Road edge extraction method based on vehicle-mounted millimeter wave radar and electronic equipment

Abstract

The invention discloses a road edge extraction method based on a vehicle-mounted millimeter wave radar and electronic equipment, and is applied to the technical field of millimeter wave radar data processing. The method comprises the following steps: acquiring driving information of a vehicle and trace point data of a millimeter wave radar; estimating a vehicle driving track according to the driving information and the trace point data; obtaining historical road edge trace points, and verifying the historical road edge trace points to eliminate invalid trace points; according to the vehicle driving track, screening and updating the current road edge trace points to obtain effective road edge points; and extracting the road edge through an RANSAC (Random Sample Consensus) algorithm under the condition that an effective road edge extraction result exists currently. According to the method, the track information of the vehicle and the trace point information detected by the millimeter wave radar are fully utilized, invalid points on the inner side and the outer side are removed through the historical road edge recognition result, the RASCAN calculation amount is reduced, meanwhile, the influence of non-road-edge obstacles on the result is avoided, the algorithm is stable, efficient and accurate, the method can well fit the road shape, and meanwhile the calculation amount is low.

Description

Technical Field

[0001] The present invention relates to the technical field of millimeter wave radar data processing, and in particular to a curb extraction method and electronic equipment based on a vehicle-mounted millimeter wave radar. Background Art

[0002] In the perception of intelligent driving environment, curb extraction is a crucial link. Curb can distinguish between driving area and non-driving area, and assist in eliminating false targets. Existing research mainly uses cameras and lidar as sensors, and uses images or laser point clouds to extract curbs, but cameras and lidar are greatly affected by the environment and weather. At the same time, the image processing and laser point cloud data volume is large and the calculation is large. Compared with optical sensors and lidar, millimeter wave radar is not easily affected by weather and has lower cost. With the development of its technology, it is gradually used in the field of intelligent driving. Therefore, it is of great significance to conduct research on curb extraction based on vehicle-mounted millimeter wave radar.

[0003] Among the existing curb extraction schemes, the point trace clustering-based method has high requirements for the compactness and continuity of the curb points, and the radar resolution on the market is difficult to meet its requirements; the curb extraction scheme based on Hough transform is generally applied to straight roads, and the traversal process is computationally intensive and consumes a lot of space resources; in addition, there is a technical solution that uses the multipath reflection relationship between multipath radar targets and real radar targets to invert multipath reflection points to obtain the road edge of the target road, but it requires the existence of multipath false targets and correctly judges the real target pairs that match them, which is difficult to implement in practical applications. Therefore, it is of great value to study a robust, efficient, and engineering-suitable vehicle-mounted millimeter-wave radar curb extraction method. Summary of the invention

[0004] In order to solve at least one of the technical problems existing in the prior art to a certain extent, the object of the present invention is to provide a curb extraction method, electronic equipment and medium based on a vehicle-mounted millimeter-wave radar.

[0005] The first technical solution adopted by the present invention is:

[0006] A method for extracting a curb based on a vehicle-mounted millimeter-wave radar comprises the following steps:

[0007] Obtain vehicle driving information and millimeter wave radar point data;

[0008] Estimating the vehicle's driving trajectory based on driving information and point trace data;

[0009] Obtain historical roadside traces, and verify the historical roadside traces to eliminate invalid traces;

[0010] According to the vehicle's driving trajectory, the current roadside points are screened and updated to obtain valid roadside points;

[0011] When there is a valid road edge extraction result, the road edge is extracted using the RANSAC algorithm.

[0012] Furthermore, the number of the millimeter-wave radars is 2, and the two millimeter-wave radars are respectively installed on both sides of the rear of the vehicle.

[0013] Furthermore, the vehicle driving information includes a vehicle speed V car and yaw angular velocity Yaw; the point trace data includes the distance R, angle θ and velocity V of the point trace;

[0014] After obtaining the point data, the distance R and angle θ in the polar coordinate system are converted to the rectangular coordinate system of the vehicle body to obtain its position information (X, Y)

[0015] X=R*cos(θ)

[0016] Y=R*sin(θ)

[0017] Convert the velocity V to the radial velocity V`:

[0018] V`=V+Vs*cos(θ)

[0019] Where Vs is the vehicle speed.

[0020] Furthermore, the vehicle driving trajectory estimation based on the driving information and the point trace data includes:

[0021] Calculate the vehicle's motion displacement dx and dy between the previous frame and the current frame:

[0022] dx=V car *T*cos(Yaw*T / 2)

[0023] dy=V car *T*sin(Yaw*T / 2)

[0024] Where T is the frame period;

[0025] Get the historical trajectory points of the vehicle, and transform the coordinates of the historical motion trajectory points to the global coordinate system according to the motion displacement. The expression is as follows:

[0026] x2=(x1-dx)*cos(-Yaw*T)-(x1-dx)*sin(-Yaw*T)

[0027] y2=(y1-dy)*cos(-Yaw*T)+(y1-dy)*sin(-Yaw*T)

[0028] Where (x1, y1) is the coordinate of the historical trajectory point before conversion, and (x2, y2) is the coordinate of the historical trajectory point after conversion;

[0029] According to the obtained coordinates, the historical trajectory points of the vehicle are fitted to obtain the vehicle driving trajectory.

[0030] Furthermore, the obtaining of historical roadside traces and verifying the historical roadside traces to eliminate invalid traces includes:

[0031] Obtain the stored historical points of the left and right curbs, and transform the obtained historical points of the curbs into the global coordinate system at the current moment by means of coordinate transformation;

[0032] According to the coordinates (x, y) of the point trace, the position difference between the historical roadside point trace and all historical point traces is calculated. If the position difference is within the preset threshold range, the historical roadside point trace is retained, otherwise it is cleared.

[0033] Furthermore, the filtering and updating of the current roadside point trace according to the vehicle driving trajectory includes:

[0034] Define the point trace motion and stillness threshold, and perform motion and stillness judgment and elimination on the current point trace: if the radial velocity V' of the point trace is greater than the preset threshold, it is determined to be a moving target and the point trace is eliminated; if the radial velocity V' of the point trace is less than or equal to the preset threshold, it is determined to be a stationary point trace and the subsequent update steps are continued;

[0035] For the static point traces, based on the vehicle driving trajectory obtained by fitting, the target point traces of the current frame are screened and distinguished on the left and right curbs, and saved in the left and right curb point sets respectively;

[0036] After determining the left and right affiliation of the current frame point trace, calculate the horizontal and vertical coordinate differences between the left point trace and all historical left roadside point traces. If the horizontal and vertical coordinate differences meet the difference threshold, the point trace is determined to be a valid left roadside point and is retained in the left roadside point set, otherwise it is deleted; calculate the horizontal and vertical coordinate differences between the right point trace and all historical right roadside point traces. If the horizontal and vertical coordinate differences meet the difference threshold, the point trace is determined to be a valid right roadside point and is retained in the right roadside point set, otherwise it is deleted.

[0037] Furthermore, when a valid road edge extraction result already exists, the road edge is extracted by using the RANSAC algorithm, including:

[0038] In the case that there are valid curb extraction results, the RANSAC algorithm is used to extract curb curves:

[0039] For the left curb, randomly select 4 curb points from the left curb points, fit a third-order curve, then traverse the left curb points, define a threshold, count the number of points within the distance curve threshold, and then define a number threshold. When searching for points that meet the preset number threshold, record them as possible curb curves; perform multiple iterations, find the fitting curve with the largest number of points as the final curb curve, and update the curb return;

[0040] For the right curb, four curb points are randomly selected from the right curb points, and a third-order curve is fitted. Then, the right curb points are traversed, a threshold is defined, and the number of points within the distance curve threshold is counted. Then, a number threshold is defined, and when searching for points that meet the preset number threshold, they are recorded as possible curb curves. Multiple iterations are performed to find the fitting curve with the largest number of points as the final curb curve, and the return curve of the curb is updated.

[0041] Furthermore, the curb extraction method further comprises the step of extracting an initial curb:

[0042] When there is no valid curb extraction result, the vehicle trajectory curve parameters are fitted based on the third-order polynomial;

[0043] The vehicle trajectory curve is translated to the left on the Y axis within a preset range, and the left curb point trace is traversed each time the translation is performed, a threshold is defined, and the number of left curb point traces within the threshold range of the translation curve is counted; a quantity threshold is defined, and the vehicle translation curve that meets the preset quantity threshold is determined as a valid curb curve;

[0044] The vehicle trajectory curve is translated rightward on the Y-axis within the preset range, and the right curb points are traversed each time the translation is performed. A threshold is defined, and the number of right curb points within the threshold range of the translation curve is counted; a quantity threshold is defined, and the vehicle translation curve that meets the preset quantity threshold is determined as a valid curb curve.

[0045] Furthermore, the curb extraction method further comprises the step of judging the validity of the curb:

[0046] The vehicle trajectory, the left curb extraction result, and the right curb extraction result are judged pairwise whether there is an intersection in their effective intervals; if there is an intersection, the curb extraction is determined to be invalid, and the curb extraction result is initialized.

[0047] The second technical solution adopted by the present invention is:

[0048] An electronic device comprises a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement a curb extraction method based on a vehicle-mounted millimeter-wave radar as described above.

[0049] The third technical solution adopted by the present invention is:

[0050] A computer-readable storage medium, wherein at least one instruction, at least one program, a code set or an instruction set is stored in the storage medium, wherein the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by a processor to implement a roadside extraction method based on a vehicle-mounted millimeter-wave radar as described above.

[0051] The fourth technical solution adopted by the present invention is:

[0052] A computer program product or a computer program includes computer instructions stored in a computer-readable storage medium. A processor of a computer device can read the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the above method.

[0053] The beneficial effects of the present invention are as follows: the present invention proposes a curb extraction scheme based on a vehicle-mounted millimeter-wave radar, which can make full use of the vehicle trajectory information and the point trace information detected by the millimeter-wave radar in a complex road environment, and use the vehicle's driving trajectory to quickly and robustly initialize the road edge in the initial state, and then according to the continuous extension characteristics of the road edge, remove the invalid points inside and outside through the historical road edge recognition results, reduce the amount of RASCAN calculation, and avoid the influence of non-curb obstacles on its results. The algorithm is robust, efficient, and accurate, can better fit the road shape, and has a low amount of calculation, which is very suitable for engineering applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the embodiments of the present invention or the drawings of related technical solutions in the prior art are introduced below. It should be understood that the drawings introduced below are only for the convenience of clearly describing some embodiments of the technical solutions of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0055] Figure 1 is a flow chart of a method for extracting a curb based on a vehicle-mounted millimeter-wave radar in an embodiment of the present invention;

[0056] Figure 2 1 is an installation diagram of a vehicle-mounted millimeter-wave radar in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limitations of the present invention. For the step numbers in the following embodiments, they are only provided for the convenience of explanation, and the order between the steps is not limited in any way, and the execution order of each step in the embodiment can be adaptively adjusted according to the understanding of those skilled in the art.

[0058] In the description of the present invention, it should be understood that descriptions involving orientations, such as up, down, front, back, left, right, etc., and orientations or positional relationships indicated are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0059] In the description of the present invention, "several" means one or more, "more" means more than two, "greater than", "less than", "exceed" etc. are understood as not including the number itself, and "above", "below", "within" etc. are understood as including the number itself. If there is a description of "first" or "second", it is only used for the purpose of distinguishing the technical features, and cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features.

[0060] In the description of the present invention, unless otherwise clearly defined, terms such as setting, installing, connecting, etc. should be understood in a broad sense, and technicians in the relevant technical field can reasonably determine the specific meanings of the above terms in the present invention based on the specific content of the technical solution.

[0061] Terminology explanation:

[0062] RANSAC algorithm: The abbreviation of Random Sample Consensus. It is an algorithm that calculates the mathematical model parameters of the data based on a set of sample data sets containing abnormal data to obtain valid sample data.

[0063] Example 1

[0064] like Figure 1 As shown, this embodiment provides a roadside extraction method based on a vehicle-mounted millimeter-wave radar, comprising the following steps:

[0065] S1: Acquisition of vehicle driving information and millimeter wave radar point trace data.

[0066] The vehicle driving information is read, and according to the millimeter wave radar echo data, the point trace data is obtained through the radar signal processing step and the data preprocessing step. In this embodiment, the point trace data is obtained by two vehicle-mounted millimeter wave radars, and the two radars are respectively installed at the rear of the vehicle, such as Figure 2 As shown, the center of the rear axle of the vehicle is taken as the coordinate origin, the horizontal direction is the Y axis, and the vertical direction is the X axis.

[0067] In some embodiments, step S1 specifically includes the following steps:

[0068] S1-1: Get vehicle driving information, the speed is V car , yaw angular velocity Yaw;

[0069] S1-2: The vehicle-mounted millimeter-wave radar scans the road environment and obtains millimeter-wave radar echo data. After radar signal processing and data preprocessing steps, point data is obtained, including the distance, angle, speed, signal-to-noise ratio and other information of the point, and the frame period is T. The roadside extraction mainly uses the distance R, angle θ, and speed V of the point.

[0070] S1-3: Convert the distance and angle information in the polar coordinate system to the rectangular coordinate system of the vehicle body to obtain its position information (X, Y):

[0071] X=R*cos(θ)

[0072] Y=R*sin(θ)

[0073] S1-4: Convert the velocity to the radial velocity V` relative to the ground:

[0074] V`=V+Vs*cos(θ)

[0075] S2: Road trajectory estimation.

[0076] According to the vehicle speed and yaw angle information, the vehicle's historical motion trajectory is converted to the current radar coordinate system through vehicle track calculation compensation, and the vehicle's driving trajectory is fitted using a polynomial.

[0077] As an optional implementation, step S2 includes the following steps:

[0078] S2-1: Calculate the vehicle's motion displacement dx and dy between the previous frame and the current frame:

[0079] dx=V car *T*cos(Yaw*T / 2)

[0080] dy=V car *T*sin(Yaw*T / 2)

[0081] S2-2: Extract the historical trajectory points of the vehicle, and compensate the displacement and angle transformation of the vehicle, and transform the coordinates of the historical motion trajectory points into the global coordinate system. The coordinates of the historical trajectory points of the vehicle before transformation are (x1, y1), and the coordinates after transformation are (x2, y2). The transformation process is as follows:

[0082] x2=(x1-dx)*cos(-Yaw*T)-(x1-dx)*sin(-Yaw*T)

[0083] y2=(y1-dy)*cos(-Yaw*T)+(y1-dy)*sin(-Yaw*T)

[0084] S2-3: Use the least squares algorithm to perform a third-order fitting on the historical trajectory points of the vehicle to obtain a polynomial curve of the historical trajectory of the vehicle.

[0085] S3: Update and verify historical roadside points.

[0086] Convert the historical curb points to the current coordinate system, align and verify them using the curb extraction results of the previous frame, and remove invalid points.

[0087] As an optional implementation, step S3 includes the following steps:

[0088] S3-1: Extract the stored historical points of the left and right curbs, and convert them to the global coordinate system at the current moment in the same way as in S2.

[0089] S3-2: If there is a valid roadside extraction result, perform a historical roadside point validity check. According to the x and y coordinates of the point trace, calculate the position difference between the historical roadside point trace and all historical point traces. If the difference meets the threshold range, it is retained, otherwise it is cleared.

[0090] S4: Filter and update the current roadside points.

[0091] Specifically, based on the vehicle driving trajectory obtained in step S2, the current point track is divided into left and right curb points, and then screened to eliminate moving points and invalid points, and the new information is updated to the historical curb points.

[0092] As an optional implementation, step S4 includes the following steps:

[0093] S4-1: Define the motion and stillness threshold of the point trace, and perform motion and stillness judgment and elimination on the current point trace: the point trace with V` greater than the threshold is determined as a moving target and eliminated; the point trace with V` less than or equal to the threshold is determined as a stationary point trace, and the subsequent update steps are continued.

[0094] S4-2: For the stationary points, based on the polynomial curve of the vehicle trajectory fitted in step S2, the target points of the current frame are screened and distinguished from the left and right curbs, and saved in the left and right curb point sets respectively.

[0095] The specific method is as follows: substitute the horizontal coordinate value of the point track into the polynomial curve of the vehicle trajectory to obtain the polynomial value, and compare it with the vertical coordinate of the point track. If the vertical coordinate of the point track is greater than the polynomial value, the point track is located on the left side of the vehicle trajectory curve and is determined to be a left curb point; if the vertical coordinate of the point track is less than the polynomial value, the point track is located on the right side of the vehicle trajectory curve and is determined to be a right curb point.

[0096] S4-3: After determining the left and right affiliation of the point trace in the current frame, calculate the horizontal and vertical coordinate differences between the left point trace and all historical left roadside point traces. If the horizontal and vertical coordinate differences meet the difference threshold, it indicates that the point trace is a valid left roadside point and it is retained in the left roadside point set, otherwise it is deleted; the screening method for right roadside points is the same.

[0097] S5: Curb extraction in the initial situation.

[0098] In the case that there is no valid curb extraction result, the left and right initial curbs are obtained by extrapolating the driving trajectory. Specifically, in the case that there is no valid curb extraction result, based on the vehicle trajectory curve parameters fitted by the third-order polynomial, the vehicle trajectory curve is translated to the left on the Y axis within a certain range. Each translation traverses the left curb points, defines a threshold, and counts the number of left curb points within the range of the translation curve threshold; defines a number threshold, and the vehicle translation curve that meets the number threshold is a valid curb curve; the right curb extraction method is the same.

[0099] S6: Roadside extraction based on RANSAC.

[0100] In the case that valid curb extraction results already exist, the RANSAC algorithm is used for multiple iterations to keep the best road edge status updated. Specifically, in the case that valid curb extraction results already exist, points outside the effective extraction area have been eliminated in S4-3, and the curb curve is extracted using the RANSAC algorithm. For the left curb, 4 curb points are randomly selected from the left curb points, a third-order curve is fitted, and then the left curb points are traversed, a threshold is defined, and the number of points within the distance curve threshold range is counted. A number threshold is then defined, and when searching for a curve that meets the number threshold, it is recorded as a possible curb curve. Perform multiple iterations to find the fitting curve with the largest number of points as the final curb curve, and update the return curb; the right curb extraction method is the same.

[0101] S7: Curb validity determination.

[0102] The vehicle trajectory, the left curb extraction result, and the right curb extraction result are judged to see if there is any intersection in their valid intervals. If there is an intersection, the curb extraction is considered invalid and the curb extraction result is initialized.

[0103] In summary, this embodiment is based on millimeter-wave radar point trace data and vehicle driving information. First, the vehicle driving information is used to fit the historical driving trajectory curve of the vehicle. Then, the roadside points are screened and accumulated according to the latest target point trace information. The vehicle driving trajectory is used to distinguish the left and right roadside points. In the initial recognition state, the left and right initial roadsides are obtained by extrapolating the driving trajectory. Then, the RANSAC algorithm is iterated multiple times to keep the best roadside state updated, and finally the roadside validity is verified. Through this method, the point traces detected by the vehicle-mounted millimeter-wave radar are processed, and the roadside point traces can be effectively screened and the roadside can be extracted.

[0104] Example 2

[0105] An embodiment of the present invention further provides an electronic device, the electronic device comprising a processor and a memory, the memory storing at least one instruction, at least one program, a code set or an instruction set, the at least one instruction, the at least one program, the code set or the instruction set being loaded and executed by the processor to implement the following Figure 1 A roadside extraction method based on vehicle-mounted millimeter-wave radar is shown.

[0106] It is understood that the memory may include a random access memory (RAM) or a read-only memory (ROM). Optionally, the memory includes a non-transitory computer-readable storage medium. The memory may be used to store instructions, programs, codes, code sets, or instruction sets. The memory may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function, instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data created according to the use of the server, etc.

[0107] The processor may include one or more processing cores. The processor uses various interfaces and lines to connect the various parts of the entire server, and executes various functions of the server and processes data by running or executing instructions, programs, code sets or instruction sets stored in the memory, and calling data stored in the memory. Optionally, the processor can be implemented in at least one hardware form of digital signal processing (DSP), field programmable gate array (FPGA), and programmable logic array (PLA). The processor can integrate one or a combination of a central processing unit (CPU) and a modem. Among them, the CPU mainly processes the operating system and application programs; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor, but implemented separately through a chip.

[0108] Since the electronic device is an electronic device corresponding to a curb extraction method based on a vehicle-mounted millimeter-wave radar in an embodiment of the present invention, and the principle of solving the problem by the electronic device is similar to that of the method, the implementation of the electronic device can refer to the implementation process of the above-mentioned method embodiment, and the repeated parts will not be repeated.

[0109] Example 3

[0110] The embodiment of the present invention further provides a computer-readable storage medium, wherein the storage medium stores at least one instruction, at least one program, code set or instruction set, and the at least one instruction, the at least one program, the code set or instruction set is loaded and executed by a processor to implement the following Figure 1 A roadside extraction method based on vehicle-mounted millimeter-wave radar is shown.

[0111] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0112] Since the storage medium is a storage medium corresponding to a curb extraction method based on a vehicle-mounted millimeter-wave radar in an embodiment of the present invention, and the principle of solving the problem by the storage medium is similar to that of the method, the implementation of the storage medium can refer to the implementation process of the above-mentioned method embodiment, and the repeated parts will not be repeated.

[0113] Example 4

[0114] In some possible implementations, various aspects of the method of the embodiment of the present invention may also be implemented in the form of a program product, which includes a program code. When the program product is run on a computer device, the program code is used to enable the computer device to execute the steps of a roadside extraction method based on a vehicle-mounted millimeter-wave radar according to various exemplary embodiments of the present application described above in this specification. Among them, the executable computer program code or "code" for executing each embodiment can be written in a high-level programming language such as C, C++, C#, Smalltalk, Java, JavaScript, Visual Basic, structured query language (e.g., Transact-SQL), Perl, or in various other programming languages.

[0115] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0116] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0117] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable ordinary technicians in the field to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent changes or modifications made based on the essence of the content of the present invention should be included in the protection scope of the present invention.

Claims

1. A roadside extraction method based on vehicle-mounted millimeter-wave radar, characterized in that: The following steps are involved: Obtain vehicle driving information and millimeter wave radar point data; Estimating vehicle driving trajectory based on driving information and point trace data; Obtain historical roadside traces, and verify the historical roadside traces to eliminate invalid traces; According to the vehicle's driving trajectory, the current roadside points are screened and updated to obtain valid roadside points; When there is a valid road edge extraction result, the road edge is extracted using the RANSAC algorithm.

2. The method for extracting a curb based on a vehicle-mounted millimeter-wave radar according to claim 1, characterized in that: The number of the millimeter-wave radars is 2, and the two millimeter-wave radars are respectively installed on both sides of the rear of the vehicle.

3. The method for extracting a curb based on a vehicle-mounted millimeter-wave radar according to claim 1, characterized in that: The vehicle driving information includes a vehicle speed V car and yaw angular velocity Yaw; the point trace data includes the distance R, angle θ and velocity V of the point trace; After obtaining the point data, the distance R and angle θ in the polar coordinate system are converted to the vehicle body rectangular coordinate system to obtain its position information (X, Y) X=R*cos(θ) Y=R*sin(θ) Convert the velocity V to the radial velocity V relative to the ground: V, = V + Vs * Cos (θ) Where Vs is the vehicle speed.

4. The method for extracting a curb based on a vehicle-mounted millimeter-wave radar according to claim 1, characterized in that: The vehicle driving trajectory estimation based on the driving information and the point trace data includes: Calculate the vehicle's motion displacement dx and dy between the previous frame and the current frame: dx=V car *T*cos(Yaw*T / 2) dy=V car *T*sin(You*T / 2) Where T is the frame period; Get the historical trajectory points of the vehicle, and transform the coordinates of the historical motion trajectory points to the global coordinate system according to the motion displacement. The expression is as follows: x2=(x1-dx)*cos(-Yaw*T)-(x1-dx)*sin(-Yaw*T) y2=(y1-dy)*cos(-Yaw*T)+(y1-dy)*sin(-Yaw*T) Where (x1, y1) is the coordinate of the historical trajectory point before conversion, and (x2, y2) is the coordinate of the historical trajectory point after conversion; According to the obtained coordinates, the historical trajectory points of the vehicle are fitted to obtain the vehicle driving trajectory.

5. The method for extracting a curb based on a vehicle-mounted millimeter-wave radar according to claim 1, characterized in that: The obtaining of historical roadside traces and verifying the historical roadside traces to eliminate invalid traces includes: Obtain the stored historical points of the left and right curbs, and transform the obtained historical points of the curbs into the global coordinate system at the current moment by means of coordinate transformation; According to the coordinates (x, y) of the point trace, the position difference between the historical roadside point trace and all historical point traces is calculated. If the position difference is within the preset threshold range, the historical roadside point trace is retained, otherwise it is cleared.

6. The method for extracting a curb based on a vehicle-mounted millimeter-wave radar according to claim 1, characterized in that: The filtering and updating of the current roadside points according to the vehicle driving trajectory includes: Define the point trace motion and stillness threshold, and perform motion and stillness judgment and elimination on the current point trace: if the radial velocity V' of the point trace is greater than the preset threshold, it is determined to be a moving target and the point trace is eliminated; if the radial velocity V' of the point trace is less than or equal to the preset threshold, it is determined to be a stationary point trace and the subsequent update steps are continued; For the static point traces, based on the vehicle driving trajectory obtained by fitting, the target point traces of the current frame are screened and distinguished on the left and right curbs, and saved in the left and right curb point sets respectively; After determining the left and right affiliation of the current frame point trace, calculate the horizontal and vertical coordinate differences between the left point trace and all historical left roadside point traces. If the horizontal and vertical coordinate differences meet the difference threshold, the point trace is determined to be a valid left roadside point and is retained in the left roadside point set, otherwise it is deleted; calculate the horizontal and vertical coordinate differences between the right point trace and all historical right roadside point traces. If the horizontal and vertical coordinate differences meet the difference threshold, the point trace is determined to be a valid right roadside point and is retained in the right roadside point set, otherwise it is deleted.

7. The method for extracting a curb based on a vehicle-mounted millimeter-wave radar according to claim 1, characterized in that: If there are valid roadside extraction results, the roadside is extracted using the RANSAC algorithm, including: In the case that there are valid curb extraction results, the RANSAC algorithm is used to extract curb curves: For the left curb, four curb points are randomly selected from the left curb points, and a third-order curve is fitted. Then, the left curb points are traversed, and the number of points within the distance curve threshold is counted. When searching for a point that meets the preset number threshold, it is recorded as a possible curb curve. Multiple iterations are performed to find the fitting curve with the largest number of points as the final curb curve. For the right curb, four curb points are randomly selected from the right curb points, and a third-order curve is fitted. Then, the right curb points are traversed, and the number of points within the distance curve threshold is counted. When searching for points that meet the preset number threshold, they are recorded as possible curb curves. Multiple iterations are performed to find the fitting curve with the largest number of points as the final curb curve.

8. The method for extracting a curb based on a vehicle-mounted millimeter-wave radar according to claim 1, characterized in that: The curb extraction method further comprises the step of extracting an initial curb: In the case that there is no valid curb extraction result at present, the vehicle trajectory curve parameters are fitted based on the third-order polynomial; the vehicle trajectory curve is translated to the left on the Y axis within the preset range, and the left curb point traces are traversed each time the translation is performed, and the number of left curb point traces within the range of the translation curve threshold is counted; the vehicle translation curve that meets the preset number threshold is determined as a valid curb curve; The vehicle trajectory curve is translated rightward on the Y axis within a preset range, and the right curb points are traversed each time the translation is performed. The number of right curb points within the translation curve threshold is counted; the vehicle translation curve that meets the preset number threshold is determined as a valid curb curve.

9. The method for extracting a curb based on a vehicle-mounted millimeter-wave radar according to claim 1, characterized in that: The curb extraction method further comprises the step of curb validity judgment: The vehicle trajectory, the left curb extraction result, and the right curb extraction result are judged pairwise whether there is an intersection in their effective intervals; if there is an intersection, the curb extraction is determined to be invalid, and the curb extraction result is initialized.

10. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or the instruction set is loaded and executed by the processor to implement the method described in any one of claims 1 to 9.

Images