Vehicle cut-in tendency judgment method and device, computer equipment and storage medium

Abstract

The invention relates to the technical field of intelligent driving, and discloses a vehicle cut-in tendency judgment method and device, computer equipment and a storage medium, and the method comprises the steps: obtaining environment perception data of a reference vehicle in a normal driving state in a current lane; target detection and identification are carried out based on the environment perception data to determine local feature data of the target vehicle; wherein the target vehicle refers to traffic participants around the reference vehicle; the local feature data is used for calculating the position change degree of the target vehicle under the condition of transverse movement; and judging the cut-in tendency of the target vehicle by using the local feature data to obtain a tendency result of the target vehicle cutting into the current lane. Therefore, the cut-in motion of the target vehicle is accurately captured by acquiring the real-time environment sensing data of the reference vehicle in the current lane in the normal driving state, so that whether the target vehicle has the intention of cutting in the current lane or not is identified, and a more timely and reliable decision basis is provided for an automatic driving system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent driving, and in particular to a method and device for judging a tendency of a vehicle to cut in, a computer device, and a storage medium. BACKGROUND

[0002] Research and development of new energy vehicles and autonomous driving have been a focus in the field of intelligent driving technology. As a key technology in autonomous driving, judging the intention of a target vehicle to cut in is an important factor affecting the safety of intelligent driving.

[0003] However, in the related art, the intention of a target vehicle to cut in is often judged by using the position, speed, or angle change of the target vehicle, which has limitations in accuracy and real-time performance. Therefore, there is a need for a method for quickly responding and accurately detecting the intention of a target vehicle to cut in. SUMMARY

[0004] The present application aims to at least partially solve one of the technical problems in the related art. To this end, the present application provides a method and device for judging a tendency of a vehicle to cut in, a computer device, and a storage medium. The main technical solution adopted by the present application includes: In a first aspect, the present application provides a method for judging a tendency of a vehicle to cut in, which includes: obtaining environmental perception data of a reference vehicle in a normal driving state in a current lane; wherein the environmental perception data is used to describe the passable road condition of the reference vehicle in the current lane; performing target detection and identification based on the environmental perception data to determine local feature data of a target vehicle; wherein the target vehicle refers to a surrounding traffic participant of the reference vehicle; the local feature data is used to calculate the degree of position change of the target vehicle in the case of lateral movement; and using the local feature data to judge the tendency of the target vehicle to cut in, to obtain a tendency result of the target vehicle to cut in the current lane.

[0005] Optionally, performing target detection and identification based on the environmental perception data to determine the local feature data of the target vehicle includes: determining the target vehicle within a predetermined range of the reference vehicle; and using the environmental perception data to identify the features of the target vehicle to determine the local feature data of the target vehicle.

[0006] Optionally, using the local feature data to judge the tendency of the target vehicle to cut in, to obtain the tendency result of the target vehicle to cut in the current lane includes: determining two-dimensional coordinate data of a cut-in detection point on the target vehicle based on the local feature data; and using the two-dimensional coordinate data of the cut-in detection point to judge whether the target vehicle has a tendency to cut in the current lane, and obtaining the tendency result of the target vehicle to cut in the current lane.

[0007] Optionally, the two-dimensional coordinate data of the cut-in detection point is used to determine whether the target vehicle has a tendency to cut into the current lane, including: performing coordinate system conversion on the two-dimensional coordinate data of the cut-in detection point to determine target coordinate data of the cut-in detection point in a world coordinate system; determining a binding lane line of the target vehicle based on the target coordinate data; and determining a lateral distance variation trend between the target vehicle and the binding lane line using the target coordinate data to determine whether the target vehicle has a tendency to cut into the current lane.

[0008] Optionally, the current lane includes a left lane line and a right lane line, and the binding lane line of the target vehicle is determined by: determining first distance data between the target vehicle and the left lane line and second distance data between the target vehicle and the right lane line based on the target coordinate data; determining target distance data that uniquely satisfies a preset binding condition among the first distance data and the second distance data, and taking the lane line corresponding to the target distance data as the binding lane line of the target vehicle.

[0009] Optionally, the lateral distance variation trend between the target vehicle and the binding lane line is determined using the target coordinate data, including: for each time frame in a target time range, determining a current lateral distance between the target vehicle and the binding lane line in each time frame using the target coordinate data; and integrating the current lateral distance between the target vehicle and the binding lane line in each time frame to obtain the lateral distance variation trend between the target vehicle and the binding lane line.

[0010] Optionally, whether the target vehicle has a tendency to cut into the current lane is determined by: if the lateral distance variation trend continuously satisfies a preset threshold within a preset time range, it is determined that the target vehicle has a tendency to cut in.

[0011] In a second aspect, an embodiment of the present application provides a vehicle cut-in tendency determination device, which includes: an environment perception module configured to obtain environment perception data of a reference vehicle in a normal driving state in a current lane; wherein the environment perception data is used to describe a passable road condition of the reference vehicle in the current lane; a target detection module configured to perform target detection and identification based on the environment perception data to determine local feature data of a target vehicle; wherein the target vehicle refers to a surrounding traffic participant of the reference vehicle; the local feature data is used to calculate a position change degree of the target vehicle in a lateral motion condition; and a cut-in tendency determination module configured to determine a cut-in tendency of the target vehicle using the local feature data to obtain a tendency result of the target vehicle cutting into the current lane.

[0012] In a third aspect, the present application further provides a computer device including a memory and a processor, the memory stores a computer program, and the processor implements the steps of the method of any one of the above aspects when executing the computer program.

[0013] In a fourth aspect, the present application provides a computer readable storage medium, having stored thereon a computer program, which when executed by a processor implements the steps of any of the above methods.

[0014] In a fifth aspect, the present application provides a computer program product comprising a computer program which when executed by a processor implements the steps of any of the above methods.

[0015] In the above embodiments, by acquiring real-time environment perception data of the reference vehicle in the normal driving state in the current lane, an instant road condition data source is provided for the target vehicle cut-in tendency judgment. And by identifying the local feature data of the target vehicle such as the front wheel grounding point, the target vehicle cut-in motion is accurately captured to identify whether the target vehicle has the intention to cut into the current lane, thereby providing a more timely and reliable decision basis for the automatic driving system. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the specific embodiments or prior art of the present application, the drawings needed to be used in the specific embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0017] Figure 1a a schematic diagram of vehicle cut-in motion according to an embodiment of the present application; Figure 1b a simplified model diagram of vehicle cut-in motion according to an embodiment of the present application; Figure 1c a schematic diagram of vehicle static state according to an embodiment of the present application; Figure 1d a schematic diagram of vehicle cut-in state according to an embodiment of the present application; Figure 1e a schematic diagram of trend change of vehicle lateral position according to an embodiment of the present application; Figure 1f a schematic diagram of trend change of vehicle lateral velocity according to an embodiment of the present application; Figure 1g a schematic diagram of trend change of vehicle heading angle according to an embodiment of the present application; Figure 1h a schematic diagram of trend change of vehicle front wheel position according to an embodiment of the present application; Figure 1i a flowchart of a vehicle cut-in tendency judgment method according to an embodiment of the present application; Figure 1j a single tire according to an embodiment of the present application; Figure 1k a plurality of tires according to an embodiment of the present application; Figure 2a a flowchart for obtaining a cut-in tendency result according to an embodiment of the present application; Figure 2b a schematic diagram of vehicle cut-in according to an embodiment of the present application; Figure 2c a detection result diagram of a target vehicle on a neighboring lane according to an embodiment of the present application; Figure 2d a detection result diagram of a target vehicle from roadside static to motion according to an embodiment of the present application; Figure 3 a structural block diagram of a vehicle cut-in tendency judgment device according to another embodiment of the present application; Figure 4 an internal structure diagram of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0018] To make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0019] With the increase of the number of cars, the rate of traffic accidents is also increasing, and the deterioration of the natural environment, etc. have a great impact on people's production and life. Therefore, more and more scholars and enterprises gradually increase the research and development of new energy vehicles and autonomous driving. Among them, judging the target vehicle cut-in intention is a key technology in autonomous driving. The traditional method of judging the target vehicle cut-in intention is mainly to use the lateral position change, lateral speed change or heading angle change of the target vehicle to obtain the cut-in intention, but these traditional methods will be delayed, such as when the target front wheel has a cut-in tendency, the target lateral position, lateral speed and angle have no obvious change.

[0020] Therefore, the method of using the position, speed or angle change of the target vehicle to judge the target vehicle cut-in intention has limitations in accuracy and real-time, and a judgment method that can quickly respond and accurately detect the target vehicle cut-in intention is needed.

[0021] Specifically, please refer to Figure 1a, the black rectangular block represents the physical position of the target vehicle in normal driving, the gray rectangular block represents the physical position of the target vehicle in the cut-in action, and the yellow point represents the front wheel of the target vehicle. When the target vehicle has a cut-in action, there will be a movement from the black rectangular block position to the gray rectangular block position. Simplifying this model, assuming that the vehicle makes a circular arc motion around the rear axle, the cut-in motion shown in Figure 1a can be simplified to the model shown in Figure 1b . As shown in Figure 1b , the blue point in the figure is the shape center point of the target vehicle, and the yellow point is the tire point of the front wheel of the target vehicle. When the target vehicle makes a counterclockwise rigid circular arc motion, the arc length of the yellow point motion is longer than that of the blue point motion. Therefore, based on this, it can be known that the lateral position change of the tire point is more than that of the center point of the target vehicle. Therefore, when the intelligent driving system judges the target vehicle to cut in, it can also more quickly and accurately judge the intention of the target vehicle to cut in.

[0022] Exemplarily, taking the target vehicle from static ( Figure 1c ) to motion cut-in ( Figure 1d ) as an example, the lateral position change, lateral velocity change, orientation angle change and front wheel position change of the target vehicle are analyzed. The rear axle center of the reference vehicle (the ego vehicle) is taken as the coordinate origin, the direction of the head of the reference vehicle is taken as the positive direction of the y-axis, and the direction perpendicular to the y-axis of the reference vehicle and towards the left is taken as the positive direction of the x-axis. A world coordinate system is established, and the trend change graph of each attribute of the target vehicle is drawn based on the measurement data of the coordinate system, as shown in Figures 1e to 1h . In the figure, the abscissa is the time frame number, and the ordinate is the position (m), the velocity (m / s), the angle (degree) and the position (m) respectively. Figure 1e In the figure, the lateral position has a segmented trend, but its slope changes at about 387 frames, indicating that the change of the lateral position is not uniform. Combined with the slow change of the lateral position in the initial stage, it is easy to delay the judgment, so it is excluded as an attribute for accurate judgment of the cut-in trend. Figure 1f In the figure, the lateral velocity starts to change at about 385 frames, and the subsequent change is unstable, so it can be excluded as an attribute for accurate judgment of the cut-in trend. Figure 1g In the figure, the change of the orientation angle maintains an almost stable slope change, but in the interval from 375 frames to 380 frames, the angle changes only by 1.6 degrees, which is not large enough to determine whether it is caused by model disturbance or actual angle change, so it is not suitable for judging the target vehicle cut-in. Finally, in the figure Figure 1h , the change of the front wheel position (which can be understood as the lateral position of the target vehicle front wheel tire and the binding line) in the interval from 375 frames to 380 frames, the lateral position changes by 0.16 m, which is a relatively large change in the lateral dimension and can be monitored. The subsequent front wheel position change trend is also relatively continuous and uniform, and can be used to judge the cut-in tendency of the target vehicle.

[0023] Based on this, according to the embodiment of the present application, a vehicle cut-in tendency judgment method embodiment is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from here.

[0024] A vehicle cut-in tendency judgment method is provided in the embodiment, as shown in the figure, the method comprises the following steps: Figure 1i S110, obtaining environment perception data of the reference vehicle in a normal driving state in the current lane. S120, performing target detection and identification based on the environment perception data to determine local feature data of the target vehicle.

[0025] The reference vehicle can be an autonomous vehicle or an intelligent driving vehicle that executes the method, i.e. the ego vehicle. The current lane can be the lane in which the reference vehicle is legally driving, and it is also the lane in which the surrounding target vehicle can cut in and affect the driving safety of the reference vehicle. The environment perception data can be used to describe the passable road condition state of the reference vehicle in the current lane. Specifically, the environment perception data can be raw data or fusion processed data collected by one or more of the vehicle-mounted cameras, laser radars, and millimeter wave radars, such as image data or point cloud data, etc. For example, the visual sensor system carried by the reference vehicle can be used to capture the current lane and the surrounding area in real time, and obtain image-based environment perception data containing road markings and surrounding vehicles.

[0026] S120, based on the environment perception data, performing target detection and identification to determine the local feature data of the target vehicle.

[0027] The target vehicle can be a surrounding traffic participant of the reference vehicle, such as a vehicle driving or stationary in the adjacent lane. The local feature data can be data obtained by analyzing key position points that can represent the local geometric structure of the target vehicle and can be used to track the motion of the target vehicle, and the local feature data is coordinate data in a pixel coordinate system. For example, the key position points used to obtain the local feature data can be the shape center point of the target vehicle, or the tire contact point of the target vehicle, and preferably the front wheel contact point of the target vehicle. Specifically, the local feature data is used to calculate the degree of position change of the target vehicle in the case of lateral motion. Here, the lateral motion refers to the position movement of the target vehicle perpendicular to the driving direction of the reference vehicle (i.e. the lane direction).

[0028] Optionally, based on the environment perception data, the local feature data of the target vehicle can be determined first within a predetermined range of the reference vehicle, and then the environment perception data is used to identify the features of the target vehicle to determine the local feature data of the target vehicle.

[0029] The preset range for the reference vehicle refers to a specific spatial area in front of and to both sides of the reference vehicle. This range can be set according to sensor performance and system requirements. For example, it could be within approximately 15 meters of the reference vehicle. Specifically, after determining the preset range, a target detection algorithm identifies vehicles traveling in the same direction as the reference vehicle within this range from the environmental perception data and preliminarily filters them as target vehicles.

[0030] Specifically, using environmental perception data to identify the features of a target vehicle refers to using algorithms such as image recognition or target detection to select specific parts of the target vehicle in the environmental perception data (such as an image), and extracting the pixel coordinates of those parts in the image as local feature data. These specific parts can refer to one or more key locations on the target vehicle selected to characterize its motion state, such as the tire contact point or shape center point mentioned earlier.

[0031] For example, if the front tires of the target vehicle are used as key location points, then it is first necessary to determine whether the front tires are clearly visible and whether the detection is stable. When the target vehicle is located within a specific distance in front of the vehicle and its orientation is basically the same as the vehicle's, if only one tire of the target vehicle is clearly visible in the image and located at the bottom edge of the image (e.g., ... Figure 1j If the target vehicle is fully visible in the image and both tires are clearly visible (as shown), then the tire is determined to be a front tire. Figure 1k As shown in the image, the tire closer to the top edge of the image is identified as the front wheel. This means identifying the tire with the smaller ordinate value in the pixel coordinate system (i.e., closer to the top edge of the image) as the front wheel. Then, for the identified front wheel tire bounding box, the pixel coordinates of the midpoint of its bottom edge are taken as local feature data. Optionally, if the center point of the target vehicle's shape is used as the key location point, its pixel coordinates in the image can be directly obtained based on the midpoint of the bottom edge of the target vehicle's entire bounding box, serving as local feature data. This method quickly identifies potential risky target vehicles within a preset range and extracts their coordinates in the image by flexibly selecting the most representative local key location points. This processing provides a data foundation for subsequent steps, and different key location point selection strategies (such as front wheel points or shape center points) allow this method to adapt to different perception conditions and accuracy requirements, enhancing the system's robustness and applicability.

[0032] S130. Use local feature data to determine the target vehicle's tendency to enter the current lane, and obtain the result of the target vehicle's tendency to enter the current lane.

[0033] The result of the target vehicle's tendency to cut into the current lane can be a binary result, such as having or not having a tendency to cut in, or it can be a score or probability value that indicates the likelihood of cutting in.

[0034] Specifically, the local feature data is used to determine the tendency of the target vehicle to cut in. Due to the difference between the mapping relationship of image pixel position and the actual position in the physical world, the local feature data needs to be converted to the world coordinate system (o-xyz) through inverse perspective mapping (IPM) to obtain the real coordinate data of the key position point in the world coordinate system. Then, based on the multiple real coordinate data of the key position point in the continuous multiple frames of images, the lateral distance variation trend between the key position point and the binding lane line of the lane where the ego vehicle is located is calculated to determine whether the target vehicle has the tendency to cut into the current lane.

[0035] In the above embodiment, by obtaining the real-time environment perception data of the reference vehicle in the normal driving state in the current lane, an instant road condition data source is provided for the target vehicle cut-in tendency determination. And by recognizing the local feature data of the target vehicle such as the front wheel grounding point, the target vehicle cut-in motion is accurately captured to identify whether the target vehicle has the intention to cut into the current lane, thereby providing a more timely and reliable decision basis for the automatic driving system.

[0036] In some embodiments, please refer to the accompanying Figure 2a The local feature data is used to determine the tendency of the target vehicle to cut in, and the tendency result of the target vehicle to cut into the current lane includes: S210, determining two-dimensional coordinate data of a cut-in detection point on the target vehicle based on the local feature data.

[0037] The cut-in detection point can be a key position point on the target vehicle that can early reflect the lateral cut-in motion trend and is used to determine whether to cut into the current lane. For example, the key position point of the local feature data of the target vehicle can be selected as the cut-in detection point to more early show the cut-in action of the target vehicle. Specifically, the two-dimensional coordinate data of the cut-in detection point on the target vehicle refers to the coordinate value of the key position point in the pixel coordinate system, which is directly derived from the recognition result of the local feature data by the target detection algorithm.

[0038] S220, determining whether the target vehicle has the tendency to cut into the current lane using the two-dimensional coordinate data of the cut-in detection point, and obtaining the tendency result of the target vehicle to cut into the current lane.

[0039] Specifically, the two-dimensional coordinate data of the cut-in detection point can be first converted to determine the target coordinate data of the cut-in detection point in the world coordinate system. Then, the binding lane line of the target vehicle is determined based on the target coordinate data. The lateral distance variation trend between the target vehicle and the binding lane line is determined using the target coordinate data, and finally it is determined whether the target vehicle has the tendency to cut into the current lane.

[0040] The target coordinate data can be three-dimensional coordinate data representing the actual spatial position of the cut-in detection point in a world coordinate system.

[0041] It should be noted that, due to the difference between the mapping relationship of the pixel position and the actual position in the physical world, the position in the pixel coordinate system needs to be converted to the world coordinate system through inverse perspective mapping (IPM) to realize accurate distance measurement and trend analysis.

[0042] It should be noted that the motion of the target vehicle cutting into the current lane mainly occurs in the road surface plane (i.e., the plane formed by the x-axis and the y-axis), and the change of the z-axis (vertical road surface height) is very small and has no substantial impact on the cut-in tendency judgment, so in actual application, only the x and y two-dimensional coordinate data can be focused on for subsequent judgment, without affecting the accuracy of the lateral distance change trend analysis.

[0043] Further, after determining the target coordinate data, the bound lane line of the target vehicle can be determined based on the target coordinate data.

[0044] It should be noted that the current lane includes a left lane line and a right lane line, which together define the left and right boundaries of the reference vehicle in the current driving lane. The bound lane line can refer to the lane line that is closest to the target vehicle and needs to be crossed by the target vehicle during the cut-in process, i.e., the lane line that is closer to the target vehicle in terms of lateral distance from the cut-in detection point of the target vehicle. Specifically, the bound lane line of the target vehicle can be determined as follows: first, based on the target coordinate data, a first distance data between the target vehicle and the left lane line and a second distance data between the target vehicle and the right lane line are determined.

[0045] The first distance data can refer to the distance between the cut-in detection point on the target vehicle and the left lane line in the y-axis direction of the world coordinate system, and the second distance data can refer to the distance between the cut-in detection point and the right lane line in the y-axis direction of the world coordinate system.

[0046] For example, the curve equation of the lane line in the world coordinate system is as follows: y = a0 + a1x + a2x 2 +a3x 3 In the formula, x represents the longitudinal coordinate; y represents the lateral coordinate; a0, a1, a2, and a3 represent the fitting coefficients of the lane line, which can be calculated by data fitting according to the actual shape of the left and right boundary lines of the current lane.

[0047] For example, the front wheel tire point of the target vehicle is taken as the cut-in detection point, please refer to Figure 2b , based on the coordinates (x 轮胎 , y 轮胎), first take its longitudinal coordinate x 轮胎 , respectively, into the left lane line equation and the right lane line equation of the current lane, to obtain the point coordinates (x 轮胎 , y 左 ) on the left lane line and the point coordinates (x 轮胎 , y 右 ) on the right lane line corresponding to the position. Then, the difference between y 轮胎 and y 左 is calculated to obtain the first distance data Δy 左 , and the difference between y 轮胎 and y 右 is calculated to obtain the second distance data Δy 右 .

[0048] It should be noted that the difference here can refer to the horizontal coordinate value of the point on the lane line minus the horizontal coordinate value of the cut-in detection point, or the horizontal coordinate value of the cut-in detection point minus the horizontal coordinate value of the point on the lane line. And since the first distance data and the second distance data are obtained by subtracting the horizontal coordinate values of two points, the first distance data and the second distance data have positive and negative values. Taking the target vehicle on the right side of the reference vehicle as an example, the first distance data Δy 左 may be equal to y 左 -y 轮胎 , which is negative; the second distance data Δy 右 may be equal to y 右 -y 轮胎 , which is also negative.

[0049] Next, the target distance data that uniquely satisfies the preset binding condition can be determined in the first distance data and the second distance data, and the lane line corresponding to the target distance data is taken as the binding lane line of the target vehicle.

[0050] The preset binding condition can refer to the condition for screening distance data that meets the binding from the first distance data and the second distance data. For example, the absolute value condition can be set, that is, the distance data with smaller absolute value in the first distance data and the second distance data is taken as the target distance data. After determining the target distance data, the lane line corresponding to the target distance data is the binding lane line of the target vehicle. In this way, the most likely lane line crossed by each target vehicle can be dynamically and accurately locked, avoiding the interference of the far lane line on the subsequent cut-in trend judgment, thereby ensuring the pertinence and accuracy of the subsequent cut-in tendency judgment.

[0051] Finally, the lateral distance change trend between the target vehicle and the binding lane line is determined by using the target coordinate data, to judge whether the target vehicle has the tendency to cut into the current lane.

[0052] The lateral distance change trend can refer to a rule and direction of change of the lateral distance between the target vehicle and the bound lane line over time. Since vehicle movement is a continuous process, a distance value at a single time point can only reflect the instantaneous state and cannot reflect the movement trend. It is necessary to analyze the distance change over a continuous time sequence to accurately identify the real movement intention of the vehicle.

[0053] Specifically, first, for each time frame in the target time range, the current lateral distance between the target vehicle and the bound lane line is determined using the target coordinate data.

[0054] The target time range can be a continuous multi-frame time period extending forward or backward from the current frame, and its value needs to ensure that it covers enough dynamic data to reflect the trend and avoid interference from single-frame accidental data. For example, more than ten frames can be taken. The current lateral distance can refer to the target distance data of the current time frame.

[0055] Specifically, for each time frame, the point coordinate data of the bound lane line at the corresponding position is calculated by substituting the target coordinate data of the cut-in detection point into the curve equation of the bound lane line. Then, the difference between the two in the lateral direction is calculated, and the current lateral distance of the time frame is obtained. For each frame, such calculation is performed, and the current lateral distance between the target vehicle and the bound lane line at each time frame is obtained.

[0056] Then, the current lateral distance between the target vehicle and the bound lane line at each time frame is integrated to obtain the lateral distance change trend between the target vehicle and the bound lane line.

[0057] The lateral distance change trend can refer to the overall change rule of the distance sequence formed after arranging the current lateral distances of the continuous frames in time sequence, or the slope of the fitting curve obtained by linear fitting of the sequence. Specifically, the change slope of the current lateral distance sequence of the continuous multiple time frames is calculated or the change curve is fitted, and the lateral distance change trend between the target vehicle and the bound lane line is obtained.

[0058] For example, taking the tire point of the target vehicle as the cut-in detection point, Figure 2c a detection result graph for a target vehicle on a neighboring lane, Figure 2dThe figure is a detection result of a roadside stationary to moving target vehicle. The horizontal coordinate represents the frame number, and the vertical coordinate represents the lateral position in the world coordinate system. The tire validity judgment chart in the figure represents whether the front wheel or the rear wheel is valid, where 0 represents invalid, and 1 represents valid. It should be noted that the validity only represents whether the tire is clearly identifiable in the image and meets the detection conditions. When both tires are valid, the system preferentially selects the front wheel tire point as the cutting detection point for main analysis, and the data of the rear wheel tire point can be used as auxiliary verification reference. The tire binding line judgment chart represents whether the selected lane line of the front wheel or the rear wheel is on the left side or the right side. The left lane line is 0, and the right lane line is 1. The current lateral distance of the tire represents the lateral position deviation of the front wheel or the rear wheel from the binding lane line, where a positive number indicates that the cutting detection point is located on the outside of the binding lane line, and a negative number indicates that the cutting detection point is located on the inside of the binding lane line. As can be seen from the figure, the distance between the front wheel of the target vehicle and the binding lane line shows a continuous and stable decreasing trend. This regular change pattern can effectively represent the motion intention of the target vehicle approaching the lane of the ego vehicle. By this way, the discrete distance observation value is converted into continuous trend analysis, which can more accurately reflect the real motion intention of the target vehicle, avoid misjudgment caused by single point observation, and improve the reliability and stability of the judgment.

[0059] Alternatively, whether the target vehicle has a tendency to cut into the current lane can be determined by the following method: if the lateral distance change trend continuously meets the preset threshold value within the preset time range, it is determined that the target vehicle has a tendency to cut in.

[0060] The preset time range can refer to a continuous time period for verifying the stability of the trend, which can be the same as the target time range described above, or can be adjusted to a shorter or longer time period according to the system's demand for response speed. The specific value depends on ensuring that the trend has continuity rather than accidental change. The preset threshold value can refer to a trend condition set by humans to determine whether there is a cut-in intention.

[0061] For example, in the case that the target vehicle is on the right side of the reference vehicle and the current lateral distance is obtained by subtracting the horizontal coordinate value of the cut-in detection point from the horizontal coordinate value of the point on the lane line, the preset threshold can be the change trend of the current lateral distance gradually decreasing from a positive number (indicating that the cut-in detection point is on the right side of the lane binding line) to 0 (indicating that the cut-in detection point is on the lane binding line) or a negative number (indicating that the cut-in detection point is on the left side of the lane binding line). If the change trend of the lateral distance continuously meets the preset threshold within the preset time range, it means that the target vehicle has a cut-in tendency. Similarly, in the case that the target vehicle is on the left side of the reference vehicle and the current lateral distance is also obtained by subtracting the horizontal coordinate value of the cut-in detection point from the horizontal coordinate value of the point on the lane line, the preset threshold can be the change trend of the current lateral distance gradually increasing from a negative number to 0 or a positive number. If the change trend of the lateral distance continuously meets the preset threshold within the preset time range, it means that the target vehicle has a cut-in tendency.

[0062] Optionally, if the change trend of the lateral distance is a function fitting curve, the preset threshold can also be related to the slope of the function of the fitting curve. For example, in the case that the target vehicle is on the right side of the reference vehicle and the current lateral distance is obtained by subtracting the horizontal coordinate value of the cut-in detection point from the horizontal coordinate value of the point on the lane line, the preset threshold can be that the slope of the fitting curve is always positive. Similarly, in the case that the target vehicle is on the left side of the reference vehicle and the current lateral distance is also obtained by subtracting the horizontal coordinate value of the cut-in detection point from the horizontal coordinate value of the point on the lane line, the preset threshold can be that the slope of the fitting curve is always negative. By this method of setting the threshold, false judgments caused by instantaneous fluctuations are avoided, and only obvious cut-in behaviors can trigger the early warning, greatly improving the accuracy and practicality of the judgment.

[0063] In the above embodiment, by determining the two-dimensional coordinate data of the cut-in detection point on the target vehicle based on the local feature data, and then using the two-dimensional coordinate data of the cut-in detection point to develop the subsequent cut-in tendency judgment, the key point position that can early reflect the cut-in motion of the target vehicle is accurately locked, avoiding the dependence on the vehicle position and speed and other lagging parameters that need to accumulate obvious changes, and providing core data support with preposition for the cut-in tendency judgment. Then, the change trend of the lateral distance of the target vehicle is analyzed by using the target coordinate data of the cut-in detection point, realizing accurate measurement of the distance from image perception to actual physical space and accurate judgment of the motion trend, and effectively solving the problems of insufficient judgment accuracy and real-time in related technologies.

[0064] The embodiment of the present specification also provides a vehicle cut-in tendency judgment device 300, as shown in Figure 3 The device comprises an environment perception module 310, a target detection module 320 and a cut-in tendency judgment module 330, wherein: The environment perception module 310 is configured to acquire environment perception data of the reference vehicle in a normal driving state in a current lane; the environment perception data is used to describe a passable road condition of the reference vehicle in the current lane.

[0065] The target detection module 320 is configured to perform target detection and identification based on the environment perception data to determine local feature data of a target vehicle; the target vehicle refers to a surrounding traffic participant of the reference vehicle; the local feature data is used to calculate a position change degree of the target vehicle in a case of appearing lateral motion.

[0066] The cut-in tendency judgment module 330 is configured to perform cut-in tendency judgment of the target vehicle by using the local feature data to obtain a tendency result of the target vehicle cutting into the current lane.

[0067] In some embodiments, the target detection module 320 is further configured to determine the target vehicle within a preset range of the reference vehicle; and perform feature identification on the target vehicle by using the environment perception data to determine the local feature data of the target vehicle.

[0068] In some embodiments, the cut-in tendency judgment module 330 is further configured to determine two-dimensional coordinate data of a cut-in detection point of the target vehicle based on the local feature data; and determine whether the target vehicle has a tendency to cut into the current lane by using the two-dimensional coordinate data of the cut-in detection point, and obtain the tendency result of the target vehicle cutting into the current lane.

[0069] In some embodiments, the cut-in tendency judgment module 330 is further configured to perform coordinate system conversion on the two-dimensional coordinate data of the cut-in detection point to determine target coordinate data of the cut-in detection point in a world coordinate system; determine a binding lane line of the target vehicle based on the target coordinate data; and determine a lateral distance change trend between the target vehicle and the binding lane line by using the target coordinate data to determine whether the target vehicle has a tendency to cut into the current lane.

[0070] In some embodiments, the current lane includes a left lane line and a right lane line, and the device 300 for judging a vehicle cut-in tendency further includes a binding lane line determination module configured to determine first distance data between the target vehicle and the left lane line and second distance data between the target vehicle and the right lane line based on the target coordinate data; determine target distance data that uniquely satisfies a preset binding condition from the first distance data and the second distance data, and take a lane line corresponding to the target distance data as the binding lane line of the target vehicle.

[0071] In some implementations, the cutting tendency judgment module 330 is also used to determine the current lateral distance between the target vehicle and the bound lane line in each time frame for each time frame within the target time range using target coordinate data; and to integrate the current lateral distance between the target vehicle and the bound lane line in each time frame to obtain the trend of lateral distance change between the target vehicle and the bound lane line.

[0072] In some implementations, the cutting tendency judgment module 330 is used to determine that the target vehicle has a cutting tendency if the trend of lateral distance change continues to meet a preset threshold within a preset time range.

[0073] For specific limitations regarding a vehicle approach tendency determination device, please refer to the limitations of a vehicle approach tendency determination method described above, which will not be repeated here. Each module in the aforementioned vehicle approach tendency determination device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0074] In this embodiment, a vehicle cutting tendency determination device is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0075] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application, such as... Figure 4 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 4The processor 10 can be a central processing unit, a network processing unit, or a combination thereof. The processor 10 can further include a hardware chip. The hardware chip can be an application specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device can be a complex programmable logic device, a field programmable logic device, a general array logic, or any combination thereof. The memory 20 stores instructions executable by the at least one processor 10 for performing the methods as illustrated in the above embodiments. The memory 20 can include a program storage area and a data storage area. The program storage area can store an operating system and applications required by at least one function. The data storage area can store data created according to the use of the computer device. In addition, the memory 20 can include a high-speed random access memory, and can further include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some alternative embodiments, the memory 20 can optionally include a memory disposed remotely from the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0076] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30, and the output device 40 can be connected through a bus or other means, Figure 4 The input device 30 can receive input digital or character information, and generate key signal input related to the user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touchpad, a pointing stick, one or more mouse buttons, a trackball, a joystick, and the like. The output device 40 can include a display device, an auxiliary lighting device (e.g., an LED), a tactile feedback device (e.g., a vibration motor), and the like. The display device includes, but is not limited to, a liquid crystal display, a light emitting diode, a display, and a plasma display. In some alternative embodiments, the display device can be a touch screen.

[0077] The embodiments of the present application further provide a computer readable storage medium, and the method according to the embodiments of the present application can be implemented in hardware, firmware, or recorded in a storage medium, or be implemented as computer code originally stored in a remote storage medium or a non-transitory machine readable storage medium and downloaded to a local storage medium, so that the method described herein can be processed by such software on a storage medium using a general purpose computer, a special purpose processor, or programmable or special hardware. The storage medium can be a disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid state disk, etc. Further, the storage medium can also include a combination of the above-mentioned memories. It can be understood that the computer, the processor, the microprocessor controller, or the programmable hardware includes a storage component that can store or receive software or computer code, when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0078] The embodiments of the present application provide a computer program product, which includes computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method of any of the embodiments of the present application.

[0079] Although the embodiments of the present application are described with reference to the drawings, various modifications and changes can be made by those skilled in the art without departing from the spirit and scope of the present application, and such modifications and changes fall within the scope defined by the appended claims. For the convenience of description, the above device is described as various units respectively described in function. Of course, the functions of the units can be implemented in the same or more software and / or hardware in the implementation of the present application. Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0080] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more flowchart illustrations and / or one or more block diagrams.

[0081] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0082] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0083] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0084] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

[0085] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for determining a vehicle's tendency to cut in, characterized in that, The method includes: Acquire environmental perception data of the reference vehicle under normal driving conditions in the current lane; wherein, the environmental perception data is used to describe the passable road conditions of the reference vehicle in the current lane; Target detection and identification are performed based on the environmental perception data to determine the local feature data of the target vehicle; wherein, the target vehicle refers to the traffic participants surrounding the reference vehicle; the local feature data is used to calculate the degree of position change of the target vehicle when lateral movement occurs; The local feature data is used to determine the target vehicle's tendency to enter the current lane, and the result of the target vehicle's tendency to enter the current lane is obtained.

2. The method according to claim 1, characterized in that, The step of performing target detection and recognition based on the environmental perception data to determine the local feature data of the target vehicle includes: The target vehicle is determined within the preset range of the reference vehicle; The target vehicle is identified using the environmental perception data to determine its local feature data.

3. The method according to claim 1, characterized in that, The step of using the local feature data to determine the cut-in tendency of the target vehicle and obtaining the result of the target vehicle's tendency to cut into the current lane includes: Based on the local feature data, determine the two-dimensional coordinate data of the cut-in detection point on the target vehicle; The two-dimensional coordinate data of the cut-in detection point are used to determine whether the target vehicle has a tendency to cut into the current lane, and the result of the target vehicle's tendency to cut into the current lane is obtained.

4. The method according to claim 3, characterized in that, The step of determining whether the target vehicle has a tendency to cut into the current lane using the two-dimensional coordinate data of the cut-in detection point includes: A coordinate system transformation is performed on the two-dimensional coordinate data of the cutting detection point to determine the target coordinate data of the cutting detection point in the world coordinate system; The lane markings for the target vehicle are determined based on the target coordinate data; The target coordinate data is used to determine the trend of lateral distance change between the target vehicle and the bound lane line, so as to determine whether the target vehicle has a tendency to cut into the current lane.

5. The method according to claim 4, characterized in that, The current lane includes the left lane line and the right lane line, and the lane line bound to the target vehicle is determined by the following method: Based on the target coordinate data, determine the first distance data between the target vehicle and the left lane line and the second distance data between the target vehicle and the right lane line; In the first distance data and the second distance data, a target distance data that uniquely satisfies the preset binding conditions is determined, and the lane line corresponding to the target distance data is taken as the binding lane line of the target vehicle.

6. The method according to claim 4, characterized in that, Determining the trend of lateral distance change between the target vehicle and the bound lane line using the target coordinate data includes: For each time frame within the target time range, the target coordinate data is used to determine the current lateral distance between the target vehicle and the bound lane line in each time frame; By integrating the current lateral distance between the target vehicle and the bound lane line in each time frame, the trend of lateral distance change between the target vehicle and the bound lane line is obtained.

7. The method according to claim 4, characterized in that, The following methods are used to determine whether the target vehicle has a tendency to cut into the current lane: If the trend of lateral distance change continues to meet a preset threshold within a preset time range, it is determined that the target vehicle has a tendency to cut in.

8. A device for determining vehicle cutting tendency, characterized in that, The device includes: An environmental perception module is used to acquire environmental perception data of a reference vehicle under normal driving conditions in the current lane; wherein, the environmental perception data is used to describe the passable road conditions of the reference vehicle in the current lane. The target detection module is used to perform target detection and recognition based on the environmental perception data to determine the local feature data of the target vehicle; wherein, the target vehicle refers to the traffic participants surrounding the reference vehicle; the local feature data is used to calculate the degree of position change of the target vehicle when lateral movement occurs; The cutting tendency determination module is used to determine the cutting tendency of the target vehicle using the local feature data, and to obtain the result of the target vehicle's tendency to cut into the current lane.

9. A computer device, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the method of any one of claims 1 to 7.

Images