keyboard). The optional user interface 1003 can also include a standard wired interface and a wireless interface. The network interface 1004 can optionally include a standard wired interface and a wireless interface (such as a Wi-Fi interface). The memory 1005 can be a high-speed RAM memory or a stable memory (Non-Volatile Memory), such as a magnetic disk memory. The memory 1005 can also be an independent storage device from the aforementioned processor 1001.

[0025] Those skilled in the art can understand, Figure 1The device structure shown in the figure does not constitute a limitation on the device, and can include more or fewer components than shown, or combine certain components, or different component arrangements.

[0026] As shown in Figure 1 The memory 1005 as a storage medium can include an operating device, a network communication module, a user interface module, and an all-terrain driving mode automatic switching program.

[0027] The device of the application calls the all-terrain driving mode automatic switching program stored in the memory 1005 through the processor 1001, and performs the following operations: Obtain the monitoring image of the environment where the current vehicle is located through the camera, and perform image segmentation on the monitoring image to obtain ground edge information, forward road material information and current road surface material information; According to the ground edge information, the forward road material information and the current road surface material information, the material change edge distance and the material change estimated time are obtained; According to the material change edge distance and the material change estimated time, the estimated response adjustment amount is determined, and the current vehicle is automatically switched to the all-terrain driving mode according to the estimated response adjustment amount.

[0028] The device of the application calls the all-terrain driving mode automatic switching program stored in the memory 1005 through the processor 1001, and further performs the following operations: Real-time monitoring image of the environment where the current vehicle is located is obtained through the forward medium focus camera and long focus camera of the current vehicle; The monitoring image is segmented by using a preset semantic segmentation algorithm to obtain ground edge information of the ground edge contour line of the forward road and forward road material information; The road surface material information of the current driving road of the front and rear wheels of the current vehicle is obtained by monitoring the road surface material information of the current driving road of the current vehicle through the side view camera of the current vehicle.

[0029] The device of the application calls the all-terrain driving mode automatic switching program stored in the memory 1005 through the processor 1001, and further performs the following operations: According to the ground edge information, the forward road material information and the current road surface material information, the material change edge point is located; The straight line spatial distance between the material change edge point and the center of the current vehicle is calculated in the vehicle coordinate system, and the straight line spatial distance is taken as the material change edge distance; The real-time vehicle speed of the current vehicle is obtained, and the material change estimated time is determined according to the real-time vehicle speed and the material change edge distance.

[0030] The device of the present application calls the all-terrain driving mode automatic switching program stored in the memory 1005 through the processor 1001, and also performs the following operations: The pixel coordinates of the material edge point are mapped to the world coordinate system by preset camera calibration parameters, and a three-dimensional space point is obtained; Real-time pose information of the current vehicle is obtained, and the three-dimensional space point is converted to the vehicle coordinate system according to the real-time pose information, and the three-dimensional coordinates of the material edge point in the vehicle coordinate system are obtained; The straight line space distance is solved according to the three-dimensional coordinates, and the straight line space distance is taken as the material change edge distance.

[0031] The device of the present application calls the all-terrain driving mode automatic switching program stored in the memory 1005 through the processor 1001, and also performs the following operations: The real-time vehicle speed of the current vehicle is obtained, and the material change estimated time is determined according to the real-time vehicle speed and the material change edge distance by the following formula:

[0032] Wherein, The material change estimated time, The material change edge distance, The real-time vehicle speed.

[0033] The device of the present application calls the all-terrain driving mode automatic switching program stored in the memory 1005 through the processor 1001, and also performs the following operations: According to the material change edge distance and the material change estimated time, the estimated response adjustment amount of the chassis system execution adjustment parameter of the current vehicle is estimated by a preset dynamic control algorithm; Before the material change estimated time arrives, the chassis system is controlled to execute the estimated response adjustment amount, so that the current vehicle is automatically switched from the current mode to the all-terrain driving mode.

[0034] The device of the present application calls the all-terrain driving mode automatic switching program stored in the memory 1005 through the processor 1001, and also performs the following operations: The material change edge distance is taken as the core error input through the adaptive PID controller, and the output value of the current core error after proportional gain weighting is dynamically calculated; The historical error is accumulated through the integral term to eliminate the steady-state deviation, and the change trend of the output value is predicted through the differential term; According to the change trend and the material change estimated time, a smoothing sequence is output, the smoothing sequence is weighted and fused according to the preset road condition mutation level weight, and the estimated response adjustment amount of the chassis system execution adjustment parameter of the current vehicle is obtained.

[0035] The embodiment obtains the monitoring image of the environment where the current vehicle is located through the camera, performs image segmentation on the monitoring image to obtain ground edge information, forward road material information and current road surface material information, calculates the material change edge distance and the material change estimation time according to the ground edge information, the forward road material information and the current road surface material information, determines the estimated response adjustment amount according to the material change edge distance and the material change estimation time, and controls the current vehicle to automatically switch the full terrain driving mode according to the estimated response adjustment amount. The road surface material in front of the vehicle and the current road can be detected by using the visual sensor in front of the vehicle, the full terrain driving mode switching is more intelligent, and the full terrain driving mode can better adapt to the current road material. The influence of misrecognition can be eliminated, the driving experience caused by chassis error adjustment can be avoided, the material change edge distance and the estimation time are accurately calculated, the full terrain driving mode is smoothly switched in milliseconds, the chassis hesitation at the junction of the road surface material and the mode mistriggering are effectively avoided, the control stability and the driving comfort of the vehicle in complex road conditions are significantly improved, and the speed and the efficiency of the full terrain driving mode automatic switching are improved.

[0036] Based on the above hardware structure, the full terrain driving mode automatic switching method embodiment of the application is proposed.

[0037] Reference Figure 2 , Figure 2 The flowchart of the first embodiment of the full terrain driving mode automatic switching method of the application is shown.

[0038] In the first embodiment, the full terrain driving mode automatic switching method comprises the following steps: Step S10, a monitoring image of the environment where the current vehicle is located is obtained through a camera, the monitoring image is segmented to obtain ground edge information, forward road material information and current road surface material information.

[0039] It should be noted that the monitoring image of the road environment where the current vehicle is located can be obtained through multiple visual sensors, i.e., multiple cameras. The ground edge information, the forward road material information and the current road surface material information can be obtained by performing image segmentation on the monitoring image. The ground edge information refers to the geometric contour line of the road boundary and the material junction extracted by image segmentation. The forward road material information refers to the distribution and classification information of different material regions (such as asphalt and gravel) on the road surface in front of the vehicle. The current road surface material information refers to the real-time monitoring result of the road surface material state (such as ice and snow, and mud) of the actual contact area of the front and rear wheels of the vehicle.

[0040] Step S20, according to the ground edge information, the forward road material information and the current road surface material information, the material change edge distance and the material change estimated time are obtained by calculation.

[0041] It should be understood that the material change edge distance and the material change estimated time can be obtained by calculation according to the ground edge information, the forward road material information and the current road surface material information; the material change edge distance refers to the straight line space distance from the current position of the vehicle to the material change boundary point (such as the asphalt-gravel boundary); the material change estimated time refers to the estimated time required for the vehicle to travel to the material change boundary point at the current speed.

[0042] Step S30, according to the material change edge distance and the material change estimated time, the estimated response adjustment amount is determined, and the current vehicle is automatically switched to the full terrain driving mode according to the estimated response adjustment amount.

[0043] It can be understood that the estimated response adjustment amount of the vehicle chassis adjustment can be determined according to the material change edge distance and the material change estimated time, and the current vehicle can be automatically switched to the full terrain driving mode by the estimated response adjustment amount.

[0044] The embodiment obtains the monitoring image of the environment where the current vehicle is located by the camera, performs image segmentation on the monitoring image, and obtains the ground edge information, the forward road material information and the current road surface material information; according to the ground edge information, the forward road material information and the current road surface material information, the material change edge distance and the material change estimated time are obtained by calculation; according to the material change edge distance and the material change estimated time, the estimated response adjustment amount is determined, and the current vehicle is automatically switched to the full terrain driving mode according to the estimated response adjustment amount, which can detect the road surface material in front of the vehicle and the current road by the visual sensor in front of the vehicle, make the full terrain driving mode switching more intelligent, more suitable for the current road material, eliminate the influence of misrecognition, avoid the decline of driving experience caused by chassis error adjustment, accurately calculate the material change edge distance and the estimated time, realize the millisecond level smooth switching of the full terrain driving mode, effectively avoid the chassis hesitation and mode misfire at the material change boundary, significantly improve the control stability and driving comfort of the vehicle in complex road conditions, and improve the speed and efficiency of the full terrain driving mode automatic switching.

[0045] Further, Figure 3 The flowchart of the second embodiment of the full terrain driving mode automatic switching method of the present application is shown in Figure 3 Based on the first embodiment, the second embodiment of the full terrain driving mode automatic switching method of the present application is proposed, and in this embodiment, the step S10 specifically includes the following steps: Step S11, real-time acquisition of the monitoring image of the environment where the current vehicle is located through the front mid-focus camera and the long-focus camera of the current vehicle.

[0046] It should be noted that the front mid-focus camera and the long-focus camera are used simultaneously to real-time acquisition of the monitoring image of the environment where the current vehicle is located.

[0047] Step S12, image segmentation of the monitoring image by using a preset semantic segmentation algorithm to obtain the ground edge information of the ground edge contour line of the front road and the front road material information.

[0048] It can be understood that the preset semantic segmentation algorithm based on deep learning can be used to segment the monitoring image, so as to obtain the ground edge information of the ground edge contour line of the front road and the front road material information.

[0049] It should be understood that after the monitoring image is segmented by using the preset semantic segmentation algorithm, the image result of the ground material calculated by the neural network of the input vehicle-mounted camera can be obtained; the training of the neural network uses the terrain material labeled picture as the input, and after the training is completed, the neural network can output the image segmentation identification result of the ground material of the picture; the process of extracting the edge can be calculated by using the classical contour extraction algorithm of computer vision, and the typical algorithm is, for example, the raster scanning method.

[0050] Step S13, monitoring of the road surface material condition of the front and rear wheels of the current vehicle on the current driving road by using the side-view camera of the current vehicle to obtain the current road surface material information of the current driving road.

[0051] It should be noted that the side-view camera of the current vehicle can be used to real-time monitor the road surface material condition of the front and rear wheels of the current vehicle on the current driving road, so as to obtain the current road surface material information of the current driving road.

[0052] It can be understood that the real-time monitoring image of the front and rear wheel regions by using the side-view camera, combined with the pixel-level identification of the road surface material of the tire contact surface by using the preset semantic segmentation algorithm, can dynamically obtain the current road surface material information (such as asphalt, gravel, ice and snow, or muddy) of the front and rear wheels, so as to provide the millisecond-level accurate road surface state parameters for the driving mode switching.

[0053] It should be understood that, referring to Figure 4 , Figure 4 The schematic diagram of the recognition range of multiple sensors in the full-terrain driving mode automatic switching method of the present application is as follows: Figure 4As shown, the camera image obtained by the vehicle camera includes: a front ground material recognition area covered by the front camera field of view, a left front / rear wheel ground material recognition area covered by the left side camera field of view, and a right front / rear wheel ground material recognition area covered by the right side camera field of view; the side-view camera monitors the current road surface material of the front and rear wheels, and the monitoring result is compared with the front-view camera sensing result to output the current road surface material information; the side-view camera and the front-view camera both use image segmentation technology to output the image segmentation detection result of the ground material in the camera field of view, which can be compared with the front-view camera result to further improve the accuracy of the recognition result.

[0054] In a specific implementation, the depth map obtained by the binocular distance measurement system of the vehicle front mid-focus and long-focus cameras in real time can be used as a key input to drive a preset deep learning semantic segmentation algorithm (such as a model based on a U-shaped network U-Net or a Mask Region-based Convolutional Neural Networks (Mask R-CNN) architecture, which is trained by a large number of terrain material annotation images) to perform high-precision segmentation processing on the monitoring image, accurately identify and extract the ground edge contour line (including road boundary, slope change point, and other geometric features) of the current road surface and the material distribution characteristics of various road surface materials (such as asphalt, gravel, ice and snow, and mud), and output the segmentation result containing the material category label and spatial position information to provide real-time and reliable environmental perception data support for subsequent calculation of material change estimation distance, entry time point, and driving mode switching decision.

[0055] In the embodiment, the monitoring image of the environment where the current vehicle is located is obtained in real time by the front mid-focus camera and the long-focus camera of the current vehicle; the ground edge information of the front road ground edge contour line and the front road material information are obtained by performing image segmentation on the monitoring image by using a preset semantic segmentation algorithm; the current road surface material information of the current road is obtained by monitoring the road surface material of the front and rear wheels of the current vehicle on the current road by using the side-view camera, which can realize accurate multi-source perception of the actual driving road condition of the vehicle and completely solve the driving mode switching error problem caused by single sensor misjudgment, thereby providing reliable data basis for chassis control.

[0056] Further, Figure 5 The flowchart of the third embodiment of the all-terrain driving mode automatic switching method of the present application is shown in FIG. 6. Figure 5 As shown, the third embodiment of the all-terrain driving mode automatic switching method of the present application is proposed based on the first embodiment. In this embodiment, the step S20 specifically includes the following steps: Step S21, locating a material change edge point according to the ground edge information, the forward road material information and the current road surface material information.

[0057] It should be noted that the critical region of road material change can be determined according to the ground edge information, the forward road material information and the current road surface material information, so as to locate the edge point of material change.

[0058] In a specific implementation, based on the ground edge information (including road boundary, slope change point and other geometric features) and the forward road material information (such as asphalt, gravel, ice and snow and other material distribution heat maps) output by image segmentation, the material change edge point (such as the pixel coordinates of the asphalt-gravel junction) can be accurately located. Step S22, calculating a straight line spatial distance between the material change edge point and the center of the current vehicle in the vehicle coordinate system, and taking the straight line spatial distance as the material change edge distance.

[0059] It can be understood that the straight line spatial distance between the material change edge point and the center of the current vehicle in the vehicle coordinate system is calculated, and then the straight line spatial distance is taken as the material change edge distance.

[0060] Further, the step S22 specifically includes the following steps: mapping the pixel coordinates of the material edge point to the world coordinate system through a preset camera calibration parameter to obtain a three-dimensional space point; obtaining real-time pose information of the current vehicle, converting the three-dimensional space point to the vehicle coordinate system according to the real-time pose information to obtain a three-dimensional coordinate of the material edge point in the vehicle coordinate system; solving a straight line spatial distance according to the three-dimensional coordinate, and taking the straight line spatial distance as the material change edge distance.

[0061] In a specific implementation, based on the pixel coordinates of the material change edge point output by image segmentation, the pixel coordinates are mapped to a three-dimensional space point in the world coordinate system through a preset camera calibration parameter (including an intrinsic matrix and an extrinsic rotation and translation matrix), and then combined with real-time pose information of the vehicle (obtained by fusion calculation of an IMU inertial measurement unit and a GPS, including a pitch angle, a yaw angle and a vehicle center position), the point is accurately converted to the vehicle coordinate system (the origin is located at the geometric center of the vehicle, the x-axis points to the forward direction of the vehicle, the y-axis points to the left, and the z-axis is perpendicular upward) by using a coordinate transformation matrix; then the three-dimensional coordinates (x, y, z) of the edge point in the vehicle coordinate system are calculated, and the Euclidean straight line spatial distance The distance value is the material change edge distance, and after calibration through multi-sensor data fusion, the precision is stable within ±0.1 meters, which provides an accurate spatial position reference for driving mode switching decision, and ensures smooth chassis adjustment with millisecond-level response in the material change interface area.

[0062] In step S23, the real-time vehicle speed of the current vehicle is obtained, and the material change estimated time is determined according to the real-time vehicle speed and the material change edge distance.

[0063] It should be understood that after the real-time vehicle speed of the current vehicle is obtained, the material change estimated time can be determined according to the real-time vehicle speed and the material change edge distance.

[0064] Further, the step S23 specifically includes the following steps: The real-time vehicle speed of the current vehicle is obtained, and the material change estimated time is determined according to the real-time vehicle speed and the material change edge distance through the following formula:

[0065] wherein, is the material change estimated time, is the material change edge distance, is the real-time vehicle speed.

[0066] In a specific implementation, in combination with a binocular distance measurement system (a depth map generated by calculating a parallax through a Semi-Global Block Matching (SGBM) algorithm and converted into a three-dimensional coordinate through triangulation), the straight-line spatial distance (material change edge distance, with an accuracy of ±0.1 meters) between the edge point and the center of the current vehicle is accurately calculated in the vehicle coordinate system; at the same time, the current road material information (pixel-level recognition results confirmed through semantic segmentation by a side-view camera for monitoring the front and rear wheel contact surface materials in real time) and the real-time vehicle speed are fused, and the material change estimated time (unit: seconds) is calculated in real time through a dynamic formula: estimated time = material change edge distance / vehicle speed, to ensure that the driving mode switching decision is completed within milliseconds, and to avoid chassis misadjustment in the material change interface area.

[0067] The embodiment obtains the three-dimensional space point by mapping the pixel coordinates of the material edge point to the world coordinate system through the preset camera calibration parameter; acquires real-time pose information of the current vehicle, and converts the three-dimensional space point to the vehicle coordinate system according to the real-time pose information to obtain the three-dimensional coordinates of the material edge point in the vehicle coordinate system; and solves the straight line space distance according to the three-dimensional coordinates, and takes the straight line space distance as the material change edge distance. The material change edge distance can be calculated with high precision, and a millisecond-level space reference is provided for the all-terrain driving mode switching, and chassis jerk and mode false triggering caused by distance error are completely eliminated.

[0068] Further, Figure 6 The flowchart of the fourth embodiment of the all-terrain driving mode automatic switching method of the present application is shown in Figure 6 Based on the first embodiment, the fourth embodiment of the all-terrain driving mode automatic switching method of the present application is proposed, and in the embodiment, the step S30 specifically includes the following steps. In step S31, the estimated response adjustment amount of the adjustment parameter of the chassis system of the current vehicle is estimated according to the material change edge distance and the material change estimated time through a preset dynamic control algorithm.

[0069] It should be noted that the estimated response adjustment amount can be calculated in real time through the preset dynamic control algorithm based on the material change edge distance and the material change estimated time, that is, the estimated response adjustment amount of the adjustment parameter of the chassis system of the current vehicle is estimated.

[0070] It can be understood that the estimated response adjustment amount is the accurate parameter value that the chassis system needs to execute, including but not limited to wheel speed differential distribution ratio, torque output gradient, suspension damping coefficient, etc., for example: differential lock state (such as switching from non-locking to full locking to enhance off-road traction), vehicle body height adjustment amount (such as lifting the air suspension by 5-10 mm to increase the ground clearance), or steering assist coefficient (such as dynamically increasing the assist force on a muddy road to improve the handling), etc., which are not limited in the embodiment.

[0071] Further, the step S31 specifically includes the following steps: The material change edge distance is taken as the core error input through an adaptive PID controller, and the output value of the current core error after proportional gain weighting is dynamically calculated; The historical error is accumulated through the integral term to eliminate the steady-state deviation, and the change trend of the output value is predicted through the differential term; According to the change trend and the material change estimated time, a reinforcement learning is performed to output a smooth sequence, the smooth sequence is weighted and fused according to a preset road condition mutation level weight, and the estimated response adjustment amount of the adjustment parameter of the chassis system of the current vehicle is obtained.

[0072] It should be noted that the preset dynamic control algorithm can be a parameter optimization model based on reinforcement learning or other algorithm model in addition to an adaptive proportional integral derivative (PID) controller, and the present embodiment does not limit this. Taking the adaptive PID controller as an example, the estimated response adjustment amount can be calculated in real time through the adaptive PID controller.

[0073] In a specific implementation, the system can calculate the estimated response adjustment amount through a dual-mode algorithm based on the material change edge distance (D) and the material change estimated time (T). First, the adaptive PID controller takes D as the core error input, and the proportional term (P) is dynamically calculated as Kp x D (Kp parameter is optimized in real time through Kalman filtering, and the material mutation rate is adaptively adjusted according to historical road condition data); the integral term (I) accumulates the historical error (Ki x ∫Ddt) to eliminate the steady-state deviation, and the differential term (D) predicts the change trend (Kd x dD / dt, Kd is dynamically scaled according to the real-time vehicle speed V); at the same time, the reinforcement learning model (based on a trained Deep Q-Network, input states including D, T, V and the current driving mode) outputs a smooth adjustment sequence (such as wheel speed difference allocation proportion, torque gradient percentage, suspension damping coefficient); after parallel calculation of the two algorithms, the final adjustment amount (unit: wheel speed difference ±0.5% / torque gradient ±0.3% / damping coefficient ±0.2%) is generated through weighted fusion (weights are dynamically allocated according to the road condition mutation level), ensuring that the chassis parameters realize gradual transition (such as comfortable mode→off-road mode) within a millisecond window before the estimated time T arrives, the accuracy is controlled within ±0.5%, and the jerk at the material junction is completely eliminated, providing millisecond-accurate control instructions for driving mode switching.

[0074] Step S32, before the material change estimated time arrives, controlling the chassis system to execute the estimated response adjustment amount to automatically switch the current vehicle from the current mode to the all-terrain driving mode.

[0075] It can be understood that before the material change estimated time arrives, the chassis system can be controlled to execute the estimated response adjustment amount, and the corresponding wheel speed, torque, moment, etc. Chassis parameter adjustment is performed by the chassis response, so that the current vehicle is automatically switched from the current mode to the all-terrain driving mode.

[0076] In a specific implementation, as shown in Figure 7 , Fig. 1 is a chassis adjustment flowchart of the all-terrain driving mode automatic switching method of the present application, and Figure 7 Figure 7 ​The result extracted from the edge of the front camera is the actual material change line distance from the current vehicle far point distance, which is the real-time detection result. The arrival time of the material change point can be calculated in real time from the speed and distance information. The road surface material information of the current driving road is output by monitoring the road surface material of the front and rear wheels through the side view camera and rechecking the front camera sensing result. The front and side road material information and the estimated material change information are summarized and transmitted to the chassis through a fixed interface. The chassis realizes automatic switching of the driving mode, and the suspension adjustment parameters are transmitted to the car central control. The front road information is summarized and comprehensively judged to almost eliminate the influence of misidentification, thereby avoiding the decline in driving experience caused by chassis error adjustment.

[0077] It should be understood that the vehicle control unit (VCU) automatically triggers the smooth switching of the all-terrain driving mode (such as from the comfort mode to the off-road mode) according to the adjustment amount at the millisecond level before the material change estimation time arrives, ensuring gradual adjustment of driving parameters in the road material boundary area, completely avoiding chassis jerk or mode misfire caused by material mutation, and thus ensuring the control stability and driving comfort of the vehicle in complex road conditions.

[0078] It can be understood that the vehicle control unit (VCU) receives the wheel speed difference allocation ratio (such as a dynamic gradient of ±0.5%), torque output gradient (such as a smooth change rate of ±0.3%), and suspension damping coefficient (such as a gradual adjustment value of ±0.2%) contained in the estimated response adjustment amount in real time. Through a closed-loop control algorithm, the gradual dynamic adjustment of the chassis parameters is performed within a millisecond window before the material change estimation time arrives, for example: the wheel speed difference is smoothly increased from 0% in the current comfort mode to 15% in the off-road mode, the torque gradient is exponentially decayed from 20% to 40%, and the suspension damping coefficient is gradually transitioned from soft tuning to hard tuning, achieving a seamless switching of the all-terrain driving mode (such as from the comfort mode to the off-road mode), completely eliminating chassis jerk and mode misfire at the road material boundary, and ensuring the control continuity and driving comfort of the vehicle in complex road conditions.

[0079] It should be understood that the wheel speed difference allocation ratio is used to adjust the speed difference of the front and rear wheels or the left and right wheels of the vehicle (such as 0% in the comfort mode and smoothly increased to 15% in the off-road mode), which is a quantitative indicator of wheel speed adjustment. The torque output gradient is used to dynamically control the gradient change of torque distribution (such as smoothly increasing from 20% to 40%). The suspension damping coefficient is used to adjust the damping stiffness of the suspension system (such as gradually transitioning from soft tuning to hard tuning with a change range of ±0.2%).

[0080] It should be noted that the ground material is identified by the deep learning image segmentation method in the present application, and the actual process can not be limited to this method. A series of image recognition and classification processing methods such as target detection and panoramic segmentation can achieve the purpose of ground material identification, and the present embodiment does not limit this.

[0081] In the embodiment, the material change edge distance and the material change estimation time are used to estimate the estimated response adjustment amount of the chassis system adjustment parameter of the current vehicle through a preset dynamic control algorithm. Before the material change estimation time arrives, the chassis system is controlled to execute the estimated response adjustment amount, so that the current vehicle automatically switches from the current mode to the all-terrain driving mode. The road surface material in front of the vehicle and the current road can be detected by the visual sensor in front of the vehicle, the all-terrain driving mode switching is more intelligent, and the current road material can be more adapted. The influence of misrecognition can be eliminated, so as to avoid the decline of driving experience caused by chassis error adjustment, accurately calculate the material change edge distance and the estimation time, realize the millisecond level smooth switching of the all-terrain driving mode, effectively avoid the chassis jerk at the junction of the road surface material and the mode mis-triggering, significantly improve the control stability and driving comfort of the vehicle in complex road conditions, and improve the speed and efficiency of the automatic switching of the all-terrain driving mode.

[0082] Correspondingly, the present application further provides an all-terrain driving mode automatic switching device.

[0083] Reference Figure 8 , Figure 8 It is a function module diagram of the first embodiment of the all-terrain driving mode automatic switching device of the present application.

[0084] In the first embodiment of the all-terrain driving mode automatic switching device of the present application, the all-terrain driving mode automatic switching device comprises: The image segmentation module 10 is used to obtain the monitoring image of the environment where the current vehicle is located through the camera, perform image segmentation on the monitoring image, and obtain the ground edge information, the forward road material information and the current road material information.

[0085] The data measurement module 20 is used to measure and calculate the material change edge distance and the material change estimation time according to the ground edge information, the forward road material information and the current road material information.

[0086] The mode automatic switching module 30 is used to determine the estimated response adjustment amount according to the material change edge distance and the material change estimation time, and control the current vehicle to automatically switch the all-terrain driving mode according to the estimated response adjustment amount.

[0087] The image segmentation module 10 is further configured to acquire a monitoring image of an environment where the current vehicle is located in real time through a front mid-focus camera and a long-focus camera of the current vehicle, perform image segmentation on the monitoring image by using a preset semantic segmentation algorithm to obtain ground edge information of a front road ground edge contour line and front road material information, and acquire current road surface material information of a current driving road by monitoring road surface material conditions of front and rear wheels of the current vehicle through a side-view camera of the current vehicle.

[0088] The data measurement module 20 is further configured to locate a material change edge point according to the ground edge information, the front road material information and the current road surface material information, calculate a straight-line spatial distance between the material change edge point and a center of the current vehicle in a vehicle coordinate system, take the straight-line spatial distance as a material change edge distance, acquire a real-time vehicle speed of the current vehicle, and determine a material change estimated time according to the real-time vehicle speed and the material change edge distance.

[0089] The data measurement module 20 is further configured to map pixel coordinates of the material change edge point to a world coordinate system by using a preset camera calibration parameter to obtain a three-dimensional space point, acquire real-time pose information of the current vehicle, convert the three-dimensional space point to a vehicle coordinate system according to the real-time pose information to obtain three-dimensional coordinates of the material change edge point in the vehicle coordinate system, and solve a straight-line spatial distance according to the three-dimensional coordinates, taking the straight-line spatial distance as the material change edge distance.

[0090] The data measurement module 20 is further configured to acquire a real-time vehicle speed of the current vehicle, and determine a material change estimated time according to the real-time vehicle speed and the material change edge distance by using the following formula:

[0091] wherein, the material change estimated time is t, the material change edge distance is d, the real-time vehicle speed is v.

[0092] The mode automatic switching module 30 is further configured to estimate an estimated response adjustment amount of an adjustment parameter of a chassis system of the current vehicle by using a preset dynamic control algorithm according to the material change edge distance and the material change estimated time, control the chassis system to execute the estimated response adjustment amount before the material change estimated time arrives, and automatically switch the current vehicle from a current mode to an all-terrain driving mode.

[0093] The mode automatic switching module 30 is further configured to input the material change edge distance as a core error into an adaptive PID controller, dynamically calculate an output value of the current core error after proportional gain weighting, accumulate historical errors through an integral term to eliminate steady-state deviation, and predict a change trend of the output value through a differential term; perform reinforcement learning according to the change trend and the material change estimation time, output a smooth sequence, weight and fuse the smooth sequence according to preset road condition mutation level weights, and obtain an estimated response adjustment amount of the chassis system execution adjustment parameter of the current vehicle.

[0094] The steps of the various functional modules of the all-terrain driving mode automatic switching device can refer to the various embodiments of the all-terrain driving mode automatic switching method of the present application, and will not be described here.

[0095] In addition, an embodiment of the present application further provides a storage medium, and the storage medium stores an all-terrain driving mode automatic switching program. When the all-terrain driving mode automatic switching program is executed by a processor, the following operations are implemented: Obtain the ground edge information, the forward road material information, and the current road surface material information by performing image segmentation on the monitoring image. Calculate the material change edge distance and the material change estimation time according to the ground edge information, the forward road material information, and the current road surface material information. Determine the estimated response adjustment amount according to the material change edge distance and the material change estimation time, and control the current vehicle to automatically switch the all-terrain driving mode according to the estimated response adjustment amount.

[0096] Further, when the all-terrain driving mode automatic switching program is executed by the processor, the following operations are further implemented: Real-time acquire the monitoring image of the environment where the current vehicle is located through the forward mid-focus camera and the long-focus camera of the current vehicle. Obtain the ground edge information and the forward road material information of the ground edge contour line of the forward road by performing image segmentation on the monitoring image using a preset semantic segmentation algorithm. Obtain the current road surface material information of the current driving road by monitoring the road surface material conditions of the front and rear wheels of the current vehicle on the current driving road through the side-view camera of the current vehicle.

[0097] Further, when the all-terrain driving mode automatic switching program is executed by the processor, the following operations are further implemented: Locate the material change edge point according to the ground edge information, the forward road material information, and the current road surface material information. Calculate a straight line spatial distance between the material change edge point and the center of the current vehicle in a vehicle coordinate system, and take the straight line spatial distance as a material change edge distance; Obtain a real-time vehicle speed of the current vehicle, and determine a material change estimated time according to the real-time vehicle speed and the material change edge distance.

[0098] Further, the all-terrain driving mode automatic switching program, when executed by the processor, further implements the following operations: Map the pixel coordinates of the material edge point to a world coordinate system through preset camera calibration parameters to obtain a three-dimensional space point; Obtain real-time pose information of the current vehicle, and convert the three-dimensional space point to a vehicle coordinate system according to the real-time pose information to obtain three-dimensional coordinates of the material edge point in the vehicle coordinate system; Solve a straight line spatial distance according to the three-dimensional coordinates, and take the straight line spatial distance as a material change edge distance.

[0099] Further, the all-terrain driving mode automatic switching program, when executed by the processor, further implements the following operations: Obtain a real-time vehicle speed of the current vehicle, and determine a material change estimated time according to the real-time vehicle speed and the material change edge distance through the following formula:

[0100] wherein, is the material change estimated time, is the material change edge distance, is the real-time vehicle speed.

[0101] Further, the all-terrain driving mode automatic switching program, when executed by the processor, further implements the following operations: Estimate an estimated response adjustment amount of an adjustment parameter of a chassis system of the current vehicle according to the material change edge distance and the material change estimated time through a preset dynamic control algorithm; Before the material change estimated time arrives, control the chassis system to execute the estimated response adjustment amount, so that the current vehicle is automatically switched from the current mode to the all-terrain driving mode.

[0102] Further, the all-terrain driving mode automatic switching program, when executed by the processor, further implements the following operations: Take the material change edge distance as a core error input through an adaptive PID controller, and dynamically calculate an output value of the current core error after proportional gain weighting; Eliminate steady-state deviation through an integral term to accumulate historical errors, and predict a change trend of the output value through a differential term; According to the change trend and the material quality change prediction time, reinforcement learning is performed, a smooth sequence is output, the smooth sequence is weighted and fused according to a preset road condition mutation level weight, and a predicted response adjustment amount of a chassis system execution adjustment parameter of the current vehicle is obtained.

[0103] Those skilled in the art can understand that all or part of the steps of the methods in the above embodiments can be completed by programs instructing relevant hardware, the programs are stored in a storage medium, and the programs include a plurality of instructions for causing an apparatus (which can be a single-chip microcomputer, a chip, etc.) or a processor to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium is a computer readable storage medium, including: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0104] It should be noted that, in this document, the terms "comprising", "containing" or any other variant thereof are intended to cover non-exclusive inclusions, such that a process, method, article, or apparatus that comprises a list of elements does not only include those elements, but also includes other elements not explicitly listed, or further includes elements inherent in such a process, method, article, or apparatus. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of another identical element in the process, method, article, or apparatus that includes the element.

[0105] The above-mentioned serial numbers of the embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0106] The above is only the preferred embodiments of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation obtained by using the content of the specification and drawings, or direct or indirect application in other related technical fields, is also included in the patent protection scope of the present application.

Claims

1. A method for automatically switching all-terrain driving modes, characterized in that, The automatic switching method for all-terrain driving modes includes: The system acquires monitoring images of the current vehicle environment using a camera, performs image segmentation on the monitoring images, and obtains ground edge information, forward road material information, and current road surface material information. The distance to the edge of the material change and the estimated time of the material change are calculated based on the ground edge information, the forward road material information and the current road surface material information. The estimated response adjustment amount is determined based on the distance to the edge of the material change and the estimated time of the material change, and the current vehicle is controlled to automatically switch to all-terrain driving mode based on the estimated response adjustment amount.

2. The all-terrain driving mode automatic switching method as described in claim 1, characterized in that, The process involves acquiring monitoring images of the current vehicle environment via a camera, performing image segmentation on the monitoring images to obtain ground edge information, forward road material information, and current road surface material information, including: The vehicle acquires real-time monitoring images of its environment using its forward-facing mid-range and telephoto cameras. The monitoring image is segmented using a preset semantic segmentation algorithm to obtain ground edge information and forward road material information of the ground edge contour line of the forward road. The vehicle's side-view camera monitors the road surface material of the front and rear wheels on the current road, thus obtaining information about the current road surface material.

3. The all-terrain driving mode automatic switching method as described in claim 1, characterized in that, The step of calculating the material change edge distance and estimated material change time based on the ground edge information, the forward road material information, and the current road surface material information includes: The material change edge points are located based on the ground edge information, the forward road material information, and the current road surface material information. Calculate the linear spatial distance between the material change edge point and the center of the current vehicle in the vehicle coordinate system, and use the linear spatial distance as the material change edge distance; The real-time speed of the current vehicle is obtained, and the estimated time of material change is determined based on the real-time speed and the distance to the edge of the material change.

4. The all-terrain driving mode automatic switching method as described in claim 3, characterized in that, The step of calculating the linear spatial distance between the material change edge point and the center of the current vehicle in the vehicle coordinate system, and using the linear spatial distance as the material change edge distance, includes: By using preset camera calibration parameters, the pixel coordinates of the material edge points are mapped to the world coordinate system to obtain three-dimensional spatial points. The real-time pose information of the current vehicle is obtained, and the three-dimensional spatial points are transformed to the vehicle coordinate system based on the real-time pose information to obtain the three-dimensional coordinates of the material edge points in the vehicle coordinate system. The straight-line spatial distance is calculated based on the three-dimensional coordinates, and the straight-line spatial distance is used as the material change edge distance.

5. The all-terrain driving mode automatic switching method as described in claim 3, characterized in that, The step of obtaining the real-time speed of the current vehicle and determining the estimated time of material change based on the real-time speed and the distance to the edge of the material change includes: Obtain the real-time speed of the current vehicle, and determine the estimated time of material change based on the real-time speed and the distance to the edge of the material change using the following formula: in, To estimate the time required for material changes, Distance at the edge of material change This is the real-time vehicle speed.

6. The all-terrain driving mode automatic switching method as described in claim 1, characterized in that, The step of determining the estimated response adjustment amount based on the material change edge distance and the estimated material change time, and controlling the current vehicle to automatically switch to all-terrain driving mode based on the estimated response adjustment amount, includes: Based on the material change edge distance and the estimated material change time, a preset dynamic control algorithm is used to estimate the estimated response adjustment amount of the chassis system of the current vehicle to perform adjustment parameters. Before the estimated time for material change arrives, the chassis system is controlled to perform the estimated response adjustment so that the current vehicle can automatically switch from the current mode to the all-terrain driving mode.

7. The all-terrain driving mode automatic switching method as described in claim 6, characterized in that, The step of estimating the predicted response adjustment amount of the chassis system of the current vehicle to perform adjustment parameters based on the material change edge distance and the estimated material change time using a preset dynamic control algorithm includes: The adaptive PID controller takes the distance of the material change edge as the core error input and dynamically calculates the output value of the current core error after proportional gain weighting. The steady-state deviation is eliminated by accumulating historical errors through the integral term, and the trend of the output value is predicted through the differential term. Reinforcement learning is performed based on the changing trend and the estimated time of material change to output a smooth sequence. The smooth sequence is then weighted and fused according to the preset road condition change level weights to obtain the estimated response adjustment amount of the chassis system of the current vehicle to perform adjustment parameters.

8. An all-terrain driving mode automatic switching device, characterized in that, The all-terrain driving mode automatic switching device includes: The image segmentation module is used to acquire monitoring images of the current environment of the vehicle through a camera, and to segment the monitoring images to obtain ground edge information, forward road material information and current road surface material information. The data calculation module is used to calculate the material change edge distance and the estimated material change time based on the ground edge information, the forward road material information and the current road surface material information. The automatic mode switching module is used to determine the estimated response adjustment amount based on the material change edge distance and the estimated material change time, and to control the current vehicle to automatically switch to all-terrain driving mode based on the estimated response adjustment amount.

9. An all-terrain driving mode automatic switching device, characterized in that, The all-terrain driving mode automatic switching device includes: a memory, a processor, and an all-terrain driving mode automatic switching program stored in the memory and executable on the processor, the all-terrain driving mode automatic switching program being configured to implement the steps of the all-terrain driving mode automatic switching method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores an all-terrain driving mode automatic switching program, which, when executed by a processor, implements the steps of the all-terrain driving mode automatic switching method as described in any one of claims 1 to 7.

Images