A method of controlling travel of a vehicle and a vehicle

Abstract

This application provides a method and vehicle for controlling vehicle movement, relating to the field of intelligent driving technology. In this method, a first feature value, a second feature value, and a third feature value are determined based on road surface images, spatial observation data, and driving parameters. These three feature values ​​replicate the actual road conditions from three dimensions: the planar geometry of the road surface, the distribution of obstacles, and the longitudinal continuity. This eliminates the deviation in passability caused by the driver's subjective judgment. A passability score is calculated based on the three feature values, accurately quantifying the passability of the road surface. When the passability score is high, the vehicle is allowed to continue using the original driving parameters to pass through the road surface. In other words, this method uses a comprehensive and objective road surface measurement index to replace the limitations of the human eye in detecting hidden potholes, local surface curvature, and scattered obstacles. This covers detailed road conditions to achieve accurate assessment of the passability of the road surface, thus avoiding safety hazards caused by misjudgment.

Description

Technical Field

[0001] This application relates to the field of intelligent driving technology, and more specifically, to a method for controlling vehicle driving and a vehicle in the field of intelligent driving technology. Background Technology

[0002] Currently, vehicles may encounter complex road sections such as potholes, steep slopes, muddy sections, and gravel sections during operation. Under normal circumstances, drivers cannot accurately judge the vehicle's passability. If they force their way through, it may cause dangerous situations such as scraping vehicle parts, tire slippage, or getting stuck. In severe cases, it may even lead to safety hazards such as loss of vehicle control.

[0003] Therefore, there is an urgent need for a method to control vehicle movement in order to accurately assess the passability of the road surface ahead and avoid safety hazards caused by misjudging the road surface. Summary of the Invention

[0004] This application provides a method for controlling vehicle movement and a vehicle. The method can accurately assess the passability of the road surface in front of the vehicle, avoiding safety hazards caused by misjudging the road surface.

[0005] In a first aspect, a method for controlling vehicle driving is provided, the method comprising: while the vehicle is in motion, determining a first feature value, a second feature value, and a third feature value based on a road surface image, spatial observation data, and the vehicle's driving parameters, wherein the first feature value indicates the surface curvature of the road surface, the second feature value indicates the distribution density of obstacle areas on the road surface, and the third feature value indicates the continuity of the road surface in the vertical direction; determining a passability score based on the first feature value, the second feature value, and the third feature value, wherein the passability score indicates the passability of the road surface; and controlling the vehicle to pass through the road surface with the current driving parameters if the passability score is greater than a first preset score.

[0006] In the above technical solution, based on road surface images (two-dimensional image data), spatial observation data (three-dimensional position data), and vehicle driving parameters, three feature values—a first feature value, a second feature value, and a third feature value—are determined. The first feature value quantifies the surface curvature of the road surface, the second feature value quantifies the density of obstacle areas, and the third feature value quantifies the continuity of the road surface in the vertical direction. These three feature values ​​comprehensively replicate the actual road conditions from three dimensions: the planar geometry of the road surface, the distribution of obstacles, and the longitudinal continuity. This eliminates the possibility of deviations in passability due to the driver's subjective judgment. Furthermore, a passability score is calculated based on the three feature values, accurately quantifying the passability of the road surface. When the passability score is high, the vehicle is allowed to traverse the road surface using its original driving parameters. In other words, the above solution replaces the limitations of the human eye in detecting hidden potholes, localized surface curvature, and scattered obstacles with a comprehensive and objective road surface measurement index. This covers detailed road conditions to achieve accurate assessment of the road's passability (degree of passability), avoiding problems such as tire slippage or chassis scraping caused by misjudgment. It prevents safety hazards caused by misjudgment when traversing the road surface, and maintains the original driving parameters when road conditions meet the standards, eliminating the need for frequent adjustments to power and speed, thus simplifying the vehicle's control logic. Therefore, the above solution can balance driving safety and ride smoothness.

[0007] In conjunction with the first aspect, in some possible implementations, the road surface image consists of two adjacent frames, including a first frame and a second frame. The spatial observation data comprises the spatial position parameters of each measured point on the road surface relative to the vehicle-mounted radar. Based on the road surface image, the spatial observation data, and the vehicle's driving parameters, a first feature value, a second feature value, and a third feature value are determined, including: determining the depth value corresponding to each pixel in the two frames based on the spatial observation data; for each first pixel in the first frame, determining the second pixel corresponding to the imaging position of the first pixel in the second frame, obtaining multiple sets of pixels, each set including a first pixel and a corresponding second pixel; and determining the lateral offset value corresponding to each set of pixels in the two frames. The vertical offset value and depth value are used to determine the first optical flow component in the horizontal direction, the second optical flow component in the vertical direction, and the third optical flow component in the depth direction for each group of pixels. This yields each first optical flow component, its corresponding second optical flow component, and its corresponding third optical flow component. The optical flow component represents the positional offset of each group of pixels. Based on this driving parameter, each first optical flow component, its corresponding second optical flow component, and its corresponding third optical flow component are corrected to obtain each first corrected optical flow component, its corresponding second corrected optical flow component, and its corresponding third corrected optical flow component. Based on the first corrected optical flow component, its corresponding second corrected optical flow component, and its corresponding third corrected optical flow component for each group of pixels, the first feature value, the second feature value, and the third feature value are determined.

[0008] In the above technical solution, spatial observation data of the driving road surface collected by vehicle-mounted radar is used to match the corresponding depth values ​​of each pixel in two frames of images. This compensates for the lack of spatial scale in two-dimensional images and avoids the limitation that two-dimensional images cannot estimate the undulations and elevation changes of the driving road surface in the height direction. Furthermore, multiple sets of pixels with corresponding imaging positions are matched based on the two frames of images, and three-dimensional optical flow components (first optical flow component, second optical flow component, and third optical flow component) are determined. Then, based on driving parameters, the three-dimensional optical flow components are corrected. This corrects the optical flow distortion caused by vehicle motion, resulting in corrected optical flow components that eliminate interference from vehicle motion, which can realistically map the shape changes of the driving road surface. Subsequently, based on the corrected optical flow components, the above three feature values ​​are calculated. This improves the accuracy of feature value determination, thereby ensuring the reliability of subsequent passability and reducing driving malfunctions caused by road surface misjudgment.

[0009] In conjunction with the first aspect and the above implementation methods, in some possible implementation methods, the driving parameter includes acceleration. Based on the driving parameter, each first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component are corrected to obtain each first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component. This includes: when the acceleration is a preset acceleration, determining each first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component as each first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component; when the acceleration is greater than the preset acceleration, correcting each first optical flow component based on a first reduction magnitude to obtain the corresponding first corrected optical flow component, and correcting based on a second reduction magnitude... The first reduction magnitude, the second reduction magnitude, and the third reduction magnitude are positively correlated with the acceleration. When the acceleration is less than the preset acceleration, the first optical flow component is corrected based on the first compensation magnitude to obtain the corresponding first corrected optical flow component. The second optical flow component is corrected based on the second compensation magnitude to obtain the corresponding second corrected optical flow component. The third optical flow component is corrected based on the third compensation magnitude to obtain the corresponding third corrected optical flow component. The first compensation magnitude, the second compensation magnitude, and the third compensation magnitude are negatively correlated with the acceleration.

[0010] In the above technical solution, the process of correcting each optical flow component based on driving parameters, including acceleration, is described using driving parameters. When the acceleration is at a preset acceleration, the original first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component are directly used, which saves unnecessary calculations. When the acceleration is greater than the preset acceleration, each optical flow component is corrected based on the reduction amplitude that increases synchronously with the acceleration to offset the artificial increase in optical flow caused by vehicle movement (rapid forward movement). When the acceleration is less than the preset acceleration, each optical flow component is corrected based on the compensation amplitude that increases synchronously with the decrease in acceleration to compensate for the missing optical flow components caused by vehicle movement (slow movement). This solution, based on acceleration, can mitigate the optical flow measurement deviation caused by vehicle movement, ensuring that the corrected optical flow components only retain the positional offset caused by road surface deformation, making the determination of the above three characteristic values ​​closely match the actual road conditions.

[0011] In conjunction with the first aspect and the above implementation methods, in some possible implementation methods, determining the first feature value, the second feature value, and the third feature value based on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels includes: determining a first modulus value based on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels to obtain multiple first modulus values, which are used to indicate the movement rate of the corresponding group of pixels; and determining the first feature value, the second feature value, and the third feature value based on the first corrected optical flow component, the corresponding second corrected optical flow component, the corresponding third corrected optical flow component, and the corresponding first modulus value of each group of pixels.

[0012] In the above technical solution, the first modulus value is determined based on the three types of corrected optical flow components of each group of pixels. This quantifies the offset strength (movement rate) of the corresponding group of pixels through the modulus value, thus avoiding the susceptibility to disturbances when a single component is individually valued. Subsequently, the three feature values ​​are determined based on the three types of corrected optical flow components and the corresponding first modulus values ​​of each group of pixels. This can constrain the calculation error caused by abnormal fluctuations in the optical flow components by using the movement rate, distinguishing between road surface deformation and data mutations caused by sporadic pixel noise, so that the obtained three feature values ​​can accurately reproduce the actual road conditions and ensure the reliability of subsequent passability.

[0013] In conjunction with the first aspect and the above implementation methods, in some possible implementation methods, determining the first feature value, the second feature value, and the third feature value based on the first corrected optical flow component, the corresponding second corrected optical flow component, the corresponding third corrected optical flow component, and the corresponding first modulus value of each group of pixels includes: for the first group of pixels in the plurality of pixels, determining the ratio between the first corrected optical flow component and the corresponding first modulus value of the first group of pixels, determining the ratio between the second corrected optical flow component and the corresponding first modulus value of the first group of pixels, and determining the ratio between the third corrected optical flow component and the corresponding first modulus value of the first group of pixels, thereby obtaining the normalized horizontal optical flow component of the first group of pixels. The algorithm calculates the normalized vertical optical flow component and the normalized depth optical flow component; performs divergence calculation on the normalized horizontal optical flow component, the normalized vertical optical flow component, and the normalized depth optical flow component to obtain the first feature value of the first group of pixels, thereby obtaining the first feature value of the multiple groups of pixels; performs divergence calculation on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels to determine the second feature value of each group of pixels, thereby obtaining the second feature value of the multiple groups of pixels, and determining the distribution density of the obstacle area on the driving road surface; performs Laplacian calculation on the first modulus value of each group of pixels to determine the third feature value of each group of pixels, thereby obtaining the third feature value of the multiple groups of pixels.

[0014] In the above technical solution, the optical flow component is normalized by correcting the ratio between the optical flow component and the corresponding first modulus value. This eliminates the calculation deviation caused by the amplitude difference in the movement rate of corresponding groups of pixels, and unifies the calculation benchmark of the optical flow components of each group of pixels. Furthermore, divergence calculation is performed on the normalized horizontal optical flow component, normalized vertical optical flow component, and normalized depth optical flow component to determine the first feature value, accurately capturing the horizontal shape changes of the road surface. Divergence calculation is performed on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels to determine the second feature value of each group of pixels, accurately identifying the density of obstacle areas. Laplacian calculation is performed on the first modulus value of each group of pixels to obtain the third feature value corresponding to the continuity of the road surface in the vertical direction, effectively perceiving the vertical structural changes of the road surface. The above solution adopts different determination methods for the three feature values ​​to adapt to different feature value extraction needs, avoid the feature value aliasing problem, and improve the accuracy and independence of the three feature value extraction.

[0015] Combining the first aspect and the above implementation methods, in some possible implementation methods, performing divergence calculations on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels to determine the second feature value of each group of pixels includes: performing divergence calculations on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels to obtain multiple divergence values, and determining the maximum divergence value from the multiple divergence values, wherein the multiple divergence values ​​are used to indicate the degree of aggregation of multiple groups of pixels when the position is offset; for the first group of pixels in the multiple groups of pixels, determining the ratio between the divergence value of the first group of pixels and the maximum divergence value to obtain the normalized divergence value of the first group of pixels; and determining the difference between the first preset value and the normalized divergence value of the first group of pixels as the second feature value of the first group of pixels to obtain the second feature value of each group of pixels.

[0016] In the above technical solution, divergence calculations are performed on the three types of corrected optical flow components of each group of pixels to characterize the degree of aggregation of multiple groups of pixels when their positions are offset. Furthermore, the divergence values ​​of each group of pixels are normalized based on the maximum divergence value, which eliminates computational imbalances caused by differences in the magnitude of divergence values ​​under different road surface scenarios. Subsequently, the difference between the first preset value and the normalized divergence value is determined as the second feature value of each group of pixels. This allows the density of obstacle areas to be mapped based on the physical meaning of the divergence value, avoiding the drawbacks of feature distortion caused by changes in acquisition distance, and accurately quantifying the distribution density of obstacle areas, so that the second feature value can objectively reflect the arrangement of obstacle areas.

[0017] In conjunction with the first aspect and the above-described implementation, in some possible implementations, determining the passability score based on the first feature value, the second feature value, and the third feature value includes: determining a first score, a second score, and a third score based on the first feature value, the second feature value, and the third feature value, wherein the first score indicates the passability affected by the degree of surface curvature, the second score indicates the passability affected by the distribution density of obstacle areas, and the third score indicates the passability affected by the degree of continuity in the height direction; determining a first weight, a second weight, and a third weight based on the vehicle's current driving parameters, wherein the first weight indicates the degree of trust in the first feature value when determining the passability score, the second weight indicates the degree of trust in the second feature value when determining the passability score, and the third weight indicates the degree of trust in the third feature value when determining the passability score; and weighting and fusing the first score, the second score, and the third score based on the first weight, the second weight, and the third weight to obtain the passability score.

[0018] In the above technical solution, the three feature values ​​are converted into individual scores (first score, second score, and third score). This separates the impact of three road conditions on trafficability, avoiding interference between different road condition parameters in the trafficability assessment. Subsequently, based on the vehicle's current driving parameters, the three weights are dynamically matched. This adjusts the trust weight of each road condition based on real-time vehicle conditions, addressing the evaluation bias caused by fixed weights failing to adapt to changes in vehicle speed. Furthermore, the individual scores are weighted and fused based on the three weights to obtain a trafficability score. This integrates multi-dimensional road conditions to form a unified standard for trafficability assessment, optimizing the shortcomings of single-indicator judgments and making the trafficability assessment results more closely reflect the actual driving conditions of the vehicle.

[0019] In conjunction with the first aspect and the above-described implementation, in some possible implementations, the road surface is divided into multiple road surface regions. The second feature value includes a first distribution feature value and a second distribution feature value. Based on the first feature value, the second feature value, and the third feature value, a first score, a second score, and a third score are determined, including: determining the first score based on the feature difference between a first preset feature value and the first feature value; determining the ratio between the first distribution feature value and the second distribution feature value as a first value, where the first value indicates the uniformity of the distribution of obstacle regions on the road surface, the first distribution feature value indicates the average distribution density of obstacle regions in the multiple road surface regions, and the second distribution feature value indicates the maximum distribution density of obstacle regions in the multiple road surface regions; determining the second score based on the difference between the first preset value and the first value, where the first preset value indicates that the distribution density of obstacle regions in the multiple road surface regions after the road surface is divided is consistent; and performing an exponential transformation on the opposite feature value of the third feature value to obtain the third score.

[0020] In the above technical solution, a first score is determined based on the feature difference between a first preset feature value and a first feature value. This allows for the use of a smooth road surface as a reference standard to quantify the traffic loss caused by surface curvature. The ratio between the average and maximum distribution density of the obstacle area is used to characterize the uniformity of the obstacle area's distribution. Then, a second score is determined based on the difference between a first preset value used to characterize uniform distribution density and a first value. This takes into account both the overall density of obstacle areas on the road surface and the impact of local obstacle areas. Simultaneously, an exponential transformation is performed on the opposite feature value of the third feature value to determine the third score. This uses an exponential function to adjust for abnormal scores caused by extreme fluctuations. These three differentiated scoring methods are adapted to the three types of road conditions, avoiding the drawbacks of a unified scoring formula that cannot match the characteristics of different road condition indicators, and improving the accuracy of the three scores (three individual scores).

[0021] In conjunction with the first aspect and the above implementation methods, in some possible implementation methods, the current driving parameter of the vehicle is the vehicle speed. Based on the current driving parameter of the vehicle, determining a first weight, a second weight, and a third weight includes: when the vehicle speed is less than a first preset vehicle speed, determining the first weight, the second weight, and the third weight as a first preset weight, a second preset weight, and a third preset weight, respectively, wherein the second preset weight is greater than the first preset weight, and the first preset weight is greater than the third preset weight; when the vehicle speed is greater than or equal to the first preset vehicle speed and less than the second preset vehicle speed, determining the first weight, the second weight, and the third weight as a fourth preset weight, a fifth preset weight, and a sixth preset weight, wherein the fourth preset weight is greater than the fifth preset weight, and the sixth preset weight is greater than the fifth preset weight; when the vehicle speed is greater than or equal to the second preset vehicle speed, determining the first weight, the second weight, and the third weight as a seventh preset weight, an eighth preset weight, and a ninth preset weight, wherein the ninth preset weight is greater than the seventh preset weight, and the seventh preset weight is greater than the eighth preset weight.

[0022] In the above technical solution, the current driving parameters are used as the speed to describe the process of determining the first, second, and third weights based on these parameters. Specifically, three speed ranges are divided, each with its corresponding weight. When the speed is less than the first preset speed (i.e., the low-speed phase), the second weight corresponding to the distribution density of obstacle areas is increased to match the driving pattern where the vehicle is more likely to notice obstacles on the road surface at low speeds. When the speed is greater than or equal to the first preset speed but less than the second preset speed (i.e., the medium-speed phase), the first weight corresponding to the surface curvature and the third weight corresponding to the continuity are increased to match the rapid changes in road conditions caused by bumps when the vehicle turns at medium speeds. When the speed is greater than or equal to the second preset speed (i.e., the high-speed phase), the third weight is emphasized, which addresses the characteristic that road undulations can easily induce vehicle instability at high speeds. This solution uses speed-based weight switching instead of fixed weights, eliminating the problem of fixed weights failing to match the real-time driving state of the vehicle, ensuring that the weighted passability score is more accurate. In conjunction with the first aspect and the above implementation methods, in some possible implementation methods, the method further includes: when the passability score is less than or equal to the first preset score and greater than the second preset score, determining the risk type of the driving surface based on the first feature value, the second feature value and the third feature value, wherein the risk type is used to indicate the cause of the decrease in the vehicle's passability; adjusting the current driving parameters based on the risk type to obtain the adjusted driving parameters, and controlling the vehicle to pass through the driving surface with the adjusted driving parameters.

[0023] In the aforementioned technical solution, when the passability score is less than or equal to a first preset score but greater than a second preset score (passability score within the middle range), the risk type of the driving road surface is determined based on the first, second, and third characteristic values ​​to pinpoint the cause of the vehicle's decreased passability. That is, based on each characteristic value, the specific road condition source of the deteriorating passability is identified, abandoning the approach of generalized control. Furthermore, based on the risk type, the current driving parameters are specifically adjusted to address the shortcoming that, when the passability score is low, the original driving parameters (current driving parameters) can only directly allow passage or abruptly prohibit passage. In other words, this solution can achieve differentiated vehicle control by identifying the specific road condition source of the deteriorating passability, avoiding the safety hazards of tire slippage or chassis scraping, while also avoiding blindly and drastically reducing speed, which would affect traffic efficiency.

[0024] In conjunction with the first aspect and the above implementation methods, in some possible implementation methods, the method further includes: when the passability score is less than or equal to the second preset score, controlling the vehicle to output a first reminder message, the first reminder message being used to remind that the passability of the road surface ahead is lower than the preset passability and that passage is prohibited; controlling the vehicle to decelerate and stop in a safe road surface area; and marking the road segment corresponding to the road surface as a prohibited road segment, storing the location information and risk association information of the road segment to assist other vehicles in outputting a second reminder message when the distance between them and the road segment is less than a preset distance after a preset time period, the second reminder message being used to remind that the road segment is a prohibited road segment and suggesting detour.

[0025] In the aforementioned technical solution, when the passability score is less than or equal to the second preset score (extremely low passability score), a multi-level safety response mechanism is triggered. This involves outputting a warning message, promptly informing the driver that the current road surface's capacity is insufficient, preventing the driver from attempting to pass and causing safety hazards. Simultaneously, the vehicle is slowed down and brought to a safe road surface area to completely eliminate safety accidents such as getting stuck or losing control from a driving control perspective. Furthermore, the corresponding road segment is marked as a no-entry segment, and the location information and risk association information of the segment are stored. This allows for detour reminders to nearby vehicles, achieving a comprehensive shared early warning of risks for a single road segment. This overcomes the limitations of individual vehicles' autonomous risk avoidance, effectively preventing repeated occurrences of similar road conditions and significantly improving the safety and intelligent early warning capabilities of road sections.

[0026] In a second aspect, an apparatus for controlling vehicle movement is provided, the apparatus comprising: a determining module, configured to: determine a first feature value, a second feature value, and a third feature value based on a road surface image, spatial observation data, and the vehicle's driving parameters while the vehicle is in motion; the first feature value indicating the surface curvature of the road surface, the second feature value indicating the distribution density of obstacle areas on the road surface, and the third feature value indicating the continuity of the road surface in the vertical direction; determine a passability score based on the first feature value, the second feature value, and the third feature value, the passability score indicating the passability of the road surface; and a control module, configured to control the vehicle to pass through the road surface with current driving parameters if the passability score is greater than a first preset score.

[0027] In conjunction with the second aspect, in some possible implementations, the road surface image consists of two adjacent frames, including a first frame and a second frame. The spatial observation data comprises the spatial position parameters of each measured point on the road surface relative to the vehicle-mounted radar. The determining module is specifically used for: determining the depth value corresponding to each pixel in the two frames based on the spatial observation data; for each first pixel in the first frame, determining the second pixel corresponding to the imaging position of that first pixel in the second frame, resulting in multiple sets of pixels, each set including a first pixel and a corresponding second pixel; and determining the horizontal offset, vertical offset, and depth value of each set of pixels in the two frames based on the horizontal offset, vertical offset, and depth values ​​corresponding to each set of pixels in the two frames. The first optical flow component in the direction of travel, the second optical flow component in the vertical direction, and the third optical flow component in the depth direction are used to obtain each first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component. The optical flow component is the position offset of each group of pixels. Based on the driving parameter, each first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component are corrected to obtain each first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component. Based on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels, the first feature value, the second feature value, and the third feature value are determined.

[0028] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the driving parameter includes acceleration. The determining module is further configured to: when the acceleration is a preset acceleration, determine each first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component as each first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component; when the acceleration is greater than the preset acceleration, correct each first optical flow component based on a first reduction amplitude to obtain the corresponding first corrected optical flow component; correct each second optical flow component based on a second reduction amplitude to obtain the corresponding second corrected optical flow component; and correct based on the third... The reduction magnitude is used to correct each third optical flow component to obtain a corresponding third corrected optical flow component. The first reduction magnitude, the second reduction magnitude, and the third reduction magnitude are positively correlated with the acceleration. When the acceleration is less than the preset acceleration, each first optical flow component is corrected based on the first compensation magnitude to obtain a corresponding first corrected optical flow component. Based on the second compensation magnitude, each second optical flow component is corrected to obtain a corresponding second corrected optical flow component. Based on the third compensation magnitude, each third optical flow component is corrected to obtain a corresponding third corrected optical flow component. The first compensation magnitude, the second compensation magnitude, and the third compensation magnitude are negatively correlated with the acceleration.

[0029] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the determining module is further specifically used to: determine a first modulus value based on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels, to obtain multiple first modulus values, the first modulus values ​​being used to indicate the movement rate of the corresponding group of pixels; and determine the first feature value, the second feature value, and the third feature value based on the first corrected optical flow component, the corresponding second corrected optical flow component, the corresponding third corrected optical flow component, and the corresponding first modulus value of each group of pixels.

[0030] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the determining module is further configured to: for the first group of pixels in the plurality of pixels, determine the ratio between the first corrected optical flow component and the corresponding first modulus value of the first group of pixels, determine the ratio between the second corrected optical flow component and the corresponding first modulus value of the first group of pixels, and determine the ratio between the third corrected optical flow component and the corresponding first modulus value of the first group of pixels, thereby obtaining the normalized horizontal optical flow component, the normalized vertical optical flow component, and the normalized depth optical flow component of the first group of pixels; and for the normalized horizontal optical flow component... The first feature value of the first group of pixels is obtained by performing divergence operation on the normalized vertical optical flow component and the normalized depth optical flow component, so as to obtain the first feature value of the multiple groups of pixels; the second feature value of the first corrected optical flow component, the corresponding second corrected optical flow component and the corresponding third corrected optical flow component of each group of pixels is determined by performing divergence operation on the first corrected optical flow component, the corresponding second corrected optical flow component and the corresponding third corrected optical flow component of each group of pixels, so as to obtain the second feature value of the multiple groups of pixels, and to determine the distribution density of the obstacle area on the driving road surface; the third feature value of the first modulus value of each group of pixels is determined by performing Laplacian operation on the first modulus value of each group of pixels, so as to obtain the third feature value of the multiple groups of pixels.

[0031] In conjunction with the second aspect and the above implementation, in some possible implementations, the determining module is further configured to: perform divergence calculations on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels to obtain multiple divergence values, and determine the maximum divergence value from the multiple divergence values, wherein the multiple divergence values ​​are used to indicate the degree of aggregation of multiple groups of pixels when the position is offset; for the first group of pixels in the multiple groups of pixels, determine the ratio between the divergence value of the first group of pixels and the maximum divergence value to obtain the normalized divergence value of the first group of pixels; and determine the difference between the first preset value and the normalized divergence value of the first group of pixels as the second feature value of the first group of pixels, so as to obtain the second feature value of each group of pixels.

[0032] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the determining module is further configured to: determine a first score, a second score, and a third score based on the first feature value, the second feature value, and the third feature value, wherein the first score indicates the passability affected by the degree of surface curvature, the second score indicates the passability affected by the distribution density of the obstacle area, and the third score indicates the passability affected by the degree of continuity in the height direction; determine a first weight, a second weight, and a third weight based on the vehicle's current driving parameters, wherein the first weight indicates the degree of trust in the first feature value when determining the passability score, the second weight indicates the degree of trust in the second feature value when determining the passability score, and the third weight indicates the degree of trust in the third feature value when determining the passability score; and perform weighted fusion of the first score, the second score, and the third score based on the first weight, the second weight, and the third weight to obtain the passability score.

[0033] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the driving surface is divided into multiple road surface regions, and the second feature value includes a first distribution feature value and a second distribution feature value. The determining module is further configured to: determine the first score based on the feature difference between the first preset feature value and the first feature value; determine the ratio between the first distribution feature value and the second distribution feature value as a first value, the first value being used to indicate the uniformity of the distribution of obstacle regions on the driving surface, the first distribution feature value being used to indicate the average distribution density of obstacle regions in the multiple road surface regions, and the second distribution feature value being used to indicate the maximum distribution density of obstacle regions in the multiple road surface regions; determine the second score based on the difference between the first preset value and the first value, the first preset value being used to indicate that the distribution density of obstacle regions in the multiple road surface regions after the driving surface is divided is consistent; and perform an exponential transformation on the opposite feature value of the third feature value to obtain the third score.

[0034] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the current driving parameter of the vehicle is the vehicle speed. The determining module is further configured to: when the vehicle speed is less than a first preset vehicle speed, determine the first weight, the second weight, and the third weight as a first preset weight, a second preset weight, and a third preset weight, respectively, wherein the second preset weight is greater than the first preset weight, and the first preset weight is greater than the third preset weight; when the vehicle speed is greater than or equal to the first preset vehicle speed and less than the second preset vehicle speed, determine the first weight, the second weight, and the third weight as a fourth preset weight, a fifth preset weight, and a sixth preset weight, wherein the fourth preset weight is greater than the fifth preset weight, and the sixth preset weight is greater than the fifth preset weight; when the vehicle speed is greater than or equal to the second preset vehicle speed, determine the first weight, the second weight, and the third weight as a seventh preset weight, an eighth preset weight, and a ninth preset weight, wherein the ninth preset weight is greater than the seventh preset weight, and the seventh preset weight is greater than the eighth preset weight.

[0035] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the determining module is further configured to determine the risk type of the driving surface based on the first feature value, the second feature value, and the third feature value when the passability score is less than or equal to the first preset score and greater than the second preset score. The risk type is used to indicate the cause of the decrease in the vehicle's passability. The control module is further configured to adjust the current driving parameters based on the risk type to obtain the adjusted driving parameters, and control the vehicle to pass through the driving surface with the adjusted driving parameters.

[0036] In conjunction with the second aspect and the above implementation methods, in some possible implementation methods, the control module is further configured to: control the vehicle to output a first reminder message when the passability score is less than or equal to the second preset score, the first reminder message being used to remind that the passability of the road surface ahead is lower than the preset passability and that passage is prohibited; and control the vehicle to decelerate and stop in a safe road surface area; and the device further includes: an output module, configured to mark the road segment corresponding to the road surface as a prohibited road segment, and store the location information and risk association information of the road segment to assist other vehicles in outputting a second reminder message when the distance between them and the road segment is less than a preset distance after a preset time period, the second reminder message being used to remind that the road segment is a prohibited road segment and to suggest detour.

[0037] Thirdly, a vehicle is provided, including a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, causing the vehicle to perform the methods described in the first aspect or any possible implementation thereof. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of a scenario for controlling vehicle movement provided in an embodiment of this application; Figure 2 This is a schematic flowchart illustrating a method for controlling vehicle movement provided in an embodiment of this application; Figure 3 This is a schematic flowchart illustrating another method for controlling vehicle movement provided in an embodiment of this application; Figure 4 This is a schematic flowchart illustrating another method for controlling vehicle movement provided in an embodiment of this application; Figure 5 This is a schematic diagram of a device for controlling vehicle movement provided in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of a controller provided in an embodiment of this application; Figure 7 This is a schematic diagram of the structure of a vehicle provided in an embodiment of this application. Detailed Implementation

[0039] The technical solutions in this application will be clearly and thoroughly described below with reference to the accompanying drawings. In the description of the embodiments of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. "And / or" in the text is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, in the description of the embodiments of this application, "multiple" refers to two or more than two.

[0040] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0041] Currently, vehicles may encounter complex road sections such as potholes, steep slopes, muddy sections, and gravel sections during operation. Figure 1 As shown, a driver is driving vehicle A and encounters a muddy section of road ahead. Typically, drivers judge whether they can pass through the muddy section by sight, which is usually inaccurate due to the uncertainty of the driver's subjective experience.

[0042] If a driver attempts to pass through the aforementioned complex road sections, it may result in dangerous situations such as scratches on vehicle parts, tire slippage, or getting stuck. In severe cases, it may even lead to safety hazards such as loss of vehicle control.

[0043] To address the aforementioned problems, this application proposes a method for controlling vehicle movement to accurately assess the passability of the road surface ahead, avoiding forced passage through complex road sections and thus preventing safety hazards. Specific implementation steps are as follows: Figure 2 .

[0044] Figure 2 This is a schematic flowchart illustrating a method for controlling vehicle movement provided in an embodiment of this application.

[0045] It should be understood that the method for controlling vehicle movement provided in this application embodiment can be applied to, for example... Figure 1 The vehicle shown (e.g., vehicle A) specifically illustrates that the method for controlling vehicle movement can be applied to a vehicle controller.

[0046] For example, such as Figure 2 As shown, the method 200 includes the following steps 201 to 203.

[0047] Step 201: While the vehicle is in motion, based on the road surface image, spatial observation data and the vehicle's driving parameters, determine a first feature value, a second feature value and a third feature value. The first feature value is used to indicate the surface curvature of the road surface, the second feature value is used to indicate the distribution density of obstacle areas on the road surface, and the third feature value is used to indicate the continuity of the road surface in the height direction.

[0048] It should be understood that in step 201 above, the road surface image refers to the image of the road surface ahead of the vehicle, collected by the vehicle's vision sensor. The vision sensor collects one road surface image at preset time intervals, meaning the road surface image is specifically a continuous frame image, i.e., multiple images. Spatial observation data refers to the spatial position parameters of each measured point on the road surface relative to the vehicle radar, collected by the vehicle radar (which can be a millimeter-wave radar), including distance, azimuth angle, and pitch angle. Among them, distance refers to the straight-line distance between the measured point and the vehicle radar, azimuth angle refers to the left and right deviation angle of the measured point relative to the central axis of the vehicle radar, and pitch angle refers to the height angle formed by the line of sight upwards / downwards relative to the horizontal mounting surface of the vehicle radar.

[0049] It should also be understood that at the same time, a road surface image corresponds to a spatial observation data and a driving parameter, that is, the above three data (road surface image, spatial observation data and driving parameter) should be time-aligned.

[0050] Optionally, the driving parameters include acceleration, yaw rate, pitch rate, and roll rate.

[0051] Optionally, the vision sensor is a camera, and the vehicle radar is a millimeter-wave radar.

[0052] It should also be noted that the aforementioned surface curvature focuses on the horizontal (lateral) morphological changes of the road surface, characterizing the undulations and local warping deformation along the vehicle width. This quantifies the inherent unevenness, curvature, and torsional deformation of the pavement itself, and is independent of road alignment, emphasizing abrupt changes in the road surface's contours. The aforementioned distribution density focuses on the spatial distribution of obstacle areas, statistically analyzing the density of potholes, obstacles, and other obstacles. Specifically, the distribution density can be the convergence density of obstacle areas. The aforementioned continuity focuses on the vertical structural changes of the road surface, emphasizing the discontinuity in the height direction of steps, steep slopes, etc., to measure the smoothness of the road surface's elevation transitions.

[0053] In other words, the difference between the first and third characteristic values ​​mentioned above is that the first characteristic value reflects the degree of surface curvature of the road surface in the horizontal direction (lateral direction), while the third characteristic value reflects the continuity of the road surface during transitions in the vertical direction (height direction). Furthermore, the values ​​of the first, second, and third characteristic values ​​range from 0 to 1.

[0054] Furthermore, 1) Spatial observation data: Regarding the third eigenvalue, multiple road surface images can only output pixel displacements on a two-dimensional plane, unable to directly perceive depth information. Spatial observation data provides an absolute three-dimensional coordinate system, allowing the relative motion vectors of the two-dimensional plane to be mapped to the real three-dimensional space, determining the elevation difference of the road surface in the height direction (Z-axis). Regarding the first and second eigenvalues, spatial observation data provides depth values ​​corresponding to each pixel in the road surface image. Only by combining spatial observation data can the vehicle determine the distribution density of obstacle areas in the real physical world, rather than just the pixel area in the road surface image. 2) Driving parameters: Regarding the first and second eigenvalues, the vehicle is a moving observation platform during driving. The road surface changes captured by the road surface image are the result of superimposing the vehicle's motion with the objective shape of the road surface. Only by using these driving parameters as compensation factors can the visual influence caused by the vehicle's motion be eliminated, and the relative motion components caused purely by the road surface geometry be separated, thereby accurately calculating the surface curvature and the distribution density of obstacle areas. Regarding the third eigenvalue, spatial observation data is essentially the instantaneous spatial state collected by the vehicle when passing the road surface at different time points. Focusing solely on spatial observation data only determines when a vehicle experiences a bumpy ride, but not the duration of that bumpy ride on the road surface. By incorporating driving parameters, the principle that "distance equals speed multiplied by time" can be used to convert the time intervals recorded by the onboard radar into the actual distance the vehicle travels on the road surface. Through this spatiotemporal transformation, the true physical span of potholes or slopes ahead can be determined, thus revealing the degree of continuity of the road surface in the vertical direction.

[0055] In one possible implementation, the road surface image consists of two adjacent frames, including a first frame and a second frame. The spatial observation data comprises the spatial position parameters of each measured point on the road surface relative to the vehicle-mounted radar. Step 201, based on the road surface image, spatial observation data, and the vehicle's driving parameters, determines the first, second, and third feature values, including: determining the depth value corresponding to each pixel in the two frames based on the spatial observation data; for each first pixel in the first frame, determining the second pixel corresponding to the imaging position of the first pixel in the second frame, resulting in multiple sets of pixels, each set including a first pixel and a corresponding second pixel; and determining the lateral offset value corresponding to each set of pixels in the two frames. The vertical offset value and depth value are used to determine the first optical flow component in the horizontal direction, the second optical flow component in the vertical direction, and the third optical flow component in the depth direction for each group of pixels. This yields each first optical flow component, its corresponding second optical flow component, and its corresponding third optical flow component. The optical flow component represents the positional offset of each group of pixels. Based on this driving parameter, each first optical flow component, its corresponding second optical flow component, and its corresponding third optical flow component are corrected to obtain each first corrected optical flow component, its corresponding second corrected optical flow component, and its corresponding third corrected optical flow component. Based on the first corrected optical flow component, its corresponding second corrected optical flow component, and its corresponding third corrected optical flow component for each group of pixels, the first feature value, the second feature value, and the third feature value are determined.

[0056] It should be understood that in the above scheme, the first pixel can be regarded as a specific point on the previous frame image (the first frame image) (such as a pixel corresponding to the ball in the image), and the second pixel can be regarded as the pixel corresponding to the new position left on the image in the next frame image (the second frame image) after the ball moves. This is to track the positional change of the same object or physical point in two consecutive frames. The depth value corresponding to the above pixels can reflect the distance between the vision sensor and the corresponding measured point on the road surface.

[0057] In the above technical solution, spatial observation data of the driving road surface collected by vehicle-mounted radar is used to match the corresponding depth values ​​of each pixel in two frames of images. This compensates for the lack of spatial scale in two-dimensional images and avoids the limitation that two-dimensional images cannot estimate the undulations and elevation changes of the driving road surface in the height direction. Furthermore, multiple sets of pixels with corresponding imaging positions are matched based on the two frames of images, and three-dimensional optical flow components (first optical flow component, second optical flow component, and third optical flow component) are determined. Then, based on driving parameters, the three-dimensional optical flow components are corrected. This corrects the optical flow distortion caused by vehicle motion, resulting in corrected optical flow components that eliminate interference from vehicle motion, which can realistically map the shape changes of the driving road surface. Subsequently, based on the corrected optical flow components, the above three feature values ​​are calculated. This improves the accuracy of feature value determination, thereby ensuring the reliability of subsequent passability and reducing driving malfunctions caused by road surface misjudgment.

[0058] In some embodiments, determining the depth value corresponding to each pixel in the two frames of images based on the spatial observation data includes: for any time and any measured point, converting the spatial observation data into first observation data in the radar coordinate system, the first observation data including first horizontal data, first vertical data, and first vertical data; converting the first observation data into second observation data in the camera coordinate system, the second observation data including second horizontal data, second vertical data, and second vertical data; substituting the second horizontal data, second vertical data, and second vertical data into the pinhole imaging function to obtain third horizontal data, third vertical data, and third vertical data; comparing the third horizontal data and third vertical data with the horizontal offset value and vertical offset value of each pixel in the image corresponding to that time to determine the target horizontal offset value and target vertical offset value that match the third horizontal data and third vertical data; and determining the depth value of the pixel corresponding to the target horizontal offset value and the target vertical offset value as the third vertical data.

[0059] In some embodiments, the spatial observation data is converted into first observation data in the radar coordinate system, including: determining the first observation data based on the following formula (1); (1) in, This is the first horizontal data. This is the first vertical data. This is the first vertical data. To correspond to the distance of the measured point relative to the vehicle-mounted radar, To correspond to the azimuth angle of the measured point relative to the vehicle-mounted radar, This corresponds to the elevation angle of the measured point relative to the vehicle-mounted radar.

[0060] In some embodiments, the first observation data is converted into second observation data in the camera coordinate system, including: determining the second observation data based on the following formula (2); (2) in, This is the second horizontal data. This is the second vertical data. This is the second vertical data. and These are the extrinsic parameters of a visual sensor (such as a camera), where the former is the rotation matrix and the latter is the translation vector.

[0061] In some embodiments, the second lateral data, the second longitudinal data, and the second vertical data are substituted into the pinhole imaging function to obtain the third lateral data, the third longitudinal data, and the third vertical data, including: determining the third lateral data, the third longitudinal data, and the third vertical data based on the following formula (3); (3) in, For the third horizontal data ， This is the third vertical data. This is the third vertical data. and The focal lengths of a visual sensor (such as a camera) in the horizontal and vertical directions, respectively. and These are the horizontal and vertical coordinates of the point where the optical axis of the vision sensor intersects the vision sensor perpendicularly in the pixel coordinate system.

[0062] It should be understood that the pixel grayscale of the same object or physical point remains unchanged in two consecutive frames of images, and the object or physical point only undergoes spatial movement. Therefore, the corresponding second pixel can be matched to the first pixel based on the pixel grayscale.

[0063] In some embodiments, determining a first optical flow component in the horizontal direction, a second optical flow component in the vertical direction, and a third optical flow component in the depth direction for each group of pixels based on the horizontal offset value, vertical offset value, and depth value corresponding to each group of pixels in the two frames of images includes: taking any group of pixels as a first group of pixels, the first group of pixels including a first target pixel and a second target pixel; determining the first optical flow component in the horizontal direction of the first group of pixels based on the deviation between the horizontal offset value corresponding to the second target pixel and the horizontal offset value corresponding to the first target pixel; determining the second optical flow component in the vertical direction of the first group of pixels based on the deviation between the vertical offset value corresponding to the second target pixel and the vertical offset value corresponding to the first target pixel; and determining the third optical flow component in the depth direction of the first group of pixels based on the deviation between the depth value corresponding to the second target pixel and the depth value corresponding to the first target pixel.

[0064] In one possible implementation, the driving parameter includes acceleration. Based on the driving parameter, each first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component are corrected to obtain each first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component. This includes: when the acceleration is a preset acceleration, determining each first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component as each first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component; when the acceleration is greater than the preset acceleration, correcting each first optical flow component based on a first reduction magnitude to obtain the corresponding first corrected optical flow component; and correcting each second optical flow component based on a second reduction magnitude. The optical flow components are corrected to obtain corresponding second corrected optical flow components. Based on the third reduction amplitude, each third optical flow component is corrected to obtain corresponding third corrected optical flow components. The first reduction amplitude, the second reduction amplitude, and the third reduction amplitude are positively correlated with the acceleration. When the acceleration is less than the preset acceleration, each first optical flow component is corrected based on the first compensation amplitude to obtain corresponding first corrected optical flow components. Based on the second compensation amplitude, each second optical flow component is corrected to obtain corresponding second corrected optical flow components. Based on the third compensation amplitude, each third optical flow component is corrected to obtain corresponding third corrected optical flow components. The first compensation amplitude, the second compensation amplitude, and the third compensation amplitude are negatively correlated with the acceleration.

[0065] It should be understood that in the above scheme, the preset acceleration is 0 m / s². 2The first, second, and third reduction amplitudes are all less than 1, with the second reduction amplitude being greater than the first, and the first being greater than the third. This means the reduction in each third optical flow component is the largest, followed by the reduction in each first optical flow component, and finally the reduction in each second optical flow component. The first, second, and third compensation amplitudes are all greater than 1, with the third compensation amplitude being greater than the first, and the third being greater than the second. This means the compensation for each third optical flow component is the strongest.

[0066] Furthermore, the smaller the acceleration (positive value), the smaller the reduction; the smaller the acceleration (negative value), the larger the compensation.

[0067] In the above technical solution, the process of correcting each optical flow component based on driving parameters, including acceleration, is described using driving parameters. When the acceleration is at a preset acceleration, the original first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component are directly used, which saves unnecessary calculations. When the acceleration is greater than the preset acceleration, each optical flow component is corrected based on the reduction amplitude that increases synchronously with the acceleration to offset the artificial increase in optical flow caused by vehicle movement (rapid forward movement). When the acceleration is less than the preset acceleration, each optical flow component is corrected based on the compensation amplitude that increases synchronously with the decrease in acceleration to compensate for the missing optical flow components caused by vehicle movement (slow movement). This solution, based on acceleration, can mitigate the optical flow measurement deviation caused by vehicle movement, ensuring that the corrected optical flow components only retain the positional offset caused by road surface deformation, making the determination of the above three characteristic values ​​closely match the actual road conditions.

[0068] In one possible implementation, determining the first feature value, the second feature value, and the third feature value based on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels includes: determining a first modulus value based on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels to obtain a plurality of first modulus values, the first modulus values ​​being used to indicate the movement rate of the corresponding group of pixels; and determining the first feature value, the second feature value, and the third feature value based on the first corrected optical flow component, the corresponding second corrected optical flow component, the corresponding third corrected optical flow component, and the corresponding first modulus value of each group of pixels.

[0069] It should be understood that in the above scheme, any group of pixels corresponds to a first corrected optical flow component, a second corrected optical flow component, a third corrected optical flow component, and a first modulus value.

[0070] In the above technical solution, the first modulus value is determined based on the three types of corrected optical flow components of each group of pixels. This quantifies the offset strength (movement rate) of the corresponding group of pixels through the modulus value, thus avoiding the susceptibility to disturbances when a single component is individually valued. Subsequently, the three feature values ​​are determined based on the three types of corrected optical flow components and the corresponding first modulus values ​​of each group of pixels. This can constrain the calculation error caused by abnormal fluctuations in the optical flow components by using the movement rate, distinguishing between road surface deformation and data mutations caused by sporadic pixel noise, so that the obtained three feature values ​​can accurately reproduce the actual road conditions and ensure the reliability of subsequent passability.

[0071] In some embodiments, the first modulus value is determined based on the first corrected optical flow component, the corresponding second corrected optical flow component and the corresponding third corrected optical flow component of each group of pixels, including: taking any group of pixels as the first group of pixels, and determining the first modulus value corresponding to the first group of pixels based on the following formula (4); (4) in, The first modulus value corresponding to the first group of pixels. The first corrected optical flow component for the first group of pixels. The second corrected optical flow component for the first group of pixels. This is the third corrected optical flow component for the first group of pixels.

[0072] In one possible implementation, determining the first feature value, the second feature value, and the third feature value based on the first corrected optical flow component, the corresponding second corrected optical flow component, the corresponding third corrected optical flow component, and the corresponding first modulus value of each group of pixels includes: for the first group of pixels in the plurality of pixels, determining the ratio between the first corrected optical flow component and the corresponding first modulus value of the first group of pixels, determining the ratio between the second corrected optical flow component and the corresponding first modulus value of the first group of pixels, and determining the ratio between the third corrected optical flow component and the corresponding first modulus value of the first group of pixels, thereby obtaining the normalized horizontal optical flow component and the normalized vertical optical flow component of the first group of pixels. The system calculates the optical flow components and normalized depth optical flow components; performs divergence calculations on the normalized horizontal optical flow component, the normalized vertical optical flow component, and the normalized depth optical flow component to obtain the first feature value of the first group of pixels, thereby obtaining the first feature value of the multiple groups of pixels; performs divergence calculations on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels to determine the second feature value of each group of pixels, thereby obtaining the second feature value of the multiple groups of pixels, and determining the distribution density of the obstacle area on the driving road surface; performs Laplacian calculations on the first modulus value of each group of pixels to determine the third feature value of each group of pixels, thereby obtaining the third feature value of the multiple groups of pixels.

[0073] In the above technical solution, the optical flow component is normalized by correcting the ratio between the optical flow component and the corresponding first modulus value. This eliminates the calculation deviation caused by the amplitude difference in the movement rate of corresponding groups of pixels, and unifies the calculation benchmark of the optical flow components of each group of pixels. Furthermore, divergence calculation is performed on the normalized horizontal optical flow component, normalized vertical optical flow component, and normalized depth optical flow component to determine the first feature value, accurately capturing the horizontal shape changes of the road surface. Divergence calculation is performed on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels to determine the second feature value of each group of pixels, accurately identifying the density of obstacle areas. Laplacian calculation is performed on the first modulus value of each group of pixels to obtain the third feature value corresponding to the continuity of the road surface in the vertical direction, effectively perceiving the vertical structural changes of the road surface. The above solution adopts different determination methods for the three feature values ​​to adapt to different feature value extraction needs, avoid the feature value aliasing problem, and improve the accuracy and independence of the three feature value extraction.

[0074] In some embodiments, performing a divergence operation on the normalized horizontal optical flow component, the normalized vertical optical flow component, and the normalized depth optical flow component to obtain a first feature value of the first group of pixels includes: performing a divergence operation on the normalized horizontal optical flow component, the normalized vertical optical flow component, and the normalized depth optical flow component to obtain a fifth feature value of the first group of pixels, and determining the absolute value of the fifth feature value as the first feature value of the first group of pixels.

[0075] It should be understood that in the above scheme, each group of pixels has a corresponding optical flow vector, which represents the direction of pixel movement in the actual image. The magnitude of the optical flow vector represents the speed of movement of the corresponding group of pixels. When taking a very small image area around the pixel, the essence of divergence is the net rate of change of all optical flow vectors within the image area, either converging inward or diverging outward. The unevenness of the road surface changes the three-dimensional height of the texture on the road surface. The distance from each texture to the visual sensor is not equal, which causes the projection displacement of the pixels corresponding to each texture to deflect differently, thus causing the optical flow vectors corresponding to the pixels to converge inward or diverge outward. That is, the degree of surface curvature of the road surface can be characterized by divergence.

[0076] Furthermore, a divergence less than 0 indicates that the optical flow vectors corresponding to surrounding pixels converge towards that pixel; a divergence greater than 0 indicates that the optical flow vectors corresponding to surrounding pixels diverge outwards from that pixel. When the divergence is 0, the optical flow vectors tend to be parallel, neither converging nor diverging. Simultaneously, the larger the absolute value of the divergence, the larger the first eigenvalue, the more severe the surface curvature, and the more curved the surface; the smaller the absolute value of the divergence, the smaller the first eigenvalue, the lower the surface curvature, and the flatter the surface.

[0077] Furthermore, after obtaining the first feature values ​​of each group of pixels, the first feature values ​​of multiple groups of pixels can be aggregated to obtain a first feature value used to indicate the surface curvature of the road surface. Specifically, this can be obtained by averaging the first feature values ​​of multiple groups of pixels.

[0078] In some embodiments, performing a Laplacian operation on the first modulus value of each group of pixels to obtain the third feature value includes: performing a Laplacian operation on the first modulus value of each group of pixels to obtain a sixth feature value of the first group of pixels, and determining the absolute value of the sixth feature value as the third feature value of the first group of pixels.

[0079] It should be understood that in the above scheme, the first modulus value reflects the movement rate of the corresponding group of pixels, which can be regarded as the movement rate of each texture relative to the visual sensor in imaging. When the continuity is high (the road surface is smooth), the first modulus values ​​corresponding to different groups of pixels change gradually, and the second-order rate of change corresponding to the first modulus value (i.e., the calculation result of the Laplacian operation) is relatively small. When the continuity is low (the driving road surface changes abruptly in the height direction), the first modulus values ​​corresponding to adjacent groups of pixels change drastically, with sharp jumps, and the calculation result increases. Therefore, the continuity of the driving road surface in the height direction can be quantified based on the calculation result of the Laplacian operation on the first modulus values ​​of each group of pixels, thereby characterizing the third feature value.

[0080] Furthermore, the first modulus value changes gradually, the second-order rate of change corresponding to the first modulus value is relatively small, and the third eigenvalue is smaller; the first modulus value changes drastically, and the third eigenvalue is larger.

[0081] Furthermore, after obtaining the third feature values ​​of each group of pixels, the third feature values ​​of multiple groups of pixels can be aggregated to obtain a third feature value used to indicate the continuity of the driving surface in the height direction. Specifically, this can be obtained by averaging the third feature values ​​of multiple groups of pixels.

[0082] In one possible implementation, a divergence operation is performed on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels to determine the second feature value of each group of pixels. This includes: performing a divergence operation on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels to obtain multiple divergence values, and determining the maximum divergence value from the multiple divergence values. The multiple divergence values ​​are used to indicate the degree of aggregation of multiple groups of pixels when the position is offset; for the first group of pixels in the multiple groups of pixels, the ratio between the divergence value of the first group of pixels and the maximum divergence value is determined to obtain the normalized divergence value of the first group of pixels; the difference between the first preset value and the normalized divergence value of the first group of pixels is determined as the second feature value of the first group of pixels to obtain the second feature of each group of pixels.

[0083] In the above technical solution, divergence calculations are performed on the three types of corrected optical flow components of each group of pixels to characterize the degree of aggregation of multiple groups of pixels when their positions are offset. Furthermore, the divergence values ​​of each group of pixels are normalized based on the maximum divergence value, which eliminates computational imbalances caused by differences in the magnitude of divergence values ​​under different road surface scenarios. Subsequently, the second feature value of each group of pixels is determined based on the difference between the first preset value and the normalized divergence value. This allows the density of obstacle areas to be mapped based on the physical meaning of the divergence values, avoiding the drawbacks of feature distortion caused by changes in acquisition distance, and accurately quantifying the distribution density of obstacle areas, so that the second feature value can objectively reflect the arrangement of obstacle areas.

[0084] It should be understood that in the above scheme, if the obstacle area on the driving road is dense, the obstacle area (which can be an obstacle) will obscure the surrounding road texture. During imaging, most of the surrounding road texture converges towards the center of the obstacle, resulting in a negative normalized divergence value. If the obstacles on the driving road are sparse (corresponding to a flat, open, or normal driving area far from obstacles), the corresponding normalized divergence value is positive. If the focus is on the convergence density of obstacles, determining the difference between the first preset value and the normalized divergence value of the first group of pixels (the normalized divergence value of each group of pixels) allows for a greater degree of convergence, resulting in a larger absolute value of the corresponding normalized divergence value and a second feature value closer to 1; conversely, a greater degree of divergence results in a larger normalized divergence value and a second feature value closer to 0.

[0085] Furthermore, it should be noted that both the first and second eigenvalues ​​mentioned above involve the convergence or divergence of optical flow vectors. However, their corresponding generation causes differ. The convergence and divergence of the first eigenvalue are caused by the surface unevenness of the road surface. During calculation, the absolute value of the divergence is extracted, and the degree of surface curvature is quantified only based on the strength of the convergence and divergence. The convergence and divergence of the second eigenvalue originate from the aggregation phenomenon of obstacle areas on the road surface. During calculation, the positive and negative values ​​of the divergence are retained, and the positive or negative value of the divergence is used to distinguish between convergence and divergence attributes, ultimately determining the second eigenvalue to characterize the distribution density of obstacle areas.

[0086] Furthermore, after obtaining the second feature values ​​of each group of pixels, the second feature values ​​of multiple groups of pixels can be aggregated to obtain a second feature value used to indicate the distribution density of obstacle areas on the driving road surface. Specifically, the second feature values ​​of each group of pixels can be compared with preset feature values, and the average of at least one second feature value greater than the preset feature value can be calculated to obtain the second feature value.

[0087] Step 202: Based on the first feature value, the second feature value, and the third feature value, determine the passability score, which is used to indicate the passability of the driving surface.

[0088] It should be understood that in step 202 above, the passability score is a quantitative score calculated by combining three road condition indicators: the surface curvature of the driving surface, the distribution density of obstacle areas, and the continuity of the driving surface in the vertical direction. This score represents the passability of the current driving surface. The higher the passability score, the greater the passability of the driving surface, and the easier it is to pass through the driving surface.

[0089] In one possible implementation, step 202, determining the passability score based on the first feature value, the second feature value, and the third feature value, includes: determining a first score, a second score, and a third score based on the first feature value, the second feature value, and the third feature value, wherein the first score indicates the passability affected by the degree of surface curvature, the second score indicates the passability affected by the distribution density of obstacle areas, and the third score indicates the passability affected by the degree of continuity in the height direction; determining a first weight, a second weight, and a third weight based on the vehicle's current driving parameters, wherein the first weight indicates the degree of trust in the first feature value when determining the passability score, the second weight indicates the degree of trust in the second feature value when determining the passability score, and the third weight indicates the degree of trust in the third feature value when determining the passability score; and weighting and fusing the first score, the second score, and the third score based on the first weight, the second weight, and the third weight to obtain the passability score.

[0090] It should be understood that in the above scheme, the first score can be understood as the quantitative score converted after considering the road condition index of the surface curvature of the driving road surface. The greater the surface curvature, the more severe the surface curvature, the larger the first characteristic value, and the lower the first score. The less the surface curvature, the flatter the driving road surface, the smaller the first characteristic value, and the higher the first score. That is, the first score is negatively correlated with the surface curvature. The second score can be understood as the quantitative score converted after considering the road condition index of the distribution density of obstacle areas. The greater the distribution density, the larger the second characteristic value, and the lower the second score. The less the distribution density, the smaller the second characteristic value, and the higher the second score. That is, the second score is negatively correlated with the distribution density. The third score can be understood as the quantitative score converted after considering the road condition index of the continuity of the driving road surface in the vertical direction. The greater the continuity, the lower the third characteristic value, and the higher the second score. The less the continuity, the higher the third characteristic value, and the lower the second score.

[0091] In the above technical solution, the three feature values ​​are converted into individual scores (first score, second score, and third score). This separates the impact of three road conditions on trafficability, avoiding interference between different road condition parameters in the trafficability assessment. Subsequently, based on the vehicle's current driving parameters, the three weights are dynamically matched. This adjusts the trust weight of each road condition based on real-time vehicle conditions, addressing the evaluation bias caused by fixed weights failing to adapt to changes in vehicle speed. Furthermore, the individual scores are weighted and fused based on the three weights to obtain a trafficability score. This integrates multi-dimensional road conditions to form a unified standard for trafficability assessment, optimizing the shortcomings of single-indicator judgments and making the trafficability assessment results more closely reflect the actual driving conditions of the vehicle.

[0092] In one possible implementation, the road surface is divided into multiple road surface regions. The second feature value includes a first distribution feature value and a second distribution feature value. Based on the first feature value, the second feature value, and the third feature value, a first score, a second score, and a third score are determined, including: determining the first score based on the feature difference between a first preset feature value and the first feature value; determining the ratio between the first distribution feature value and the second distribution feature value as a first value, where the first value indicates the uniformity of the distribution of obstacle regions on the road surface, the first distribution feature value indicates the average distribution density of obstacle regions in the multiple road surface regions, and the second distribution feature value indicates the maximum distribution density of obstacle regions in the multiple road surface regions; determining the second score based on the difference between the first preset value and the first value, where the first preset value indicates that the distribution density of obstacle regions in the multiple road surface regions after the road surface is divided is consistent; and performing an exponential transformation on the opposite feature value of the third feature value to obtain the third score.

[0093] It should be understood that in the above scheme, the first preset feature value is specifically used to indicate that the surface of the driving road is free of curvature and the driving road is relatively flat. The first preset feature value can be 1. When the distribution density of the obstacle area is consistent in each road area after the driving road is divided, the maximum distribution density of the obstacle area in multiple road areas is the same as the average distribution density, that is, the first preset value is 1.

[0094] In the above technical solution, a first score is determined based on the feature difference between a first preset feature value and a first feature value. This allows for the use of a smooth road surface as a reference standard to quantify the traffic loss caused by surface curvature. The ratio between the average and maximum distribution density of the obstacle area is used to characterize the uniformity of the obstacle area's distribution. Then, a second score is determined based on the difference between a first preset value used to characterize uniform distribution density and a first value. This takes into account both the overall density of obstacle areas on the road surface and the impact of local obstacle areas. Simultaneously, an exponential transformation is performed on the opposite feature value of the third feature value to determine the third score. This uses an exponential function to adjust for abnormal scores caused by extreme fluctuations. These three differentiated scoring methods are adapted to the three types of road conditions, avoiding the drawbacks of a unified scoring formula that cannot match the characteristics of different road condition indicators, and improving the accuracy of the three scores (three individual scores).

[0095] In some embodiments, determining the first score based on the feature difference between a first preset feature value and the first feature value includes: determining the first score as the product of the feature difference and a preset benchmark score.

[0096] In some embodiments, determining the first score based on the feature difference between the first preset feature value and the first feature value includes: determining the first score based on the following formula (5); (5) in, For this first score, The first preset feature value, It is the first eigenvalue.

[0097] In some embodiments, determining the second score based on the difference between a first preset value and the first value includes: determining the second score as the product of the difference and the benchmark score.

[0098] In some embodiments, determining the second score based on the difference between a first preset value and the first value includes: determining the second score based on the following formula (6); (6) in, For this second score, This is the first preset value. This is the characteristic value of the first distribution. This is the characteristic value of the second distribution. This is the first value.

[0099] In some embodiments, performing an exponential transformation on the opposite eigenvalue of the third eigenvalue to obtain the third score includes: performing an exponential transformation on the opposite eigenvalue of the third eigenvalue to obtain a fourth eigenvalue; and determining the product between the fourth eigenvalue and the benchmark score as the third score.

[0100] In some embodiments, the third score is obtained by performing an exponential transformation on the opposite eigenvalue of the third eigenvalue, including: determining the third score based on the following formula (7); (7) in, For this third score, This is the third eigenvalue. It is the opposite eigenvalue of the third eigenvalue.

[0101] In the above formulas (5) to (7), 100 is the preset benchmark score, which is used to convert different feature values ​​(feature differences) or numerical values ​​into scores.

[0102] In one possible implementation, the vehicle's current driving parameter is its speed. Based on this current driving parameter, a first weight, a second weight, and a third weight are determined, including: when the vehicle speed is less than a first preset speed, the first weight, the second weight, and the third weight are determined as a first preset weight, a second preset weight, and a third preset weight, respectively, where the second preset weight is greater than the first preset weight, and the first preset weight is greater than the third preset weight; when the vehicle speed is greater than or equal to the first preset speed but less than the second preset speed, the first weight, the second weight, and the third weight are determined as a fourth preset weight, a fifth preset weight, and a sixth preset weight, where the fourth preset weight is greater than the fifth preset weight, and the sixth preset weight is greater than the fifth preset weight; when the vehicle speed is greater than or equal to the second preset speed, the first weight, the second weight, and the third weight are determined as a seventh preset weight, an eighth preset weight, and a ninth preset weight, where the ninth preset weight is greater than the seventh preset weight, and the seventh preset weight is greater than the eighth preset weight.

[0103] It should be understood that in the above scheme, the sum of the first weight, the second weight and the third weight is 1, the sum of the first preset weight, the second preset weight and the third preset weight is 1, the sum of the fourth preset weight, the fifth preset weight and the sixth preset weight is 1, and the sum of the seventh preset weight, the eighth preset weight and the ninth preset weight is 1.

[0104] In the above technical solution, the current driving parameters are used as the speed to describe the process of determining the first, second, and third weights based on these parameters. Specifically, three speed ranges are divided, each with its corresponding weight. When the speed is less than the first preset speed (i.e., the low-speed phase), the second weight corresponding to the distribution density of obstacle areas is increased to match the driving pattern where the vehicle is more likely to notice obstacles on the road surface at low speeds. When the speed is greater than or equal to the first preset speed but less than the second preset speed (i.e., the medium-speed phase), the first weight corresponding to the surface curvature and the third weight corresponding to the continuity are increased to match the rapid changes in road conditions caused by bumps when the vehicle turns at medium speeds. When the speed is greater than or equal to the second preset speed (i.e., the high-speed phase), the third weight is emphasized, which addresses the characteristic that road undulations can easily induce vehicle instability at high speeds. This solution uses speed-based weight switching instead of fixed weights, eliminating the problem of fixed weights failing to match the real-time driving state of the vehicle, ensuring that the weighted passability score is more accurate.

[0105] For example, the first preset vehicle speed is 10 km / h and the second preset vehicle speed is 40 km / h. That is, when the vehicle speed is less than the first preset vehicle speed, the first weight is 0.34, the second weight is 0.5, and the third weight is 0.16; when the first preset vehicle speed is less than or equal to the second preset vehicle speed, the first weight is 0.36, the second weight is 0.3, and the third weight is 0.34; when the vehicle speed is greater than or equal to the second preset vehicle speed, the first weight is 0.32, the second weight is 0.28, and the third weight is 0.4.

[0106] It should be understood that the current driving parameters also include other parameters such as yaw rate, acceleration, and suspension height. The above scheme provides a specific process for determining the first weight, second weight, and third weight based on vehicle speed. The following describes the specific process for determining the first weight, second weight, and third weight based on other parameters.

[0107] In some embodiments, the current driving parameter is the yaw rate. Based on the current driving parameter of the vehicle, determining a first weight, a second weight, and a third weight includes: when the absolute value of the yaw rate is less than or equal to a first preset angular velocity, determining the first weight, the second weight, and the third weight as a tenth preset weight, an eleventh preset weight, and a twelfth preset weight, where the eleventh preset weight is greater than the tenth preset weight, and the tenth preset weight is greater than the twelfth preset weight; when the absolute value of the yaw rate is greater than the first preset angular velocity and less than the second preset angular velocity, determining the first weight, the second weight, and the third weight as a thirteenth preset weight, a fourteenth preset weight, and a fifteenth preset weight, where the thirteenth preset weight is greater than the fourteenth preset weight, and the fourteenth preset weight is greater than the fifteenth preset weight; when the absolute value of the yaw rate is greater than or equal to the second preset angular velocity, determining the first weight, the second weight, and the third weight as a sixteenth preset weight, a seventeenth preset weight, and an eighteenth preset weight, where the sixteenth preset weight is greater than the eighteenth preset weight, and the eighteenth preset weight is greater than the seventeenth preset weight.

[0108] For example, the first preset angular velocity is 3° / s and the second preset angular velocity is 8° / s. That is, when the absolute value of the yaw rate is less than or equal to the first preset angular velocity, the first weight is 0.35, the second weight is 0.45, and the third weight is 0.2; when the first preset angular velocity is less than or equal to the absolute value of the yaw rate and less than the second preset angular velocity, the first weight is 0.42, the second weight is 0.33, and the third weight is 0.25; when the absolute value of the yaw rate is greater than or equal to the second preset angular velocity, the first weight is 0.55, the second weight is 0.2, and the third weight is 0.25.

[0109] In some embodiments, the current driving parameter is acceleration. Based on the current driving parameter of the vehicle, determining a first weight, a second weight, and a third weight includes: when the absolute value of the acceleration is less than or equal to a first preset acceleration, determining the first weight, the second weight, and the third weight as a nineteenth preset weight, a twentieth preset weight, and a twenty-first preset weight, respectively, where the twentieth preset weight is greater than the nineteenth preset weight, and the nineteenth preset weight is greater than the twenty-first preset weight; when the absolute value of the acceleration is greater than the first preset acceleration and less than the second preset acceleration, determining the first weight, the second weight, and the third weight as a twenty-second preset weight, a twenty-third preset weight, and a twenty-fourth preset weight, where the twenty-third preset weight is greater than the twenty-second preset weight, and the twenty-second preset weight is greater than the twenty-fourth preset weight; when the absolute value of the acceleration is greater than or equal to the second preset acceleration, determining the first weight, the second weight, and the third weight as a twenty-fifth preset weight, a twenty-sixth preset weight, and a twenty-seventh preset weight, where the twenty-seventh preset weight is greater than the twenty-sixth preset weight, and the twenty-sixth preset weight is greater than the twenty-fifth preset weight.

[0110] For example, the first preset acceleration is 0.5 m / s². 2 The second preset acceleration is 1.5 m / s². 2 Specifically, when the absolute value of acceleration is less than or equal to the first preset acceleration, the first weight is 0.35, the second weight is 0.45, and the third weight is 0.2; when the first preset acceleration is less than the absolute value of acceleration and less than the second preset acceleration, the first weight is 0.32, the second weight is 0.38, and the third weight is 0.3; when the absolute value of acceleration is greater than or equal to the second preset acceleration, the first weight is 0.25, the second weight is 0.3, and the third weight is 0.45.

[0111] Step 203: If the passability score is greater than the first preset score, control the vehicle to pass through the road surface with the current driving parameters.

[0112] It should be understood that in step 203 above, the first preset score refers to the critical score for passing through the road surface, specifically the minimum score for the vehicle to safely (road conditions meet the standards, with no risk of passage) pass through the road surface. Optionally, this first preset score is 85 points. Furthermore, "the vehicle passes through the road surface with the current driving parameters" means that the vehicle maintains its current speed, accelerator pedal, suspension height, and other driving parameters unchanged, without decelerating, accelerating, or adjusting its body posture, and normally passes through the road surface according to the original operating conditions.

[0113] It should also be understood that the pass score may be greater than the first preset score, and similarly, the pass score may be less than or equal to the first preset score. These two cases are further described below.

[0114] Scenario 1 :like Figure 3 As shown, the pass score is less than or equal to the first preset score and greater than the second preset score. Step 301: If the passability score is less than or equal to the first preset score and greater than the second preset score, determine the risk type of the road surface based on the first feature value, the second feature value, and the third feature value. The risk type indicates the cause of the vehicle's decreased passability. Step 302: Based on the risk type, adjust the current driving parameters to obtain adjusted driving parameters, and control the vehicle to pass through the road surface with the adjusted driving parameters.

[0115] It should be understood that in the above scheme, the risk types can be divided into three types, corresponding to the first risk type, the second risk type and the third risk type. The first risk type can be regarded as the risk type where the surface curvature exceeds the standard, the second risk type can be regarded as the risk type where the distribution density of the obstacle area exceeds the standard, and the third risk type can be regarded as the risk type where the continuity of the driving road surface in the height direction does not meet the standard.

[0116] It should be noted here that... Figure 3 The scheme shown corresponds to weight 10.

[0117] In the aforementioned technical solution, when the passability score is less than or equal to a first preset score but greater than a second preset score (passability score within the middle range), the risk type of the driving road surface is determined based on the first, second, and third characteristic values ​​to pinpoint the cause of the vehicle's decreased passability. That is, based on each characteristic value, the specific road condition source of the deteriorating passability is identified, abandoning the approach of generalized control. Furthermore, based on the risk type, the current driving parameters are specifically adjusted to address the shortcoming that, when the passability score is low, the original driving parameters (current driving parameters) can only directly allow passage or abruptly prohibit passage. In other words, this solution can achieve differentiated vehicle control by identifying the specific road condition source of the deteriorating passability, avoiding the safety hazards of tire slippage or chassis scraping, while also avoiding blindly and drastically reducing speed, which would affect traffic efficiency.

[0118] Optionally, the second preset score is 60 points.

[0119] In some embodiments, determining the risk type of the driving surface based on the first feature value, the second feature value, and the third feature value includes: determining a first score, a second score, and a third score based on the first feature value, the second feature value, and the third feature value; if the first score is less than a third preset score, determining the risk type includes a first risk type, which indicates that the cause of the reduced vehicle passability is that the surface curvature is greater than a preset curvature; if the second score is less than the third preset score, determining the risk type includes a second risk type, which indicates that the cause of the reduced vehicle passability is that the distribution density of the obstacle area is greater than a preset distribution density; if the third score is less than the third preset score, determining the risk type includes a third risk type, which indicates that the cause of the reduced vehicle passability is that the continuity of the driving surface in the height direction is less than a preset continuity.

[0120] That is, the aforementioned risk types include at least one of the first risk type, the second risk type, and the third risk type. Optionally, the third pre-set score is 60 points.

[0121] In some embodiments, adjusting the current driving parameters based on the risk type to obtain adjusted driving parameters includes: when the risk type includes a first risk type, determining a first deviation score between a third preset score and a first score; determining the ratio between the first deviation score and the third preset score to obtain a first reduction magnitude; reducing the vehicle speed in the current driving parameters based on the first reduction magnitude, and reducing the response sensitivity of the power steering torque when the steering wheel is turned in the current driving parameters based on the first reduction magnitude; when the risk type includes a second risk type, determining a second deviation score between the third preset score and the second score; and determining the second deviation score. The ratio between the second increase and the third preset score is used to obtain the second increase magnitude; based on the second increase magnitude, the brake preload oil pressure in the current driving parameter is increased at a first time, the first time being related to the second increase magnitude, and the degree of increase in the brake preload oil pressure is the second increase magnitude, the brake preload oil pressure being the preset oil pressure inside the brake wheel cylinder in the vehicle used to eliminate the gap between the brake pads and the brake disc; if the risk type includes a third risk type, a third deviation score is determined between the third preset score and the third score; the ratio between the third deviation score and the third preset score is determined to obtain the third increase magnitude; based on the third increase magnitude, the damping force of the shock absorber in the current driving parameter is increased.

[0122] It should be understood that in the above scheme, the lower the first score, the greater the reduction in the first degree of speed reduction, and the greater the reduction in the response sensitivity of the power steering torque. Specifically, the response sensitivity of the power steering torque when the steering wheel is turned can be considered as the speed at which the power steering system outputs power steering torque in response to the driver's steering wheel turn. For the same steering wheel angle, higher response sensitivity results in a faster increase in power steering torque, making steering easier; conversely, lower response sensitivity slows the rate of increase in power steering torque, requiring the driver to apply greater grip force to change direction.

[0123] In some embodiments, the method for determining the first time includes: determining the actual distance between the vehicle and the obstacle area; determining the time taken for the vehicle to reach the obstacle area based on the actual distance and the vehicle's speed; determining the first time by multiplying the second increase by the time taken; determining the second time by the difference between the time taken and the first time; and obtaining the first time by adding the second time to the current time as the starting point.

[0124] Scenario 2 :like Figure 4 As shown, the pass score is less than or equal to the second preset score. Step 401: If the passability score is less than or equal to the second preset score, control the vehicle to output a first reminder message. This first reminder message is used to remind the driver that the passability of the road ahead is lower than the preset passability and that passage is prohibited. Step 402: Control the vehicle to decelerate and stop in a safe road area. Step 403: Mark the road segment corresponding to the passability as a prohibited section, and store the location information and risk association information of the section. This will assist other vehicles in outputting a second reminder message when the distance between them and the section is less than a preset distance after a preset time period. This second reminder message is used to remind the driver that the section is a prohibited section and suggest an alternative route.

[0125] It should be understood that in the above scheme, the location information of the road segment is used to locate the spatial range of the corresponding road segment, and may include the name, start location, and end location of the corresponding road segment. Risk-related information is used to characterize the supporting working conditions that prevent normal passage of the corresponding road segment, including a first characteristic value, a second characteristic value, a third characteristic value, a first score, a second score, a third score, a risk type, a passability score, and a second time for determining the passability score. Furthermore, the preset duration is specifically the duration after the second time, and this preset duration is relatively short.

[0126] It should be noted here that... Figure 4 The scheme shown corresponds to weight 11.

[0127] Optionally, the preset duration is 7 days. The preset duration is the effective duration of the calculated risk-related information starting from the second time. It is a critical time limit determined based on the natural changes in road conditions such as road surface damage. Within this critical time limit, the road conditions of the corresponding road segment remain unchanged by default. After the critical time limit is exceeded, the risk-related information becomes invalid and is no longer used for warnings of other vehicles.

[0128] In the aforementioned technical solution, when the passability score is less than or equal to the second preset score (extremely low passability score), a multi-level safety response mechanism is triggered. This involves outputting a warning message, promptly informing the driver that the current road surface's capacity is insufficient, preventing the driver from attempting to pass and causing safety hazards. Simultaneously, the vehicle is slowed down and brought to a safe road surface area to completely eliminate safety accidents such as getting stuck or losing control from a driving control perspective. Furthermore, the corresponding road segment is marked as a no-entry segment, and the location information and risk association information of the segment are stored. This allows for detour reminders to nearby vehicles, achieving a comprehensive shared early warning of risks for a single road segment. This overcomes the limitations of individual vehicles' autonomous risk avoidance, effectively preventing repeated occurrences of similar road conditions and significantly improving the safety and intelligent early warning capabilities of road sections.

[0129] Figure 5 This is a schematic diagram of a device for controlling vehicle movement provided in an embodiment of this application.

[0130] For example, such as Figure 5 As shown, the device 500 includes: Determine module 501, used for: While the vehicle is in motion, based on the road surface image, spatial observation data and the vehicle's driving parameters, a first feature value, a second feature value and a third feature value are determined. The first feature value is used to indicate the surface curvature of the road surface, the second feature value is used to indicate the distribution density of obstacle areas on the road surface, and the third feature value is used to indicate the continuity of the road surface in the height direction. Based on the first feature value, the second feature value, and the third feature value, a passability score is determined, which is used to indicate the degree of passability of the driving surface; The control module 502 is used to control the vehicle to pass through the road surface with the current driving parameters when the passability score is greater than the first preset score.

[0131] Optionally, the road surface image consists of two adjacent frames, including a first frame and a second frame. The spatial observation data consists of the spatial position parameters of each measured point on the road surface relative to the vehicle-mounted radar. The determining module 501 is specifically used for: determining the depth value corresponding to each pixel in the two frames based on the spatial observation data; for each first pixel in the first frame, determining the second pixel corresponding to the imaging position of the first pixel in the second frame, obtaining multiple sets of pixels, each set of pixels including a first pixel and a corresponding second pixel; and determining the first pixel in the horizontal direction of each set of pixels based on the lateral offset value, vertical offset value, and depth value corresponding to each set of pixels in the two frames. The optical flow component, the second optical flow component in the vertical direction, and the third optical flow component in the depth direction are used to obtain each first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component in a plurality of first optical flow components. The optical flow component is the position offset of each group of pixels. Based on the driving parameter, each first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component are corrected to obtain each first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component. Based on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels, the first feature value, the second feature value, and the third feature value are determined.

[0132] Optionally, the driving parameter includes acceleration. The determining module 501 is further configured to: when the acceleration is a preset acceleration, determine each first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component as each first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component; when the acceleration is greater than the preset acceleration, correct each first optical flow component based on a first reduction magnitude to obtain the corresponding first corrected optical flow component; correct each second optical flow component based on a second reduction magnitude to obtain the corresponding second corrected optical flow component; and correct each first optical flow component based on a third reduction magnitude. The three optical flow components are corrected to obtain a corresponding third corrected optical flow component. The first reduction amplitude, the second reduction amplitude, and the third reduction amplitude are positively correlated with the acceleration. When the acceleration is less than the preset acceleration, each first optical flow component is corrected based on the first compensation amplitude to obtain a corresponding first corrected optical flow component. Each second optical flow component is corrected based on the second compensation amplitude to obtain a corresponding second corrected optical flow component. Each third optical flow component is corrected based on the third compensation amplitude to obtain a corresponding third corrected optical flow component. The first compensation amplitude, the second compensation amplitude, and the third compensation amplitude are negatively correlated with the acceleration.

[0133] Optionally, the determining module 501 is further configured to: determine a first modulus value based on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels, thereby obtaining a plurality of first modulus values, wherein the first modulus values ​​are used to indicate the moving rate of the corresponding group of pixels; and determine the first feature value, the second feature value, and the third feature value based on the first corrected optical flow component, the corresponding second corrected optical flow component, the corresponding third corrected optical flow component, and the corresponding first modulus value of each group of pixels.

[0134] Optionally, the determining module 501 is further configured to: for a first group of pixels in the plurality of pixels, determine the ratio between the first corrected optical flow component and the corresponding first modulus value of the first group of pixels, determine the ratio between the second corrected optical flow component and the corresponding first modulus value of the first group of pixels, and determine the ratio between the third corrected optical flow component and the corresponding first modulus value of the first group of pixels, thereby obtaining the normalized horizontal optical flow component, the normalized vertical optical flow component, and the normalized depth optical flow component of the first group of pixels; and for the normalized horizontal optical flow component and the normalized vertical optical flow component... The first feature value of the first group of pixels is obtained by performing a divergence operation on the component and the normalized depth optical flow component, so as to obtain the first feature value of the multiple groups of pixels; the second feature value of the first corrected optical flow component, the corresponding second corrected optical flow component and the corresponding third corrected optical flow component of each group of pixels are diverged to determine the second feature value of each group of pixels, so as to obtain the second feature value of the multiple groups of pixels, and to determine the distribution density of the obstacle area on the driving road surface; the third feature value of the first modulus value of each group of pixels is determined by performing a Laplacian operation on the first modulus value of each group of pixels, so as to obtain the third feature value of the multiple groups of pixels.

[0135] Optionally, the determining module 501 is further configured to: perform divergence calculation on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels to obtain multiple divergence values, and determine the maximum divergence value from the multiple divergence values, wherein the multiple divergence values ​​are used to indicate the degree of aggregation of multiple groups of pixels when the position is offset; for the first group of pixels in the multiple groups of pixels, determine the ratio between the divergence value of the first group of pixels and the maximum divergence value to obtain the normalized divergence value of the first group of pixels; and determine the difference between the first preset value and the normalized divergence value of the first group of pixels as the second feature value of the first group of pixels to obtain the second feature value of each group of pixels.

[0136] Optionally, the determining module 501 is further configured to: determine a first score, a second score, and a third score based on the first feature value, the second feature value, and the third feature value, wherein the first score indicates the passability affected by the degree of surface curvature, the second score indicates the passability affected by the distribution density of the obstacle area, and the third score indicates the passability affected by the degree of continuity in the height direction; determine a first weight, a second weight, and a third weight based on the vehicle's current driving parameters, wherein the first weight indicates the degree of trust in the first feature value when determining the passability score, the second weight indicates the degree of trust in the second feature value when determining the passability score, and the third weight indicates the degree of trust in the third feature value when determining the passability score; and perform weighted fusion of the first score, the second score, and the third score based on the first weight, the second weight, and the third weight to obtain the passability score.

[0137] Optionally, the road surface is divided into multiple road surface regions. The second feature value includes a first distribution feature value and a second distribution feature value. The determining module 501 is further configured to: determine the first score based on the feature difference between the first preset feature value and the first feature value; determine the ratio between the first distribution feature value and the second distribution feature value as a first value, the first value indicating the uniformity of the distribution of obstacle regions on the road surface, the first distribution feature value indicating the average distribution density of obstacle regions in the multiple road surface regions, and the second distribution feature value indicating the maximum distribution density of obstacle regions in the multiple road surface regions; determine the second score based on the difference between the first preset value and the first value, the first preset value indicating that the distribution density of obstacle regions in the multiple road surface regions after the road surface is divided is consistent; and perform an exponential transformation on the opposite feature value of the third feature value to obtain the third score.

[0138] Optionally, the current driving parameter of the vehicle is the vehicle speed. The determining module 501 is further configured to: when the vehicle speed is less than a first preset vehicle speed, determine the first weight, the second weight, and the third weight as a first preset weight, a second preset weight, and a third preset weight, respectively, wherein the second preset weight is greater than the first preset weight, and the first preset weight is greater than the third preset weight; when the vehicle speed is greater than or equal to the first preset vehicle speed and less than the second preset vehicle speed, determine the first weight, the second weight, and the third weight as a fourth preset weight, a fifth preset weight, and a sixth preset weight, wherein the fourth preset weight is greater than the fifth preset weight, and the sixth preset weight is greater than the fifth preset weight; when the vehicle speed is greater than or equal to the second preset vehicle speed, determine the first weight, the second weight, and the third weight as a seventh preset weight, an eighth preset weight, and a ninth preset weight, wherein the ninth preset weight is greater than the seventh preset weight, and the seventh preset weight is greater than the eighth preset weight.

[0139] Optionally, the determining module 501 is further configured to determine the risk type of the driving surface based on the first feature value, the second feature value, and the third feature value when the passability score is less than or equal to the first preset score and greater than the second preset score. The risk type is used to indicate the cause of the decrease in the vehicle's passability. The control module is further configured to adjust the current driving parameters based on the risk type to obtain the adjusted driving parameters, and control the vehicle to pass through the driving surface with the adjusted driving parameters.

[0140] Optionally, the control module 502 is further configured to: when the passability score is less than or equal to the second preset score, control the vehicle to output a first reminder message, the first reminder message being used to remind that the passability of the road ahead is lower than the preset passability and that passage is prohibited; and control the vehicle to decelerate and stop in a safe road area; and the device 500 further includes: an output module, configured to mark the road segment corresponding to the road surface as a prohibited road segment, and store the location information and risk association information of the road segment to assist other vehicles in outputting a second reminder message when the distance between them and the road segment is less than a preset distance after a preset time period, the second reminder message being used to remind that the road segment is a prohibited road segment and to suggest detouring.

[0141] Figure 6 This is a schematic diagram of the structure of a controller provided in an embodiment of this application.

[0142] For example, such as Figure 6 As shown, the controller 600 includes a storage module 601 and a processing module 602. The storage module 601 stores executable program code 603, and the processing module 602 is used to call and execute the executable program code 603 to perform a method for controlling the driving of a vehicle.

[0143] Figure 7 This is a schematic diagram of the structure of a vehicle provided in an embodiment of this application.

[0144] For example, such as Figure 7 As shown, the vehicle 700 includes a memory 701 and a processor 702. The memory 701 stores executable program code 703, and the processor 702 is used to call and execute the executable program code 703 to perform a method for controlling the vehicle's movement.

[0145] Furthermore, embodiments of this application also protect an apparatus that may include a memory and a processor, wherein the memory stores executable program code, and the processor is used to call and execute the executable program code to perform a method for controlling vehicle driving provided in embodiments of this application.

[0146] This embodiment can divide the device into functional modules based on the above method example. For example, each module can correspond to a separate function, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0147] When each functional module is divided according to its corresponding function, the device may also include a determining module, a controlling module, and an output module. It should be noted that all relevant content in the above method embodiments can be referenced from the functional descriptions of the corresponding functional modules, and will not be repeated here.

[0148] It should be understood that the device provided in this embodiment is used to execute the above-described method for controlling vehicle movement, and therefore can achieve the same effect as the above-described implementation method.

[0149] When using an integrated unit, the device may include a processing module and a storage module. When the device is applied to a vehicle, the processing module can be used to control and manage the vehicle's movements. The storage module can be used to support the vehicle in executing relevant executable program code.

[0150] The processing module may be a processor or a controller, which can implement or execute various exemplary logic blocks, modules, and circuits shown in conjunction with the disclosure of this application. The processor may also be a combination of functions that implement computing capabilities, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc., and the storage module may be a memory.

[0151] In addition, the device provided in the embodiments of this application may specifically be a chip, component or module. The chip may include a connected processor and a memory. The memory is used to store instructions. When the processor calls and executes the instructions, the chip can execute a method for controlling vehicle driving provided in the above embodiments.

[0152] This embodiment also provides a computer-readable storage medium storing executable program code. When the executable program code is run on a computer, the computer performs the aforementioned method steps to implement the method for controlling vehicle driving provided in the above embodiment.

[0153] This embodiment also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned steps to implement a method for controlling vehicle movement provided in the above embodiment.

[0154] In this embodiment, the device, computer-readable storage medium, computer program product, or chip are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0155] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0156] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0157] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for controlling vehicle movement, characterized in that, The method includes: While the vehicle is in motion, based on the road surface image, spatial observation data, and the vehicle's driving parameters, a first feature value, a second feature value, and a third feature value are determined. The first feature value is used to indicate the surface curvature of the road surface, the second feature value is used to indicate the distribution density of obstacle areas on the road surface, and the third feature value is used to indicate the continuity of the road surface in the height direction. Based on the first feature value, the second feature value, and the third feature value, a passability score is determined, which is used to indicate the passability of the driving surface. If the passability score is greater than the first preset score, the vehicle is controlled to pass through the road surface with the current driving parameters.

2. The method according to claim 1, characterized in that, The road surface image consists of two adjacent frames, including a first frame and a second frame. The spatial observation data comprises the spatial position parameters of each measured point on the road surface relative to the vehicle-mounted radar. Determining the first, second, and third feature values ​​based on the road surface image, spatial observation data, and the vehicle's driving parameters includes: Based on the spatial observation data, the depth value corresponding to each pixel in the two frames of images is determined; For each first pixel in the first frame image, determine the second pixel at the corresponding imaging position of the first pixel in the second frame image to obtain multiple sets of pixels, each set of pixels including the first pixel and the corresponding second pixel; Based on the horizontal offset value, vertical offset value and depth value corresponding to each group of pixels in the two frames of images, the first optical flow component in the horizontal direction, the second optical flow component in the vertical direction and the third optical flow component in the depth direction of each group of pixels are determined, and each first optical flow component, the corresponding second optical flow component and the corresponding third optical flow component in the multiple first optical flow components are obtained. The optical flow component is the position offset of each group of pixels. Based on the driving parameters, each first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component are corrected to obtain each first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component. Based on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels, the first feature value, the second feature value, and the third feature value are determined.

3. The method according to claim 2, characterized in that, The driving parameters include acceleration. Based on the driving parameters, each first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component are corrected to obtain each first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component, including: When the acceleration is a preset acceleration, each first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component are determined as each first optical flow component, the corresponding second optical flow component, and the corresponding third optical flow component. When the acceleration is greater than the preset acceleration, each first optical flow component is corrected based on the first reduction magnitude to obtain the corresponding first corrected optical flow component; each second optical flow component is corrected based on the second reduction magnitude to obtain the corresponding second corrected optical flow component; and each third optical flow component is corrected based on the third reduction magnitude to obtain the corresponding third corrected optical flow component. The first reduction magnitude, the second reduction magnitude, and the second reduction magnitude are positively correlated with the acceleration. When the acceleration is less than the preset acceleration, each first optical flow component is corrected based on the first compensation amplitude to obtain the corresponding first corrected optical flow component. Each second optical flow component is corrected based on the second compensation amplitude to obtain the corresponding second corrected optical flow component. Each third optical flow component is corrected based on the third compensation amplitude to obtain the corresponding third corrected optical flow component. The first compensation amplitude, the second compensation amplitude, and the third compensation amplitude are negatively correlated with the acceleration.

4. The method according to claim 2, characterized in that, The determination of the first feature value, the second feature value, and the third feature value based on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels includes: Based on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels, a first modulus value is determined, resulting in multiple first modulus values. The first modulus value is used to indicate the movement rate of the corresponding group of pixels. Based on the first corrected optical flow component, the corresponding second corrected optical flow component, the corresponding third corrected optical flow component, and the corresponding first modulus value of each group of pixels, the first feature value, the second feature value, and the third feature value are determined.

5. The method according to claim 4, characterized in that, The determination of the first feature value, the second feature value, and the third feature value based on the first corrected optical flow component, the corresponding second corrected optical flow component, the corresponding third corrected optical flow component, and the corresponding first modulus value of each group of pixels includes: For the first group of pixels in the plurality of pixels, the ratio between the first corrected optical flow component and the corresponding first modulus value of the first group of pixels is determined, the ratio between the second corrected optical flow component and the corresponding first modulus value of the first group of pixels is determined, and the ratio between the third corrected optical flow component and the corresponding first modulus value of the first group of pixels is determined, so as to obtain the normalized horizontal optical flow component, the normalized vertical optical flow component and the normalized depth optical flow component of the first group of pixels. Divergence calculations are performed on the normalized horizontal optical flow component, the normalized vertical optical flow component, and the normalized depth optical flow component to obtain the first feature value of the first group of pixels, thereby obtaining the first feature value of the multiple groups of pixels. Divergence calculations are performed on the first corrected optical flow component, the corresponding second corrected optical flow component, and the corresponding third corrected optical flow component of each group of pixels to determine the second feature value of each group of pixels, so as to obtain the second feature value of the multiple groups of pixels and determine the distribution density of the obstacle area on the driving road surface. A Laplacian operation is performed on the first modulus value of each group of pixels to determine the third feature value of each group of pixels, so as to obtain the third feature value of the multiple groups of pixels.

6. The method according to claim 1, characterized in that, The process of determining the passability score based on the first feature value, the second feature value, and the third feature value includes: Based on the first feature value, the second feature value, and the third feature value, a first score, a second score, and a third score are determined. The first score is used to indicate the degree of passage affected by the degree of surface curvature, the second score is used to indicate the degree of passage affected by the distribution density of the obstacle area, and the third score is used to indicate the degree of passage affected by the degree of continuity in the height direction. Based on the vehicle's current driving parameters, a first weight, a second weight, and a third weight are determined. The first weight is used to indicate the degree of trust in the first feature value when determining the passability score. The second weight is used to indicate the degree of trust in the second feature value when determining the passability score. The third weight is used to indicate the degree of trust in the third feature value when determining the passability score. Based on the first weight, the second weight, and the third weight, the first score, the second score, and the third score are weighted and fused to obtain the passability score.

7. The method according to claim 6, characterized in that, The driving surface is divided into multiple road surface regions. The second feature value includes a first distribution feature value and a second distribution feature value. Determining a first score, a second score, and a third score based on the first feature value, the second feature value, and the third feature value includes: The first score is determined based on the feature difference between the first preset feature value and the first feature value; The ratio between the first distribution characteristic value and the second distribution characteristic value is determined as the first value. The first value is used to indicate the uniformity of the distribution of obstacle areas on the driving road surface. The first distribution characteristic value is used to indicate the average distribution density of obstacle areas in the multiple road surface areas. The second distribution characteristic value is used to indicate the maximum distribution density of obstacle areas in the multiple road surface areas. The second score is determined based on the difference between the first preset value and the first value. The first preset value is used to indicate that the distribution density of the obstacle area is consistent in the multiple road areas after the driving road surface is divided. The third score is obtained by performing an exponential transformation on the opposite eigenvalue of the third eigenvalue.

8. The method according to claim 1, characterized in that, The method further includes: If the passability score is less than or equal to the first preset score and greater than the second preset score, the risk type of the driving surface is determined based on the first feature value, the second feature value and the third feature value. The risk type is used to indicate the cause of the decrease in vehicle passability. Based on the risk type, the current driving parameters are adjusted to obtain the adjusted driving parameters, and the vehicle is controlled to pass through the road surface with the adjusted driving parameters.

9. The method according to claim 8, characterized in that, The method further includes: If the passability score is less than or equal to the second preset score, the vehicle is controlled to output a first reminder message. The first reminder message is used to remind that the passability of the road ahead is lower than the preset passability and that passage is prohibited. And, control the vehicle to decelerate and stop in a safe road area; In addition, the road segment corresponding to the driving road surface is marked as a prohibited road segment, and the location information and risk association information of the road segment are stored to assist other vehicles in outputting a second reminder message when the distance between them and the road segment is less than a preset distance after a preset time. The second reminder message is used to remind that the road segment is a prohibited road segment and suggest detouring.

10. A vehicle, characterized in that, The vehicles include: Memory, used to store executable program code; A processor for calling and running the executable program code from the memory, causing the vehicle to perform the method as described in any one of claims 1 to 9.

Images