Stop line prediction method and device, medium and equipment

Abstract

The invention relates to a stop line prediction method and device, a medium and equipment. The stop line prediction method comprises the following steps: acquiring current stop line point cloud data at a current moment and historical stop line point cloud data at at least one historical moment under a world coordinate system by using data acquired by a vehicle-mounted sensor; fitting a historical stop line of at least one historical moment by using the historical stop line point cloud data of at least one historical moment; calculating a deviation value between a point cloud point in the point cloud data of the current stop line and each historical stop line, and determining an associated historical stop line associated with the point cloud data of the current stop line by using the deviation value; and fitting a fusion stop line at the current moment by using the current stop line point cloud data and the historical stop line point cloud data corresponding to the associated historical stop line. According to the technical scheme, the output of the stop line can be more stable, and the technical problem that the stop line is unstable in the prior art is solved.

Description

Technical Field

[0001] The present disclosure relates to the field of automobile technology, and in particular to a stop line prediction method, device, medium and equipment. Background Art

[0002] As the degree of automobile intelligence increases, the popularity of autonomous driving functions also increases. However, there are still many deficiencies in the current autonomous driving functions. For example, during the autonomous driving process, the vehicle computer needs to control the vehicle according to the position of the stop line in the lane. The stop line is currently mostly determined by using the Bird's Eye View (BEV) technology, which mainly obtains environmental information through sensors and other equipment, and then identifies the stop line from the environmental information and determines the positional relationship between the stop line and the vehicle. However, there are a large number of complex situations in the actual road environment, such as stop lines that are not perpendicular to the lane lines, stop lines that are blocked by other vehicles or obstacles, and changes in lighting. These situations are likely to cause the recognition of the same stop line in adjacent frames to have sudden changes in elements such as position, length, and line type, which in turn causes the stop line in the bird's eye view to flicker and the position to be erratic, and other problems such as unstable stop line output. Therefore, reducing the occurrence of unstable stop line output has become an urgent problem to be solved. Summary of the invention

[0003] In order to solve the above technical problems, the present disclosure provides a stop line prediction method, device, medium and equipment to solve the technical problem of unstable stop line in the prior art.

[0004] The present disclosure provides a stop line prediction method, comprising:

[0005] Using data collected by the vehicle-mounted sensor, obtain the current stop line point cloud data at the current moment and the historical stop line point cloud data at at least one historical moment in the world coordinate system;

[0006] Using historical stop line point cloud data of at least one historical moment to fit a historical stop line at least one historical moment;

[0007] Calculate the deviation value between the point cloud point in the current stop line point cloud data and each historical stop line, and use the deviation value to determine the associated historical stop line associated with the current stop line point cloud data;

[0008] The fused stop line at the current moment is fitted using the current stop line point cloud data and the historical stop line point cloud data corresponding to the associated historical stop line.

[0009] In some embodiments, fitting the fused stop line at the current moment using the current stop line point cloud data and the historical stop line point cloud data corresponding to the associated historical stop line includes:

[0010] The point cloud points in the current stop line point cloud data and the point cloud points in the historical stop line point cloud data corresponding to the associated historical stop line are combined into a point cloud set. For each point cloud point in the point cloud set, the point cloud point weight is generated using the acquisition time of the point cloud point and / or the distance between the point cloud point and the vehicle; the point cloud set and the point cloud point weight are used to fit the fused stop line at the current moment.

[0011] In some embodiments, the step of generating a point cloud point weight using the acquisition time of the point cloud point and / or the distance between the point cloud point and the vehicle includes:

[0012] Get the acquisition time of the point cloud point and the median acquisition time of all the point cloud points in the point cloud set, calculate the first difference between the acquisition time of the point cloud point and the median acquisition time and the second difference between the acquisition time of the point cloud point and the current time, and use the reciprocal of the sum of the first difference and the second difference as the point cloud point weight.

[0013] In some embodiments, the step of generating a point cloud point weight using the acquisition time of the point cloud point and / or the distance between the point cloud point and the vehicle includes:

[0014] The first distance between the point cloud point and the vehicle is obtained using the coordinates of the point cloud point in the world coordinate system, and the reciprocal of the first distance is used as the weight of the point cloud point.

[0015] In some embodiments, the step of generating a point cloud point weight using the acquisition time of the point cloud point and / or the distance between the point cloud point and the vehicle includes:

[0016] Obtain the acquisition time of the point cloud point and the median acquisition time of all the point cloud points in the point cloud set, calculate the first difference between the acquisition time of the point cloud point and the median acquisition time and the second difference between the acquisition time of the point cloud point and the current time, and use the reciprocal of the sum of the first difference and the second difference as the time difference; use the coordinates of the point cloud point in the world coordinate system to obtain the first distance between the point cloud point and the vehicle, and use the reciprocal of the first distance as the distance difference; use the sum of the product of the preset time weight and the time difference and the product of the preset distance weight and the distance difference as the point cloud point weight.

[0017] In some embodiments, the determining of the associated historical stop line associated with the current stop line point cloud data using the deviation value includes:

[0018] For each historical stop line, the deviation value between each point cloud point in the current stop line point cloud data and the historical stop line is calculated respectively, the deviation value of each point cloud point is accumulated and averaged to obtain the deviation mean, and the inverse of the deviation mean is used as the matching weight between the historical stop line and the current stop line; the matching algorithm is used to combine the matching weights of the historical stop line and the current stop line to obtain the associated historical stop line associated with the current stop line point cloud data.

[0019] In some embodiments, the method further comprises:

[0020] Determine whether the number of times the fusion stop line moment is obtained continuously before the current moment reaches the preset number, and if so, output the fusion stop line of the current moment.

[0021] The present disclosure also provides a stop line prediction device, comprising:

[0022] A collection module, used to obtain the current stop line point cloud data at the current moment and the historical stop line point cloud data at at least one historical moment in the world coordinate system using the data collected by the vehicle-mounted sensor;

[0023] A fitting module, used to fit a historical stop line at at least one historical moment using historical stop line point cloud data at at least one historical moment;

[0024] An association module, used to calculate the deviation value between the point cloud point in the current stop line point cloud data and each historical stop line, and determine the associated historical stop line associated with the current stop line point cloud data using the deviation value;

[0025] The fusion module is used to fit the fused stop line at the current moment by using the current stop line point cloud data and the historical stop line point cloud data corresponding to the associated historical stop line.

[0026] The present disclosure also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a program or instruction, wherein the program or instruction enables a computer to execute the steps of any of the above methods.

[0027] The present disclosure also provides an electronic device, comprising:

[0028] one or more processors;

[0029] A memory for storing one or more programs or instructions;

[0030] The processor is used to execute the steps of any of the above methods by calling the program or instruction stored in the memory.

[0031] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages:

[0032] The technical solution provided by the embodiment of the present disclosure utilizes data collected by vehicle-mounted sensors to obtain the current stop line point cloud data of the current moment and the historical stop line point cloud data of at least one historical moment in the world coordinate system, and fits the historical stop line of at least one historical moment, and then calculates the deviation value between the point cloud point in the current stop line point cloud data and each historical stop line, and utilizes the deviation value to determine the associated historical stop line associated with the current stop line point cloud data; utilizes the current stop line point cloud data and the historical stop line point cloud data corresponding to the associated historical stop line to fit the fused stop line at the current moment, which can avoid frequent jumps of the stop line, make the output of the stop line more stable, and solve the technical problem of unstable stop line in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0035] Figure 1 A flowchart of a stop line prediction method provided by an embodiment of the present disclosure;

[0036] Figure 2 It is a schematic diagram of stopping line output in an embodiment of the present disclosure;

[0037] Figure 3 A structural block diagram of a stop line prediction device provided by an embodiment of the present disclosure;

[0038] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0039] In order to more clearly understand the above-mentioned objectives, features and advantages of the present disclosure, the scheme of the present disclosure will be further described below. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0040] In the following description, many specific details are set forth to facilitate a full understanding of the present disclosure, but the present disclosure may also be implemented in other ways different from those described herein; it is obvious that the embodiments in the specification are only part of the embodiments of the present disclosure, rather than all of the embodiments.

[0041] The current stop line in the bird's-eye view may jump too much and frequently, change in length suddenly, appear or disappear suddenly, etc., causing technical problems such as unstable stop line, resulting in parking frustration and uncertainty in parking range.

[0042] Figure 1 This is a flow chart of a stop line prediction method provided by an embodiment of the present disclosure. This method is applicable to vehicle intelligent driving, can solve the technical problem of unstable stop lines, and can be applied to autonomous driving vehicles and manned vehicles. This method can be executed by a stop line prediction device, which can be implemented in software and / or hardware. Figure 1 As shown, the method comprises the following steps:

[0043] S110: Using data collected by the vehicle-mounted sensor, obtain current stop line point cloud data at the current moment and historical stop line point cloud data at at least one historical moment in the world coordinate system.

[0044] Optionally, data collected by vehicle-mounted sensors, where the vehicle-mounted sensors may be cameras, lidars, etc., and the vehicle computer may use the raw information collected by the sensors to obtain data containing information about the environment in which the vehicle is located; the number of sensors may be one or more, for example, a camera or lidar may be set in front of the vehicle, and the sensor may be used to collect images or point cloud information in front of the vehicle, and then generate data containing information about the environment in front of the vehicle, or multiple fisheye cameras may be set at multiple different positions of the vehicle, and then the collected multiple images may be used to generate data containing information about the environment around the vehicle. It should be noted that point cloud data is required in subsequent steps, and the process of obtaining data containing point cloud data based on the raw information collected by the sensors may use existing methods, such as obtaining depth information at each position in the image from the image to obtain point cloud data, etc., which is not limited here.

[0045] In addition, when the vehicle computer uses the data collected by the on-board sensors, the collected data can be stored, and the data can be distinguished by the time of data collection, such as the number of times the sensor collects data per unit time combined with the corresponding time, or the specific time can be used to distinguish the data.

[0046] Optionally, obtain the current stop line point cloud data at the current moment in the world coordinate system, wherein the vehicle computer can determine the point cloud data corresponding to the stop line at the current moment from the data collected by the on-board sensor, that is, the current stop line point cloud data at the current moment. In addition, since the position of the sensor on the vehicle is fixed, the specific external parameters of the sensor can also be determined, and then the specific position of the sensor in the vehicle coordinate system can be obtained by the existing method, and the internal parameters of the sensor itself are known, and the position of the point cloud corresponding to the original information collected by the sensor in the sensor coordinate system can be obtained at the time of collection. In this way, when the internal and external parameters of the sensor are determined, the current stop line point cloud data at the current moment in the vehicle coordinate system can be obtained by the existing coordinate system conversion method. In addition, the conversion relationship between the vehicle coordinate system and the world coordinate system can be obtained by existing methods such as vehicle positioning. At this time, the coordinates of each point cloud in the current stop line point cloud data in the world coordinate system can be determined by the existing coordinate system conversion method.

[0047] Optionally, historical stop line point cloud data of at least one historical moment in the world coordinate system is obtained, wherein the vehicle computer can obtain data corresponding to at least one historical moment from the stored data, and obtain the vehicle posture change information between the historical moment and the current moment for each historical moment, and use the posture change information and the internal and external parameters of the sensor to obtain the historical stop line point cloud data of the historical moment in the vehicle coordinate system; it should be noted that the vehicle coordinate system is a coordinate system established by the vehicle itself. As the actual position of the vehicle changes and its own posture (or orientation) changes, the relative position between the stop line and the vehicle at the historical moment will also change. Therefore, when obtaining the historical stop line point cloud data of the historical moment in the world coordinate system, it is necessary to introduce information such as vehicle positioning and posture change to determine the conversion relationship between the vehicle coordinate system corresponding to the historical moment and the world coordinate system; the vehicle position positioning and posture change information can be obtained through the odometer posture data in the vehicle; and then the posture change information, the internal and external parameters of the sensor, and the existing coordinate system conversion method can be used to obtain the historical stop line point cloud data of the historical moment in the world coordinate system.

[0048] S120: Fitting a historical stop line at at least one historical moment using the historical stop line point cloud data at at least one historical moment.

[0049] Optionally, the least squares method is used to fit the historical stop line point cloud data of each historical moment to form at least one historical stop line at a historical moment, such as using the coordinates of multiple points in the world coordinate system to fit a straight line as the stop line. Since the process of fitting multiple points to obtain a straight line or curve using the least squares method is an existing method, it is not repeated here. In addition, it is necessary to fit the stop line in each historical moment separately to avoid confusion between the historical stop line point cloud data of different historical moments.

[0050] The stop line can be described by a linear curve and the positions of the start and end points, that is, it can be described by the characteristic vector [c1, c0, start, end]. The linear curve expression of the historical stop line in the world coordinate system can be y=c1*x+c0, where c1 is a first-order coefficient, c0 is a constant, start is the coordinate of the starting point of the historical stop line in the world coordinate system, and end is the coordinate of the end point of the historical stop line in the world coordinate system.

[0051] like Figure 2 As shown, Figure 2 is a schematic diagram of the stop line output in the embodiment of the present disclosure, specifically a bird's-eye view of an intersection, where L1, L2, and L3 are historical stop lines at three intersections at a historical moment, and the historical stop lines are fitted using the historical stop line point cloud data. The intersection is a T-junction, so the historical stop lines of the three intersections can be identified at a historical moment; in addition, the vehicle can also only collect point cloud information within a certain distance in front of the vehicle, so that only the stop lines within a certain distance in front of the vehicle can be identified. If there are historical stop lines at multiple historical moments, there may be multiple stop lines in front of each intersection in the world coordinate system.

[0052] M is the vehicle. The vehicle is about to drive to the intersection and needs to obtain the accurate position of the stop line. The X-axis and Y-axis are the world coordinate system. The stop lines are mainly divided into XY lines and YX lines.

[0053] XY line: In the world coordinate system, calculate |Δx| / |Δy| from the start point to the end point of the stop line, where Δx is the difference between the start point and the end point of the stop line on the X-axis, and Δy is the difference between the start point and the end point of the stop line on the Y-axis. The line with a ratio greater than 1 is the XY line. In the figure, L2 is the XY line.

[0054] YX line: In the world coordinate system, calculate |Δx| / |Δy| from the start point to the end point of the stop line. The line with a ratio less than 1 is the YX line. In the figure, L1 and L3 are the YX lines.

[0055] like Figure 2 As shown, in the case where there are multiple intersections in the point cloud data, the intersections can be determined using the existing method, and then the corresponding current stop line point cloud data can be determined for each intersection. For example, B1, B2, and B3 are the current stop line point cloud data at the current moment. The current stop line point cloud data is composed of a bird's-eye view point cloud, and each current stop line point cloud data is usually composed of multiple point cloud points.

[0056] S130: Calculate the deviation value between the point cloud point in the current stop line point cloud data and each historical stop line, and use the deviation value to determine the associated historical stop line associated with the current stop line point cloud data.

[0057] In some embodiments, first, the deviation value between the point cloud point in the current stop line point cloud data and each historical stop line is calculated. Since there may be multiple historical stop lines, that is, there are historical stop lines at multiple historical moments, it is necessary to determine the deviation value between each point cloud point in the current stop line point cloud data and each of the multiple historical stop lines, and then determine the historical stop line associated with the current stop line point cloud data from the multiple historical stop lines as the associated historical stop line, so it is necessary to perform the following calculation on each current stop line point cloud data:

[0058] Optionally, for each historical stop line, the deviation value between each point cloud point in the current stop line point cloud data and the historical stop line is calculated respectively, and the deviation value of each point cloud point is accumulated and averaged to obtain the deviation mean; the historical stop line whose deviation mean is less than the preset mean is used as the associated historical stop line associated with the current stop line point cloud data.

[0059] During the calculation process, the distance between the point cloud point and the historical stop line in the world coordinate system can be used as the deviation value between the point cloud point and the historical stop line; after the deviation mean is obtained for each historical stop line, the preset mean (the specific value is not limited) can be used to compare with each historical stop line, and the historical stop line with a deviation mean less than the preset mean can be used as the associated historical stop line associated with the current stop line point cloud data.

[0060] For example, the deviation value between each point cloud point of the current stop line point cloud data B1 and the historical stop lines L1, L2, and L3 is calculated.

[0061] For the historical stop line, it is the YX line, such as L1 and L3:

[0062] Calculate the distance between the point cloud point and the end point of the historical stop line. If the Y-axis coordinate value of the point cloud point is greater than the Y-axis coordinate value of the end point of the historical stop line, and the corresponding distance between them exceeds the preset distance (usually set to 5m, which can be adaptively changed according to the actual width of the intersection), the distance between the point cloud point and the end point of the historical stop line is used as the deviation value between the point cloud point and the historical stop line; or

[0063] Calculate the distance between the point cloud point and the starting point of the historical stop line. If the Y-axis coordinate value of the point cloud point is less than the Y-axis coordinate value of the starting point of the historical stop line, and the corresponding distance between them exceeds the preset distance, the distance between the point cloud point and the starting point of the historical stop line is used as the deviation value between the point cloud point and the historical stop line; or

[0064] Calculate the distance between the point cloud point and the historical stop line along the X-axis direction of the world coordinate system as the deviation value;

[0065] Then, the sum of the deviation values ​​is divided by the number of point cloud points constituting the current stop line (eg, 20 points) to obtain the deviation mean.

[0066] For the historical stop line is the XY line, such as L2:

[0067] Calculate the distance between the point cloud point and the end point of the historical stop line. If the X-axis coordinate value of the point cloud point is greater than the X-axis coordinate value of the end point of the historical stop line, and the corresponding distance between them exceeds the preset distance, the distance between the point cloud point and the end point of the historical stop line is used as the deviation value between the point cloud point and the historical stop line; or

[0068] Calculate the distance between the point cloud point and the starting point of the historical stop line. If the X-axis coordinate value of the point cloud point is less than the X-axis coordinate value of the starting point of the historical stop line, and the corresponding distance between them exceeds the preset distance, the distance between the point cloud point and the starting point of the historical stop line is used as the deviation value between the point cloud point and the historical stop line; or

[0069] Calculate the distance between each point cloud point and the historical stop line along the Y-axis direction of the world coordinate system as the deviation value;

[0070] Then, the sum of the deviation values ​​is divided by the number of point cloud points constituting the current stop line (eg, 20 points) to obtain a deviation mean.

[0071] The above calculation method is only one method of calculating the deviation value between the point cloud point in the current stop line point cloud data and each historical stop line, and the specific calculation method is not limited here.

[0072] The method for determining the associated historical stop line associated with the current stop line point cloud data can be to compare the deviation mean with the preset mean, and the historical stop line less than the preset mean can be used as the associated historical stop line associated with the current stop line point cloud data.

[0073] For each historical stop line, the deviation value between each point cloud point in the current stop line point cloud data and the historical stop line is calculated respectively, and the deviation value of each point cloud point is accumulated and averaged to obtain the deviation mean, and the inverse of the deviation mean is used as the matching weight of the historical stop line and the current stop line; the matching algorithm is used to combine the matching weights of the historical stop line and the current stop line to obtain the associated historical stop line associated with the current stop line point cloud data. This method can then use the KM (Kuhn-Munkras) algorithm to associate each current stop line point cloud data with the corresponding historical stop line, that is, determine the associated historical stop line associated with each current stop line point cloud data. If the KM algorithm is used to calculate the matching method with the largest weight, and remove the matching pairs with a sum of weights less than 0.5, the associated current stop line point cloud data corresponds to the same real stop line as the historical stop line, that is, B1 is associated with L1, B2 is associated with L2, and B3 is associated with L3.

[0074] In the disclosed embodiment, the historical stop lines L1, L2, and L3 are all associated historical stop lines. In other embodiments, only a part of the multiple historical stop lines may be determined as associated historical stop lines. For example, after calculation, 6 of the 8 historical stop lines are determined to be associated historical stop lines.

[0075] S140: Fitting the fused stop line at the current moment by using the current stop line point cloud data and the historical stop line point cloud data corresponding to the associated historical stop line.

[0076] Optionally, the point cloud points in the current stop line point cloud data and the point cloud points in the historical stop line point cloud data corresponding to the associated historical stop are combined into a point cloud set, and for each point cloud point in the point cloud set, a point cloud point weight is generated using the collection time of the point cloud point and / or the distance between the point cloud point and the vehicle; the point cloud set and the point cloud point weight are used to fit the fused stop line at the current moment.

[0077] Among them, the point cloud points in the current stop line point cloud data and the point cloud points in the historical stop line point cloud data corresponding to the associated historical stop are both obtained, each point cloud point has coordinates in the world coordinate system, and these point cloud points can also form a point cloud set.

[0078] The point cloud point weight can be generated by using the acquisition time of the point cloud point, which includes obtaining the acquisition time of the point cloud point and the median acquisition time of all the point cloud points in the point cloud set, calculating the first difference between the acquisition time of the point cloud point and the median acquisition time and the second difference between the acquisition time of the point cloud point and the current time, and taking the reciprocal of the sum of the first difference and the second difference as the point cloud point weight. In this process, since the point cloud point is obtained by using the data collected by the vehicle-mounted sensor, the time when the sensor collects the data can be used as the acquisition time of the point cloud point.

[0079] The point cloud point weight is generated by using the distance between the point cloud point and the vehicle, and the first distance between the point cloud point and the vehicle is obtained by using the coordinates of the point cloud point in the vehicle coordinate system, and the reciprocal of the first distance is used as the point cloud point weight. In this process, the origin of the vehicle coordinate system can be used as the location of the vehicle, and the distance from the point cloud point to the origin of the vehicle coordinate system is used as the distance between the point cloud point and the vehicle.

[0080] The point cloud weights generated by using the acquisition time of the point cloud and / or the distance between the point cloud and the vehicle can be as follows: the acquisition time of the point cloud and the middle value of the acquisition time of all the point clouds in the point cloud set are obtained, the first difference between the acquisition time of the point cloud and the middle value of the acquisition time and the second difference between the acquisition time of the point cloud and the current time are calculated, and the reciprocal of the sum of the first difference and the second difference is used as the time difference; the first distance between the point cloud and the vehicle is obtained by using the coordinates of the point cloud in the vehicle coordinate system, and the reciprocal of the first distance is used as the distance difference; the sum of the product of the preset time weight and the time difference and the product of the preset distance weight and the distance difference is used as the point cloud weight. The preset time weight and the preset distance weight can be fixed values, or they can be adjusted according to the vehicle speed, such as the preset distance weight is proportional to the vehicle speed at the current moment, and the preset time weight is inversely proportional to the vehicle speed at the current moment, that is, the higher the vehicle speed, the greater the preset distance weight and the smaller the preset time weight, and vice versa, the smaller the vehicle speed, the smaller the preset distance weight and the larger the preset time weight.

[0081] Optionally, based on the weight of each point, curve fitting is performed on the point cloud in the point cloud set, that is, the current stop line and the associated historical stop line are fitted by the least square method to obtain a linear curve equation of the fused stop line.

[0082] By using filtering (such as Kalman filtering), the fused stop line can be denoised to obtain the fused stop line to be output. Kalman filtering is an optimal estimation algorithm, which means that the deviation between the obtained state and the true value of the system state is minimal. The estimation process is to calculate the optimal quantity at the current moment based on the instrument measurement value at the current moment and the predicted value at the previous moment, and predict the value at the next moment based on the optimal quantity. It can be seen that the data required by Kalman filtering is only the optimal value at the previous moment and the measurement value at the current moment. It is a forward recursive filtering algorithm, and the calculation and storage requirements are very small. It can be processed in real time, so it is very easy to implement in engineering. Considering the noise of the associated historical stop line and the current stop line, Kalman filtering is used to perform denoising to obtain the fused stop line to be output.

[0083] In some embodiments, the stop line prediction method further includes:

[0084] Determine whether the number of times the fusion stop line moment is obtained continuously before the current moment reaches the preset number, and if so, output the fusion stop line of the current moment.

[0085] Considering that the stop line parameters may need convergence time under Kalman filtering, they will be displayed only after the cumulative number of associated frames reaches a preset number, which can be set to 5-10 times, preferably 6 times. If the preset number is reached, the fused stop line is output, that is, the fused stop line obtained at the current moment is output and used as the historical stop line at the next moment, which is equivalent to updating Figure 2 L1, L2, and L3 in it.

[0086] The technical solution provided by the disclosed embodiment associates the current stop line point cloud data with the historical stop line point cloud data, and fuses the current stop line point cloud data with the historical stop line point cloud data to obtain the fused stop line at the current moment, thereby improving the stability of the position, direction and length of the stop line, and can avoid the discontinuity of vehicle planning and control, the sense of frustration, and the uncertainty of the parking range caused by the frequent jumping of the stop line, and also avoids the change of the length of the stop line, and the sudden appearance or disappearance of the stop line for a few frames, which causes the instability of the relationship between the stop line and the active lane, so that the L1, L2, and L3 output in each frame are more stable, which solves the technical problem of unstable stop line in the prior art.

[0087] Corresponding to the stop line prediction method provided by the embodiment of the present disclosure, the embodiment of the present disclosure also provides a stop line prediction device. Figure 3 A structural block diagram of a stop line prediction device provided by an embodiment of the present disclosure, such as Figure 3 As shown, the stop line prediction device comprises:

[0088] The acquisition module 31 is used to obtain the current stop line point cloud data at the current moment and the historical stop line point cloud data at at least one historical moment in the world coordinate system using the data collected by the vehicle-mounted sensor;

[0089] A fitting module 32, configured to fit a historical stop line at least at a historical moment using the historical stop line point cloud data at at least one historical moment;

[0090] An association module 33, used to calculate the deviation value between the point cloud point in the current stop line point cloud data and each historical stop line, and determine the associated historical stop line associated with the current stop line point cloud data using the deviation value;

[0091] The fusion module 34 is used to fit the fused stop line at the current moment by using the current stop line point cloud data and the historical stop line point cloud data corresponding to the associated historical stop line.

[0092] In some embodiments, the fusion module 34 is specifically used to:

[0093] The point cloud points in the current stop line point cloud data and the point cloud points in the historical stop line point cloud data corresponding to the associated historical stop line are combined into a point cloud set. For each point cloud point in the point cloud set, the point cloud point weight is generated using the acquisition time of the point cloud point and / or the distance between the point cloud point and the vehicle; the point cloud set and the point cloud point weight are used to fit the fused stop line at the current moment.

[0094] In some embodiments, the fusion module 34 is further configured to:

[0095] Get the acquisition time of the point cloud point and the median acquisition time of all the point cloud points in the point cloud set, calculate the first difference between the acquisition time of the point cloud point and the median acquisition time and the second difference between the acquisition time of the point cloud point and the current time, and use the reciprocal of the sum of the first difference and the second difference as the point cloud point weight.

[0096] In some embodiments, the fusion module 34 is further configured to:

[0097] The first distance between the point cloud point and the vehicle is obtained using the coordinates of the point cloud point in the world coordinate system, and the reciprocal of the first distance is used as the weight of the point cloud point.

[0098] In some embodiments, the fusion module 34 is further configured to:

[0099] Obtain the acquisition time of the point cloud point and the median acquisition time of all the point cloud points in the point cloud set, calculate the first difference between the acquisition time of the point cloud point and the median acquisition time and the second difference between the acquisition time of the point cloud point and the current time, and use the reciprocal of the sum of the first difference and the second difference as the time difference; use the coordinates of the point cloud point in the world coordinate system to obtain the first distance between the point cloud point and the vehicle, and use the reciprocal of the first distance as the distance difference; use the sum of the product of the preset time weight and the time difference and the product of the preset distance weight and the distance difference as the point cloud point weight.

[0100] In some embodiments, the association module 33 is specifically used to:

[0101] For each historical stop line, the deviation value between each point cloud point in the current stop line point cloud data and the historical stop line is calculated respectively, the deviation value of each point cloud point is accumulated and averaged to obtain the deviation mean, and the inverse of the deviation mean is used as the matching weight between the historical stop line and the current stop line; the matching algorithm is used to combine the matching weights of the historical stop line and the current stop line to obtain the associated historical stop line associated with the current stop line point cloud data.

[0102] In some embodiments, the fusion module 34 is further configured to:

[0103] Determine whether the number of times the fusion stop line moment is obtained continuously before the current moment reaches the preset number, and if so, output the fusion stop line of the current moment.

[0104] The stop line prediction device disclosed in the above embodiments can execute the stop line prediction method disclosed in the above embodiments, and has the same or corresponding beneficial effects. To avoid repetition, it will not be described again here.

[0105] The embodiment of the present disclosure also provides a computer-readable storage medium, which stores a program or instruction, and the program or instruction enables a computer to execute the steps of any of the above methods.

[0106] Using historical stop line point cloud data of at least one historical moment to fit a historical stop line at least one historical moment;

[0107] Calculate the deviation value between the point cloud point in the current stop line point cloud data and each historical stop line, and use the deviation value to determine the associated historical stop line associated with the current stop line point cloud data;

[0108] The fused stop line at the current moment is fitted using the current stop line point cloud data and the historical stop line point cloud data corresponding to the associated historical stop line.

[0109] Optionally, when executed by a computer processor, the computer executable instructions may also be used to execute the technical solution of any of the above-mentioned stop line prediction methods provided in the embodiments of the present disclosure, thereby achieving corresponding beneficial effects.

[0110] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the embodiments of the present disclosure can be implemented with the help of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the embodiments of the present disclosure is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment of the present disclosure.

[0111] The embodiment of the present disclosure also provides an electronic device, including: one or more processors; a memory for storing one or more programs or instructions; the processor calls the programs or instructions stored in the memory to execute the steps of any of the above methods to achieve corresponding beneficial effects.

[0112] Figure 4 Schematic diagram of the hardware structure of the electronic device provided by the embodiment of the present disclosure. Figure 4As shown, the electronic device includes one or more processors 401 and a memory 402 .

[0113] The processor 401 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0114] The memory 402 may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 401 may run the program instructions to implement the stop line prediction method of the embodiment of the present disclosure described above, and / or other desired functions. Various contents such as input signals, signal components, noise components, etc. may also be stored in the computer-readable storage medium.

[0115] In one example, the electronic device may further include: an input device 403 and an output device 404, and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0116] In addition, the input device 403 may also include, for example, a keyboard, a mouse, and the like.

[0117] The output device 404 can output various information to the outside, including the determined distance information, direction information, etc. The output device 404 can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, and the like.

[0118] Of course, to simplify, Figure 4 Only some of the components related to the present disclosure in the electronic device are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application situations, the electronic device may further include any other appropriate components.

[0119] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0120] The above description is only a specific embodiment of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A stop line prediction method, It is characterized in that include: Using data collected by the vehicle-mounted sensor, obtain the current stop line point cloud data at the current moment and the historical stop line point cloud data at at least one historical moment in the world coordinate system; Using historical stop line point cloud data of at least one historical moment to fit a historical stop line at least one historical moment; Calculate the deviation value between the point cloud point in the current stop line point cloud data and each historical stop line, and use the deviation value to determine the associated historical stop line associated with the current stop line point cloud data; The fused stop line at the current moment is fitted using the current stop line point cloud data and the historical stop line point cloud data corresponding to the associated historical stop line.

2. The method according to claim 1, It is characterized in that The method of fitting the fused stop line at the current moment by using the current stop line point cloud data and the historical stop line point cloud data corresponding to the associated historical stop line includes: The point cloud points in the current stop line point cloud data and the point cloud points in the historical stop line point cloud data corresponding to the associated historical stop line are combined into a point cloud set. For each point cloud point in the point cloud set, the point cloud point weight is generated using the acquisition time of the point cloud point and / or the distance between the point cloud point and the vehicle; the point cloud set and the point cloud point weight are used to fit the fused stop line at the current moment.

3. The method according to claim 2, It is characterized in that The step of generating a point cloud point weight by using the acquisition time of the point cloud point and / or the distance between the point cloud point and the vehicle includes: Get the acquisition time of the point cloud point and the median acquisition time of all the point cloud points in the point cloud set, calculate the first difference between the acquisition time of the point cloud point and the median acquisition time and the second difference between the acquisition time of the point cloud point and the current time, and use the reciprocal of the sum of the first difference and the second difference as the point cloud point weight.

4. The method according to claim 2, It is characterized in that The step of generating a point cloud point weight by using the acquisition time of the point cloud point and / or the distance between the point cloud point and the vehicle includes: The first distance between the point cloud point and the vehicle is obtained using the coordinates of the point cloud point in the world coordinate system, and the reciprocal of the first distance is used as the weight of the point cloud point.

5. The method according to claim 2, It is characterized in that The step of generating a point cloud point weight by using the acquisition time of the point cloud point and / or the distance between the point cloud point and the vehicle includes: Obtain the acquisition time of the point cloud point and the median acquisition time of all the point cloud points in the point cloud set, calculate the first difference between the acquisition time of the point cloud point and the median acquisition time and the second difference between the acquisition time of the point cloud point and the current time, and use the reciprocal of the sum of the first difference and the second difference as the time difference; use the coordinates of the point cloud point in the world coordinate system to obtain the first distance between the point cloud point and the vehicle, and use the reciprocal of the first distance as the distance difference; use the sum of the product of the preset time weight and the time difference and the product of the preset distance weight and the distance difference as the point cloud point weight.

6. The method according to claim 1, It is characterized in that The method of using the deviation value to determine the associated historical stop line associated with the current stop line point cloud data includes: For each historical stop line, the deviation value between each point cloud point in the current stop line point cloud data and the historical stop line is calculated respectively, the deviation value of each point cloud point is accumulated and averaged to obtain the deviation mean, and the inverse of the deviation mean is used as the matching weight between the historical stop line and the current stop line; the matching algorithm is used to combine the matching weights of the historical stop line and the current stop line to obtain the associated historical stop line associated with the current stop line point cloud data.

7. The method according to claim 1, It is characterized in that Also includes: Determine whether the number of times the fusion stop line moment is obtained continuously before the current moment reaches the preset number, and if so, output the fusion stop line of the current moment.

8. A stop line prediction device, It is characterized in that include: A collection module, used to obtain the current stop line point cloud data at the current moment and the historical stop line point cloud data at at least one historical moment in the world coordinate system using the data collected by the vehicle-mounted sensor; A fitting module, used to fit a historical stop line at at least one historical moment using historical stop line point cloud data at at least one historical moment; An association module, used to calculate the deviation value between the point cloud point in the current stop line point cloud data and each historical stop line, and determine the associated historical stop line associated with the current stop line point cloud data using the deviation value; The fusion module is used to fit the fused stop line at the current moment by using the current stop line point cloud data and the historical stop line point cloud data corresponding to the associated historical stop line.

9. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a program or instruction, and the program or instruction enables a computer to execute the steps of the method according to any one of claims 1 to 7.

10. An electronic device, It is characterized in that include: one or more processors; A memory for storing one or more programs or instructions; The processor is used to execute the steps of the method according to any one of claims 1 to 7 by calling the program or instruction stored in the memory.

Images