Road guardrail prediction method and device, and storage medium

Abstract

The application provides a road guardrail prediction method and device, a storage medium and a computer program product, wherein the method comprises: receiving point cloud data collected by a target vehicle-mounted radar, the point cloud data comprising position information and speed information of each radar point; performing curve fitting on the point cloud data with a Doppler speed less than a target speed value to obtain a fitting curve; and determining the position of the guardrail according to the fitting curve in the case that the confidence of the fitting curve is greater than a threshold value. In the technical scheme of the application, the fitting curve can represent the shape of the guardrail, and the position of the guardrail can be determined according to the fitting curve in the case that the confidence of the fitting curve is greater than the threshold value, so that the confidence and accuracy of the predicted guardrail position and shape are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer processing, and in particular to a road guardrail prediction method and device and a storage medium. BACKGROUND

[0002] With the development of intelligent driving technology, intelligent driving systems gradually transit from auxiliary driving systems to advanced automatic driving systems. For higher-level automatic driving systems, the output characteristics of sensors are no longer limited to the detection and identification of obstacles on the road. For higher-level automatic driving functions, the environmental perception system needs to not only perceive conventional obstacles on the road such as vehicles, pedestrians, and animals, but also pay attention to the position or shape of the road edge, such as the guardrail.

[0003] How to provide a high-precision road guardrail prediction scheme is a topic generally considered in the industry. SUMMARY

[0004] The present application provides a road guardrail prediction method, device and storage medium to solve the low accuracy of existing road guardrail prediction in the prior art and achieve a high-precision road guardrail prediction scheme.

[0005] The present application provides a road guardrail prediction method, which comprises:

[0006] Receiving point cloud data collected by a target vehicle-mounted radar, the point cloud data comprising position information and speed information of each radar point;

[0007] Performing curve fitting on the point cloud data with a Doppler speed less than a target speed value to obtain a fitting curve;

[0008] In a case where the confidence of the fitting curve is greater than a threshold value, determining the position of the guardrail according to the fitting curve.

[0009] According to the road guardrail prediction method provided by the present application, before performing curve fitting on the point cloud data with a Doppler speed less than a target speed value to obtain a fitting curve, the method further comprises:

[0010] Dividing the point cloud data on at least one side of the vehicle in the lateral direction of the vehicle into point cloud intervals along the driving direction of the vehicle;

[0011] Comparing the number of radar points in each point cloud interval, and determining the point cloud data in the point cloud interval that satisfies the target number condition as target point cloud data;

[0012] The curve fitting on the point cloud data with a Doppler speed less than a target speed value comprises:

[0013] Fitting a curve using the target point cloud data with a Doppler velocity less than a target velocity value to obtain a fitting curve.

[0014] According to the road guardrail prediction method provided in the application, the target quantity condition is that the number of radar points corresponding to the target point cloud data is greater than the number of radar points in other point cloud intervals.

[0015] According to the road guardrail prediction method provided in the application, the point cloud data on at least one side of the vehicle in the lateral direction is divided into point cloud intervals along the driving direction of the vehicle, including:

[0016] The point cloud data of the target frame number on at least one side of the vehicle in the lateral direction is fused.

[0017] The fused point cloud data is divided into point cloud intervals along the driving direction of the vehicle.

[0018] According to the road guardrail prediction method provided in the application, the point cloud data with a Doppler velocity less than a target velocity value is fitted to obtain a fitting curve, including:

[0019] A quadratic curve is fitted using the target point cloud data, and the starting point and the ending point of the quadratic curve are determined.

[0020] Analog point clouds are inserted between the starting point and the ending point, and the analog point clouds form the fitting curve.

[0021] According to the road guardrail prediction method provided in the application, in the case that the confidence of the fitting curve is greater than a threshold value, before determining the position of the guardrail according to the fitting curve, the method further includes:

[0022] A test value is calculated based on the fitting curve for the target point cloud data, and the error between the test value and the true value of the target point cloud data is calculated, and an exponential function error is calculated for the mean square value of each test value error, and the exponential function error is taken as the confidence;

[0023] The confidence is compared with the threshold value.

[0024] According to the road guardrail prediction method provided in the application, the method further includes:

[0025] In the case that the confidence does not exceed the threshold value, the point cloud data on at least one side of the vehicle in the lateral direction is returned to be divided into point cloud intervals along the driving direction of the vehicle.

[0026] The application also provides a road guardrail prediction device, including:

[0027] A collection module collects point cloud data through a vehicle-mounted radar, and the point cloud data includes position information and speed information of each radar point.

[0028] a fitting module, configured to perform curve fitting on the point cloud data with the Doppler velocity less than the target speed value to obtain a fitting curve;

[0029] a determining module, configured to determine the position of the guardrail according to the fitting curve in a case where a confidence degree of the fitting curve is greater than a threshold value.

[0030] According to the road guardrail prediction device provided by the present application, the device further comprises:

[0031] a dividing module, configured to divide the point cloud data of at least one side of the ego vehicle in the lateral direction of the ego vehicle into point cloud intervals before performing curve fitting on the point cloud data with the Doppler velocity less than the target speed value to obtain a fitting curve;

[0032] a first comparing module, configured to compare the number of radar points in each point cloud interval, and determine the point cloud data in the point cloud interval in which the number of radar points meets a target number condition as target point cloud data;

[0033] The fitting module is specifically configured to:

[0034] perform curve fitting on the target point cloud data with the Doppler velocity less than the target speed value to obtain a fitting curve.

[0035] According to the road guardrail prediction device provided by the present application, the dividing module is specifically configured to:

[0036] fuse the point cloud data of the target frame number on at least one side of the ego vehicle in the lateral direction;

[0037] divide the fused point cloud data into point cloud intervals in the driving direction of the ego vehicle.

[0038] According to the road guardrail prediction device provided by the present application, the fitting module is specifically configured to:

[0039] fit a quadratic curve using the target point cloud data, and determine the starting point and the ending point of the quadratic curve;

[0040] generate the fitting curve by inserting simulated point clouds between the starting point and the ending point.

[0041] According to the road guardrail prediction device provided by the present application, the device further comprises:

[0042] a calculating module, configured to, before determining the position of the guardrail according to the fitting curve in a case where a confidence degree of the fitting curve is greater than a threshold value, calculate a test value based on the fitting curve for the target point cloud data, and calculate an exponential function error based on the error between the test value and the true value of the target point cloud data and the mean square value of each test value error, and take the exponential function error as the confidence degree;

[0043] The second comparison module compares the confidence level with the threshold value.

[0044] The present application also provides an electronic device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the road barrier prediction method according to any one of the above when executing the program.

[0045] The present application also provides a non-transitory computer readable storage medium, which stores a computer program, wherein the computer program is executable by a processor to implement the steps of the road barrier prediction method according to any one of the above.

[0046] The present application also provides a computer program product, comprising a computer program, wherein the computer program is executable by a processor to implement the steps of the road barrier prediction method according to any one of the above.

[0047] The present application provides a road barrier prediction method, device, storage medium and computer program product, wherein the road barrier prediction method comprises: receiving point cloud data collected by a target vehicle-mounted radar, the point cloud data comprising position information and speed information of each radar point; performing curve fitting on the point cloud data with a Doppler speed less than a target speed value to obtain a fitting curve; and determining the position of the barrier according to the fitting curve in a case where the confidence level of the fitting curve is greater than a threshold value.

[0048] In the technical scheme of the present application, the fitting curve can represent the shape of the barrier, and the position of the barrier can be determined according to the fitting curve in a case where the confidence level of the fitting curve is greater than a threshold value, so as to improve the credibility and accuracy of the predicted barrier position and shape. BRIEF DESCRIPTION OF DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0050] Figure 1 is one of the flowcharts of the road barrier prediction method provided by the present application;

[0051] Figure 2 is another flowchart of the road barrier prediction method provided by the present application;

[0052] Figure 3 is a third flowchart of the road barrier prediction method provided by the present application;

[0053] Figure 4 is a fourth flowchart of the road guardrail prediction method provided by the present application;

[0054] Figure 5 is a curve fitting schematic diagram in the road guardrail prediction method provided by the present application;

[0055] Figure 6 is a first structural schematic diagram of the road guardrail prediction device provided by the present application;

[0056] Figure 7 is a second structural schematic diagram of the road guardrail prediction device provided by the present application;

[0057] Figure 8 is a third structural schematic diagram of the road guardrail prediction device provided by the present application;

[0058] Figure 9 is a structural schematic diagram of the electronic device provided by the present application. DETAILED DESCRIPTION

[0059] In order to make the objectives, technical solutions and advantages of the present application clearer, the technical solutions in the present application will be clearly and completely described below with reference to the drawings in the present application. Obviously, the described embodiments are some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0060] The road guardrail prediction method of the present application will be described below. Figures 1-4 The execution subject of the present method is a vehicle-mounted terminal or an automatic driving system arranged in the vehicle-mounted terminal or a road guardrail prediction system installed in the automatic driving system.

[0061] With reference to Figure 1 , the road guardrail prediction method provided by the present application includes the following steps:

[0062] Step 110: receiving point cloud data collected by a target vehicle-mounted radar, the point cloud data including position information and speed information of each radar point;

[0063] Step 120: performing curve fitting on the point cloud data with a Doppler speed less than a target speed value to obtain a fitting curve;

[0064] Step 130: in the case that the confidence of the fitting curve is greater than a threshold value, determining the position of the guardrail according to the fitting curve.

[0065] In the present embodiment, the vehicle-mounted radar collects road point cloud data in real time.

[0066] Specifically, the vehicle-mounted radar can be a millimeter wave radar, which is a radar working in a millimeter wave band. The working principle of the millimeter wave radar is to generate electromagnetic waves of a specific modulation frequency (FMCW) by using a high-frequency circuit, and to send and receive electromagnetic waves through an antenna, and to calculate various parameters of the target by parameters of the sent and received electromagnetic waves.

[0067] The received electromagnetic waves are converted into a point cloud form, and the point cloud data includes position information and speed information of each radar point. The position information of the radar point cloud can be represented by coordinate values in the ego vehicle coordinate system, for example, in an x-y coordinate system, the x direction is the driving direction of the ego vehicle, and the y direction is the lateral direction of the ego vehicle. Each point cloud can be mapped to the x-y coordinate system and represented by corresponding coordinate values.

[0068] Optionally, the vehicle-mounted radar can also be a laser radar or other radar. The millimeter wave radar can maintain sufficient signal strength of the antenna in a smaller integrated space.

[0069] Optionally, the angle of the vehicle-mounted radar is not limited, and can be at the rear of the ego vehicle, the lateral sides, or the front of the ego vehicle.

[0070] In the embodiment of the present application, the fitting curve can represent the shape of the guardrail, and the position of the guardrail can be determined according to the fitting curve when the confidence of the fitting curve is greater than a threshold, thereby improving the credibility and accuracy of the predicted guardrail position and shape.

[0071] Optionally, with reference to Figure 2 The road guardrail prediction method provided in the embodiment includes the following steps:

[0072] Step 210: receiving point cloud data collected by a target vehicle-mounted radar, the point cloud data including position information and speed information of each radar point;

[0073] Step 220: dividing the point cloud data on at least one lateral side of the ego vehicle into point cloud intervals along the driving direction of the ego vehicle;

[0074] Step 230: comparing the number of radar points in each point cloud interval, and determining the point cloud data in the point cloud interval satisfying a target number condition as target point cloud data;

[0075] Step 240: performing curve fitting on the target point cloud data with a Doppler speed less than a target speed value to obtain a fitting curve;

[0076] Step 250: determining the position of the guardrail according to the fitting curve when the confidence of the fitting curve is greater than a threshold.

[0077] Specifically, the embodiment divides the point cloud data on both lateral sides of the ego vehicle into multiple point cloud intervals along the driving direction, such as setting a point cloud interval every 2 meters or other distances laterally, and selects the target point cloud data contained in the point cloud interval that meets the target number condition of the number of radar points as the basis for curve fitting.

[0078] Optionally, the at least one lateral side of the ego vehicle includes one side or both sides.

[0079] Optionally, the target number condition is that the number of radar points corresponding to the target point cloud data is greater than the number of radar points in other point cloud intervals, i.e., the number of radar points corresponding to the target point cloud data is the most.

[0080] Optionally, with reference to Figure 3 dividing the point cloud data on the at least one lateral side of the ego vehicle into point cloud intervals along the driving direction of the ego vehicle, specifically including the following steps:

[0081] Step 310: fusing the point cloud data of the target frame number on the at least one lateral side of the ego vehicle;

[0082] Step 320: dividing the fused point cloud data into point cloud intervals along the driving direction of the ego vehicle.

[0083] Through point cloud data fusion, a larger number of radar point clouds can be obtained, and the higher the concentration of radar point clouds corresponding to the position of the guardrail, the higher the accuracy of the fitted curve in the later stage.

[0084] Optionally, with reference to Figure 4 fitting a curve using point cloud data with a Doppler velocity less than a target speed value to obtain a fitted curve, specifically including the following steps:

[0085] Step 410: fitting a quadratic curve using the target point cloud data and determining the starting point and ending point of the quadratic curve;

[0086] Step 420: inserting simulated point clouds between the starting point and the ending point, and the simulated point clouds form the fitted curve.

[0087] wherein the quadratic curve can be in the form of a parabola, and the corresponding parameters are determined using the least squares method, and the purpose of fitting is to determine the shape of the parabola, including its starting point and ending point.

[0088] wherein on both lateral sides of the ego vehicle, each side of the quadratic curve portion has a starting point and an ending point, and the simulated point clouds are inserted between the starting point and the ending point to form the fitted curve, and the fitted curve portions on both sides are regarded as the shape and position of the guardrail.

[0089] wherein the fitted curve can be represented as ax^2+bx+c=y in the ego vehicle coordinate system x-y.

[0090] Optionally, the method further comprises the following steps:

[0091] In the case that the confidence of the fitting curve is greater than a threshold, before determining the position of the guardrail according to the fitting curve, a test value is calculated based on the fitting curve for the target point cloud data, and an error between the test value and a true value of the target point cloud data is calculated, and an exponential function error is calculated for a mean square value of each test value error, and the exponential function error is taken as the confidence;

[0092] The confidence is compared with the threshold.

[0093] The confidence represents the reliability of the current fitting curve, and if it is higher than the threshold, it indicates that the reliability of the fitting curve is high, and otherwise the confidence is insufficient.

[0094] Reference Figure 5 As shown, the test values y1, y2, y3 are calculated by using the fitting curve y=ax2+bx+c and the coordinates x1, x2, x3 of the corresponding radar points in the target point cloud data along the x-axis direction, and the errors are calculated by using the test values y1, y2, y3 and the true values y1', y2', y3' corresponding to x1, x2, x3 in the target point cloud data, respectively. This is an example, and more radar point errors will be calculated in actual application, and the mean square value is calculated, which can improve the accuracy of the confidence judgment of the fitting curve.

[0095] Optionally, the road guardrail prediction method provided by the embodiment of the present application further comprises the following steps:

[0096] In the case that the confidence does not exceed the threshold, the point cloud data on at least one side of the vehicle in the lateral direction can be returned to divide the point cloud interval along the driving direction of the vehicle, and the point cloud interval is re-divided.

[0097] The road guardrail prediction device provided by the present application is described below, and the road guardrail prediction device described below can be correspondingly referred to the road guardrail prediction method described above.

[0098] Reference Figure 6 The road guardrail prediction device provided by the embodiment of the present application comprises:

[0099] The acquisition module 610 collects point cloud data through a vehicle-mounted radar, and the point cloud data comprises position information of each radar point;

[0100] The fitting module 620 performs curve fitting on the point cloud data with a Doppler velocity less than a target speed value to obtain a fitting curve;

[0101] The determination module 630 determines the position of the guardrail according to the fitting curve in the case that the confidence of the fitting curve is greater than a threshold.

[0102] Optionally, referring to Figure 7 , compared with Figure 6 , the device further comprises:

[0103] The division module 710 divides the point cloud data on at least one side of the ego vehicle in the driving direction of the ego vehicle before performing curve fitting on the point cloud data with a Doppler velocity less than a target velocity value to obtain a fitting curve.

[0104] The first comparison module 720 compares the number of radar points in each point cloud interval, and determines the point cloud data in the point cloud interval that meets the target number condition as target point cloud data.

[0105] The fitting module 730 is specifically configured to:

[0106] Perform curve fitting on the target point cloud data with a Doppler velocity less than a target velocity value to obtain a fitting curve.

[0107] Optionally, the division module 710 is specifically configured to:

[0108] Fuse the point cloud data of the target frame number on at least one side of the ego vehicle;

[0109] Divide the fused point cloud data in the driving direction of the ego vehicle.

[0110] Optionally, the fitting module 730 is specifically configured to:

[0111] Fit a quadratic curve using the target point cloud data, and determine the starting point and the ending point of the quadratic curve;

[0112] Generate the fitting curve by inserting simulated point clouds between the starting point and the ending point.

[0113] Optionally, referring to Figure 8 , compared with Figure 7 , the device further comprises:

[0114] The calculation module 810, in the case that the confidence degree of the fitting curve is greater than a threshold value, before determining the position of the guardrail according to the fitting curve, calculates a test value based on the fitting curve from the target point cloud data, and compares the error between the test value and the true value of the target point cloud data, and calculates an exponential function error of the mean square value of each test value error, and takes the exponential function error as the confidence degree.

[0115] The second comparison module 820 compares the confidence degree with the threshold value.

[0116] Figure 9 An example of an electronic device is shown in the physical structure diagram, asFigure 9 As shown, the electronic device can include a processor 910, a communications interface 920, a memory 930, and a communications bus 940, wherein the processor 910, the communications interface 920, and the memory 930 complete mutual communication through the communications bus 940. The processor 910 can invoke a logical instruction in the memory 930 to execute a road guardrail prediction method, which includes:

[0117] receiving point cloud data collected by a target vehicle-mounted radar, the point cloud data including position information and speed information of each radar point;

[0118] performing curve fitting on the point cloud data with a Doppler speed less than a target speed value to obtain a fitting curve;

[0119] in a case where a confidence degree of the fitting curve is greater than a threshold value, determining a position of the guardrail according to the fitting curve.

[0120] In addition, the logical instruction in the memory 930 described above can be implemented in the form of a software function unit and sold or used as an independent product, and can be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0121] On the other hand, the present application also provides a computer program product, the computer program product includes a computer program, the computer program can be stored on a non-transitory computer readable storage medium, when the computer program is executed by a processor, the computer can execute the road guardrail prediction method provided by the above-mentioned methods, the method includes:

[0122] receiving point cloud data collected by a target vehicle-mounted radar, the point cloud data including position information and speed information of each radar point;

[0123] performing curve fitting on the point cloud data with a Doppler speed less than a target speed value to obtain a fitting curve;

[0124] In a case where a confidence of the fitting curve is greater than a threshold value, a position of the guardrail is determined according to the fitting curve.

[0125] In another aspect, the present application also provides a non-transitory computer readable storage medium having stored thereon a computer program, which, when executed by a processor, implements a road guardrail prediction method provided by each of the above methods, and the method comprises:

[0126] Receiving point cloud data collected by a target vehicle-mounted radar, the point cloud data comprising position information and speed information of each radar point;

[0127] Performing curve fitting on the point cloud data with a Doppler speed less than a target speed value to obtain a fitting curve;

[0128] In a case where a confidence of the fitting curve is greater than a threshold value, a position of the guardrail is determined according to the fitting curve.

[0129] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., they can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement it without creative labor.

[0130] From the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be realized by means of software and the necessary general hardware platform, and of course, it can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, server, or network device, etc.) execute the methods described in each embodiment or some parts of the embodiment.

[0131] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for predicting road guardrails, characterized in that, include: Receive point cloud data collected by the target vehicle-mounted radar, wherein the point cloud data includes the position information and speed information of each radar point; The point cloud data with Doppler velocity less than the target velocity value is used for curve fitting to obtain the fitted curve; If the confidence level of the fitted curve is greater than a threshold, the position of the guardrail is determined based on the fitted curve. Before performing curve fitting on the point cloud data where the Doppler velocity is less than the target velocity value to obtain the fitted curve, the method further includes: Divide the point cloud data of at least one side of the vehicle into point cloud intervals along the vehicle's driving direction; Compare the number of radar points in each point cloud interval, and determine the point cloud data in the point cloud interval where the number of radar points meets the target number condition as the target point cloud data. The step of using point cloud data with Doppler velocity less than the target velocity value to perform curve fitting to obtain a fitted curve includes: Curve fitting is performed using the target point cloud data where the Doppler velocity is less than the target velocity value to obtain the fitted curve; The point cloud data of at least one side of the vehicle is divided into point cloud intervals along the vehicle's driving direction, including: Point cloud data fusion for at least one target frame on the lateral side of the vehicle; The merged point cloud data is divided into point cloud intervals along the vehicle's driving direction; The step of using point cloud data with Doppler velocity less than the target velocity value to perform curve fitting to obtain a fitted curve includes: A quadratic curve is obtained by fitting the target point cloud data, and the starting point and ending point of the quadratic curve are determined. A simulated point cloud is inserted between the start and end points, and the simulated point cloud forms the fitted curve. Before determining the position of the guardrail based on the fitted curve if the confidence level of the fitted curve is greater than a threshold, the method further includes: Test values ​​are calculated based on the fitted curve for the target point cloud data, and the error between the test values ​​and the true values ​​of the target point cloud data is calculated. An exponential function error is then calculated based on the mean square value of the error of each test value, and the exponential function error is used as the confidence level. Compare the confidence level with the threshold.

2. The road guardrail prediction method according to claim 1, characterized in that, The target quantity condition is that the number of radar points corresponding to the target point cloud data is greater than the number of radar points in other point cloud intervals.

3. The road guardrail prediction method according to claim 1, characterized in that, The method further includes: If the confidence level does not exceed the threshold, the point cloud data of at least one side of the vehicle is returned and divided into point cloud intervals along the vehicle's driving direction.

4. A road guardrail prediction device, characterized in that, include: The acquisition module collects point cloud data through vehicle-mounted radar. The point cloud data includes the position and speed information of each radar point. The fitting module uses the point cloud data where the Doppler velocity is less than the target velocity value to perform curve fitting and obtain the fitted curve. The determination module determines the position of the guardrail based on the fitted curve if the confidence level of the fitted curve is greater than a threshold. Before performing curve fitting on the point cloud data with a Doppler velocity less than the target velocity value to obtain the fitted curve, the point cloud data of at least one side of the vehicle's lateral direction is divided into point cloud intervals along the vehicle's driving direction. The first comparison module compares the number of radar points in each point cloud interval and determines the point cloud data in the point cloud interval where the number of radar points meets the target number condition as the target point cloud data. The fitting module is specifically used for: Curve fitting is performed using the target point cloud data where the Doppler velocity is less than the target velocity value to obtain the fitted curve; The partitioning module is specifically used for: Point cloud data fusion for at least one target frame on the lateral side of the vehicle; The merged point cloud data is divided into point cloud intervals along the vehicle's driving direction; The fitting module is specifically used for: A quadratic curve is obtained by fitting the target point cloud data, and the starting point and ending point of the quadratic curve are determined. The fitted curve is generated by inserting a simulated point cloud between the start and end points; The device further includes: The calculation module calculates test values ​​for the target point cloud data based on the fitted curve before determining the position of the guardrail according to the fitted curve when the confidence level of the fitted curve is greater than a threshold. It also calculates the error between the test values ​​and the true values ​​of the target point cloud data, and calculates the exponential function error for the mean square value of the error of each test value. The exponential function error is then used as the confidence level. The second comparison module compares the confidence level with the threshold.

5. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the road guardrail prediction method as described in any one of claims 1 to 3.

6. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the road guardrail prediction method as described in any one of claims 1 to 3.

7. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the road guardrail prediction method as described in any one of claims 1 to 3.

Images