Road boundary detection method and system based on vehicle-mounted millimeter-wave radar

Abstract

The present invention discloses a road boundary detection method and system based on an in-vehicle millimeter-wave radar. The method first preliminarily screens the input static target points of the in-vehicle millimeter-wave radar to obtain relatively stable static target points, and screens all possible road boundary points from the stable static target points by using a strategy of clustering all boundary points with the first starting point by setting a search range. Then, combined with the boundary points screened in the previous frame of the radar, a quadratic curve fitting is performed by using the least squares method. Finally, considering that the road boundary will not jump in a short time, the Kalman filtering algorithm is used to filter the three coefficients in the obtained quadratic curve, and a relatively stable curve output in the time domain is obtained, thereby improving the accuracy of road boundary detection.

Description

Technical Field

[0001] The present invention relates to the technical field of road detection, and particularly to a road boundary detection method and system based on an in-vehicle millimeter-wave radar. Background Art

[0002] Traditional road boundary detection is generally achieved by using lidar. However, lidar is expensive and cannot be popularized. In-vehicle millimeter-wave radar is inexpensive, has good ranging and speed measurement capabilities for targets, good penetration ability for rain and fog, and is not affected by light intensity, making it an irreplaceable sensor choice in intelligent driving solutions.

[0003] Due to the interference of vehicles and the like on highways and urban roads, the output of the road boundary line obtained from the radar data collected by the in-vehicle millimeter-wave radar using the existing detection method is often unstable in the time domain, resulting in low detection accuracy. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a road boundary detection method and system based on an in-vehicle millimeter-wave radar, which solves the problem that the output of the road boundary line obtained from the radar data collected by the in-vehicle millimeter-wave radar in the prior art is unstable in the time domain and has low detection accuracy.

[0005] On the one hand, the present invention provides a road boundary detection method based on an in-vehicle millimeter-wave radar, including:

[0006] Obtaining all static target points of the current frame output by the in-vehicle millimeter-wave radar;

[0007] Screening out a plurality of stable static target points from all the static coordinate points according to an initial screening range, where the initial screening range is determined according to the radar detection ability;

[0008] Taking the stable static target points located within a preset screening range as alternative boundary starting points; wherein, with the radar central axis as the symmetry axis, two search ranges are set within the field of view angle range of the in-vehicle millimeter-wave radar, and the two set search ranges are used as the preset screening range. Each search range takes the sector radius boundary corresponding to the field of view angle range of the in-vehicle millimeter-wave radar as one boundary, and the other boundary is parallel to the sector radius boundary;

[0009] Sorting the alternative boundary starting points in ascending order of the distance from the radar origin, and taking the alternative boundary starting point with the smallest distance as the boundary starting point;

[0010] Starting from the boundary starting point, search for boundary points among the stable static target points within a preset search range, and use the found boundary points as new starting points to continue searching for the next boundary point until no next boundary point can be found;

[0011] Use the least squares method to perform quadratic curve fitting on all the found boundary points to obtain the quadratic curve coefficients of the current frame;

[0012] Use the Kalman filtering algorithm to perform filtering processing on the quadratic curve coefficients of the current frame and the quadratic curve coefficients of the previous frame to obtain the target quadratic curve coefficients and the target road boundary curve.

[0013] Optionally, it further includes: if multiple boundary points are found within the preset search range, calculate the distance between each boundary point and the radar origin;

[0014] Sort all the boundary points in ascending order according to the obtained distances, and use the boundary point corresponding to the minimum distance as the new starting point.

[0015] Optionally, the distance calculation formula between the boundary point and the radar origin is expressed as:

[0016] d 2 =αP x 2 +βP y 2 ---(1)

[0017] In the formula: d is the calculated distance between the boundary point and the radar origin, α is the horizontal scale factor, β is the vertical scale factor, P x is the abscissa of the boundary point, P y is the ordinate of the boundary point.

[0018] Optionally, it further includes:

[0019] If all the found boundary points meet the set quantity threshold and length threshold, then proceed to the step of using the least squares method to perform quadratic curve fitting on all the found boundary points to obtain the quadratic curve coefficients of the current frame;

[0020] Otherwise, remove the current boundary starting point, and use the alternative boundary starting point that is the next one after the current boundary starting point in the sorting as the new boundary starting point.

[0021] Optionally, the step of removing the current boundary starting point and using the alternative boundary starting point that is the next one after the current boundary starting point in the sorting as the new boundary starting point includes:

[0022] Count the number of the remaining alternative boundary starting points;

[0023] If the number of the remaining alternative boundary starting points is greater than zero, use the alternative boundary starting point that is the next one after the current boundary starting point in the sorting as the new boundary starting point.

[0024] Optionally, in the radar coordinate system, obtain the initial screening range by setting a lateral threshold and a longitudinal threshold; wherein, the lateral threshold is obtained according to the road width and the radar detection capability in the real scene; the longitudinal threshold is determined according to the radar detection capability.

[0025] Optionally, use the least squares method to perform quadratic curve fitting on all the found boundary points to obtain the quadratic curve coefficients of the current frame, including:

[0026] Obtain all the boundary points found in the previous frame;

[0027] Fuse all the boundary points found in the current frame and all the boundary points found in the previous frame;

[0028] Use the least squares method to perform quadratic curve fitting on all the fused boundary points to obtain the quadratic curve coefficients of the current frame.

[0029] Optionally, the larger the ordinate of the starting point, the larger the lateral width and the longitudinal width of the corresponding preset search range.

[0030] On the other hand, the present invention also provides a road boundary detection system based on an in-vehicle millimeter-wave radar, including:

[0031] A target point acquisition module configured to acquire all the static target points of the current frame output by the in-vehicle millimeter-wave radar;

[0032] A preliminary screening module configured to screen out a plurality of stable static target points from all the static coordinate points according to the initial screening range, and the initial screening range is determined according to the radar detection capability;

[0033] An alternative starting point acquisition module configured to use the stable static target points located within the preset screening range as alternative boundary starting points; wherein, with the radar central axis as the axis of symmetry, set two search ranges within the field of view angle range of the in-vehicle millimeter-wave radar, and use the set two search ranges as the preset screening range. Each search range takes the fan-shaped radius boundary corresponding to the field of view angle range of the in-vehicle millimeter-wave radar as one boundary, and the other boundary is parallel to the fan-shaped radius boundary;

[0034] An initial starting point determination module configured to sort the alternative boundary starting points in ascending order of the distance from the radar origin, and use the alternative boundary starting point with the smallest distance as the boundary starting point;

[0035] A search module, configured to start from the boundary starting point, search for boundary points among the stable static target points within a preset search range, and use the found boundary points as new starting points to continue searching for the next boundary point until no next boundary point can be found;

[0036] A curve fitting module, configured to perform quadratic curve fitting on all the found boundary points by using the least squares method to obtain the quadratic curve coefficients of the current frame;

[0037] And a filtering processing module, configured to perform filtering processing on the quadratic curve coefficients of the current frame and the quadratic curve coefficients of the previous frame by using the Kalman filtering algorithm to obtain target quadratic curve coefficients and a target road boundary curve.

[0038] On the other hand, the present invention further provides an electronic device, including: a processor and a memory, where a computer-readable instruction is stored on the memory, and when the computer-readable instruction is executed by the processor, the road boundary detection method based on an in-vehicle millimeter-wave radar as described above is implemented.

[0039] On another hand, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the road boundary detection method based on an in-vehicle millimeter-wave radar as described above is implemented.

[0040] The road boundary detection method based on an in-vehicle millimeter-wave radar of the present invention first preliminarily screens the input static target points of the in-vehicle millimeter-wave radar to obtain relatively stable static target points, and uses the strategy of setting a search range and using the first starting point to perform clustering to find all boundary points to screen all possible road boundary points among the stable static target points. Then, combined with the boundary points screened in the previous frame of the radar, quadratic curve fitting is performed by using the least squares method. Finally, considering that the road boundary will not change suddenly in a short time, the Kalman filtering algorithm is used to perform filtering processing on the three coefficients in the obtained quadratic curve to obtain a relatively stable curve output in the time domain, thereby improving the accuracy of road boundary detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0042] Figure 1 It is a schematic flowchart of some embodiments of the road boundary detection method based on an in-vehicle millimeter-wave radar of the present invention;

[0043] Figure 2 is a schematic flowchart of some other embodiments of the road boundary detection method based on an in-vehicle millimeter-wave radar according to the present invention;

[0044] Figure 3 is a schematic flowchart of still some other embodiments of the road boundary detection method based on an in-vehicle millimeter-wave radar according to the present invention;

[0045] Figure 4 is a schematic diagram of road boundary point tracking of the road boundary detection method based on an in-vehicle millimeter-wave radar according to the present invention;

[0046] Figure 5 is a flowchart of the Kalman filtering algorithm of the road boundary detection method based on an in-vehicle millimeter-wave radar according to the present invention;

[0047] Figure 6 is a structural block diagram of some embodiments of the road boundary detection system based on an in-vehicle millimeter-wave radar according to the present invention. Specific embodiments

[0048] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0049] On the one hand, as shown in Figure 1 the embodiments of the present invention provide a road boundary detection method based on an in-vehicle millimeter-wave radar, including:

[0050] Step 100: Obtain all static target points of the current frame output by the in-vehicle millimeter-wave radar.

[0051] In this step, the in-vehicle millimeter-wave radar can be a millimeter-wave radar with a function of directly outputting static coordinate points; of course, it can also be an in-vehicle millimeter-wave radar without this function. In this case, it is necessary to first obtain the current frame radar echo signal output by the in-vehicle millimeter-wave radar, and then analyze all static target points of the current frame from the current frame radar echo signal. The static target point is the position data / coordinate data of the static target point.

[0052] Step 200: Screen out a plurality of stable static target points from all static coordinate points according to an initial screening range, and the initial screening range is determined according to the radar detection ability.

[0053] In this step, the static target points within the initial screening range are directly screened out to obtain multiple stable static target points; the initial screening range is determined according to the radar detection ability. For example: if the data reliability of the static target points detected by the vehicle-mounted millimeter-wave radar within L meters is high, then the static target points detected within L meters are screened out as stable static target points. Here, the initial screening range can be a rectangle or a triangle. When it is a rectangle, the longitudinal distance of the initial screening range is set to L meters, the transverse distance of the initial screening range is set to the preset width, and the static target points within the preset width range are screened out, without being too wide; it can also be a triangle, and only the longitudinal distance threshold is set. Generally, the preset width can cover all or most of the roads with guardrails in the real scenario.

[0054] Step 300: Use the stable static target points within the preset screening range as the alternative boundary starting points; among them, with the radar central axis as the axis of symmetry, two search ranges are set within the field of view angle of the vehicle-mounted millimeter-wave radar, and the two set search ranges are used as the preset screening range. Each search range takes the fan-shaped radius boundary corresponding to the field of view angle range of the vehicle-mounted millimeter-wave radar as one boundary, and the other boundary is parallel to the fan-shaped radius boundary.

[0055] In this step, the distance between the two boundaries of each search range should be able to meet the detection requirements of boundaries such as guardrails. The extension length of the search range in the fan-shaped radius direction is related to the boundary characteristics of various different roads, generally an empirical value. Usually, the projection length of the extension length of the search range in the fan-shaped radius direction in the radar central axis direction is less than the length corresponding to the initial screening range in the radar central axis direction. In practical applications, after obtaining the stable static target points, they can be sorted in ascending order according to the ordinate to facilitate the subsequent search for boundary points.

[0056] Step 400: Sort the alternative boundary starting points in ascending order according to the distance from the radar origin, and use the alternative boundary starting point with the smallest distance as the boundary starting point.

[0057] Before searching for boundary points, calculate the distance between each alternative boundary starting point and the radar origin, and sort the calculated distances in ascending order. Use the alternative boundary starting point with the smallest distance as the boundary starting point.

[0058] Step 500: Starting from the boundary starting point, search for boundary points from the stable static target points within the preset search range, and use the found boundary points as the new starting points to continue searching for the next boundary point until no next boundary point can be found.

[0059] In this step, first determine a boundary starting point as the starting point, and search for the next boundary starting point within a preset search range. When the next boundary starting point is found, use this boundary starting point as the new starting point to search for the next boundary point, that is, use the stable static target points found within the preset search range as boundary points to achieve the tracking of road boundary points. It should be noted that each starting point corresponds to a preset search range.

[0060] Step 600: Use the least squares method to perform quadratic curve fitting on all the boundary points found to obtain the quadratic curve coefficients of the current frame.

[0061] In this step, use the least squares method to perform quadratic curve fitting on the points obtained in step 400. The fitting formula is shown as follows:

[0062] P x = aP y 2 + bP y + c---(2)

[0063] In the formula: a is the quadratic term coefficient of the fitting function, b is the linear term coefficient of the fitting function, and c is the constant term of the fitting function. P x is the abscissa of the boundary point, and P y is the ordinate of the boundary point.

[0064] Step 700: Use the Kalman filter algorithm to perform filtering on the quadratic curve coefficients of the current frame and the quadratic curve coefficients of the previous frame to obtain the target quadratic curve coefficients and the target road boundary curve.

[0065] In this step, use the three coefficients a, b, and c obtained from the current frame and the previous frame, and use the Kalman filter algorithm to perform filtering on them to obtain a relatively stable coefficient output in the time domain.

[0066] The road boundary detection method based on in-vehicle millimeter-wave radar in the embodiment of the present invention first preliminarily screens the input in-vehicle millimeter-wave radar static target points to obtain relatively stable static target points. By setting a search range and using the first starting point for clustering to find all boundary points, all possible road boundary points are screened out among the stable static target points. Then, combined with the boundary points screened from the previous frame of the radar, the least squares method is used for quadratic curve fitting. Finally, considering that the road boundary will not change suddenly in a short time, the Kalman filter algorithm is used to filter the three coefficients in the obtained quadratic curve to obtain a relatively stable curve output in the time domain, thereby improving the accuracy of road boundary detection. The road boundary detection method based on in-vehicle millimeter-wave radar in the embodiment of the present invention has very important engineering applications in unmanned driving and assisted driving.

[0067] In some embodiments, the road boundary detection method based on in-vehicle millimeter-wave radar of the present invention further includes:

[0068] If multiple boundary points are found within a preset search range, the distance between each boundary point and the radar origin is calculated;

[0069] All boundary points are sorted in ascending order according to the calculated distances, and the boundary point corresponding to the minimum distance is used as the new starting point.

[0070] In this embodiment, generally one boundary point will be found through the preset search range. When multiple boundary points are found within the same preset search range, the distance d between each boundary point and the radar origin is calculated, and the boundary point closest to the radar origin is used as the new starting point. In addition, during the actual sorting process, it can also be set through a program, that is, the search will stop when one point is found, and then this point is used for the next step of screening.

[0071] In some embodiments, in the road boundary detection method based on in-vehicle millimeter-wave radar of the present invention, the calculation formula for the distance d between the boundary point and the radar origin is expressed as:

[0072] d 2 = αP x 2 + βP y 2 ---(1)

[0073] In the formula: d is the calculated distance between the boundary point and the radar origin, α is the horizontal scale factor, β is the vertical scale factor, the value ranges of both α and β are [0, 1], and α = 1 - β. α and β can be empirical values or calibrated values, P x is the abscissa of the boundary point, and P y is the ordinate of the boundary point.

[0074] In this embodiment, considering the characteristics of the boundary point, an influence factor is added to the calculation formula when calculating the distance, and the values of α and β can be used to adjust the influence of P x and P y on d.

[0075] In some embodiments, as shown in Figure 2 the road boundary detection method based on in-vehicle millimeter-wave radar of the present invention further includes:

[0076] Step 800: If all the found boundary points meet the set quantity threshold and length threshold, then transfer to the step of using the least squares method to perform quadratic curve fitting on all the found boundary points to obtain the quadratic curve coefficients of the current frame;

[0077] Step 900: Otherwise, remove the current boundary starting point, and use the alternative boundary starting point that is the next one in the sorting as the new boundary starting point.

[0078] In this embodiment, for all the boundary points found in step 500, it is necessary to perform a validity check, that is, to determine whether all the boundary points meet the set quantity threshold and length threshold. If they do not meet the requirements, the current boundary starting point is removed. If the found boundary points meet the requirements, then step 600 is performed. It should be noted that the set quantity threshold and length threshold here are empirical values or calibrated values. In some other embodiments, since quadratic curve fitting is used, the minimum number of boundary point data is 3. The quantity threshold can also be comprehensively determined in combination with the number of point clouds output by the millimeter-wave radar. For example, if the radar resolution is high and the number of output point clouds is large, this threshold can be appropriately increased. The setting of the length threshold is related to the radar resolution and the actual scenario. Here, the threshold is set to 8m. In some other embodiments, it is also possible to transfer to step 600 when all the found boundary points meet the set quantity threshold.

[0079] In some embodiments, as shown in Figure 2 Step 900 in the road boundary detection method based on in-vehicle millimeter-wave radar of the present invention includes:

[0080] Step 901: Count the number of remaining alternative boundary starting points;

[0081] Step 902: If the number of remaining alternative boundary starting points is greater than zero, use the alternative boundary starting point that is the next one in the sorting as the new boundary starting point.

[0082] In this embodiment, if all the found boundary points do not meet the set quantity threshold and length threshold, the current boundary starting point is removed. Determine whether the remaining possible alternative boundary starting points obtained in step 400 are empty. If they are not empty, select the next nearest point as the boundary starting point and return to step 500. If they are empty, output empty, and then directly transfer to step 100 to process the static target points output by the in-vehicle millimeter-wave radar in the next frame.

[0083] Optionally, in the road boundary detection method based on an in-vehicle millimeter-wave radar according to an embodiment of the present invention, in the radar coordinate system, an initial screening range is obtained by setting a lateral threshold and a longitudinal threshold. Among them, the lateral threshold is obtained according to the road width and the radar detection ability in the real scene; the longitudinal threshold is determined according to the radar detection ability. That is, static target points corresponding to the radar detection ability are screened out to obtain reliable static target points as stable static target points. It should be noted that in this embodiment, the lateral threshold should be able to cover all or most of the roads with boundaries such as guardrails in the real scene, and the longitudinal threshold should be able to cover the distance range that the radar can accurately detect.

[0084] In some embodiments, referring to Figure 3 as shown, step 600 in the road boundary detection method based on an in-vehicle millimeter-wave radar of the present invention includes:

[0085] Step 601: Obtain all boundary points found in the previous frame;

[0086] Step 602: Fuse all boundary points found in the current frame and all boundary points found in the previous frame;

[0087] Step 603: Use the least squares method to perform quadratic curve fitting on all the fused boundary points to obtain the quadratic curve coefficients of the current frame.

[0088] In this embodiment, each static target point has an ID or a mark, and the static target points with duplicate IDs or marks are removed through fusion.

[0089] Optionally, in the road boundary detection method based on an in-vehicle millimeter-wave radar according to an embodiment of the present invention, the larger the ordinate of the starting point, the larger the lateral width and the longitudinal width of the corresponding preset search range. It should be noted that the corresponding relationship between the ordinate and the lateral width and the longitudinal width of the preset search range can be obtained according to experience or calibration. In this embodiment, the preset search range is a threshold box, and the starting point is the boundary starting point. A threshold box is set on its left and right and upward, and the next point is searched within the box.

[0090] The implementation process of the road boundary detection method based on an in-vehicle millimeter-wave radar is specifically described below. Refer to Figure 4 as shown:

[0091] 1) Obtain all static target points of the current frame of the radar;

[0092] 2) In the radar coordinate system, by setting the horizontal and vertical thresholds, an initial screening range is obtained to preliminarily screen all input static target points and obtain stable static target points. Among them, the horizontal and vertical thresholds are determined according to indicators such as the road width and radar detection ability in the real scene. For example, the horizontal threshold should cover all or most of the roads with boundaries such as guardrails in the real scene, and the vertical threshold should cover the distance range that the radar can accurately detect. After obtaining the stable static target points, they are sorted according to the magnitude of the ordinate.

[0093] 3) Determine the area within the field of view (FOV) of the radar within the initial screening range. Set a search range on both sides of this area with the radar central axis as the symmetry axis. Define the static target points within the search range as the candidate boundary starting points of the road. When the candidate boundary starting points are not empty, perform step 4). Among them, each search range takes the sector radius boundary corresponding to the field of view range as one boundary, and the other boundary is parallel to the sector radius boundary. The distance between the two boundaries should meet the detection requirements for boundaries such as guardrails.

[0094] 4) Calculate the distance from all obtained candidate boundary starting points to the radar origin. Due to the characteristics of the boundary starting points, an influence factor is added in this invention when calculating the distance, and its calculation formula is shown in Equation (1):

[0095] d 2 =αP x 2 +βP y 2 ---(1)

[0096] In the formula: d is the calculated distance from the candidate boundary starting point to the radar origin, α is the horizontal scale factor, β is the vertical scale factor, P x is the abscissa of the candidate boundary starting point, P y is the ordinate of the candidate boundary starting point. Therefore, the values of α and β can be used to adjust the influence of P x and P y on d. Take the minimum value of the distances d of all candidate boundary starting points as the boundary starting point.

[0097] 5) Use the obtained boundary starting point to search for the next boundary point among the stable static target points within a certain range in its horizontal and vertical directions. Then take the found boundary point as the new starting point and further search for the next boundary point, and so on. When no next boundary point can be found, enter step 6).

[0098] 6) Determine the validity of the boundary points found in step 5, and check whether the number and length of the boundary points found in step 5) meet the set threshold; if not, remove the current alternative boundary starting point, and check whether the remaining possible alternative boundary starting points obtained in step (3) are empty. If not, select the next nearest alternative boundary starting point as the boundary starting point, and go back to step 5). If it is empty, output as empty. If the found boundary points meet the requirements, proceed to step 7);

[0099] 7) Use the least squares method to perform quadratic curve fitting on the points obtained in step 6), and the fitting formula is shown in formula (2):

[0100] P x =aP y 2 +bP y +c---(2)

[0101] where a is the quadratic term coefficient of the fitting function, b is the linear term coefficient of the fitting function, and c is the constant term of the fitting function; P x is the abscissa of the boundary point, and P y is the ordinate of the boundary point.

[0102] 8) Then use the three coefficients a, b, and c obtained from the current frame and the previous frame to perform filtering processing on them using the idea of Kalman filtering to obtain a relatively stable coefficient output in the time domain.

[0103] See Figure 5 As shown, taking a certain side boundary of the road as an example, the principle of Kalman filtering is explained. Let the quadratic curve fitting the road boundary at time k and time k - 1 be shown in formula (3):

[0104]

[0105] where a k , b k , c k and a k-1 , b k-1 , c k-1 are the polynomial coefficients obtained by fitting at time k and time k - 1 respectively, P k x , P k -1 x are the abscissas of the boundary points at time k and time k - 1 respectively, and P k y , P k-1 y are the ordinates of the boundary points at time k and time k - 1 respectively. Therefore, the state vector x can be set as [a b c] T, since the on-vehicle millimeter-wave radar can measure the coordinate information of road boundary points, the measurement vector can be set as the abscissa values of all boundary points obtained by clustering each frame, and n is the number of boundary points. After obtaining the two state vectors and the measurement vector, the state equation and the observation equation are shown in equations (4) and (5) respectively:

[0106] x k = Ax k-1 + w k-1 ---(4)

[0107] z k = Hx k + v k ---(5)

[0108] In the formula, A is the state transition matrix, W k-1 is the covariance matrix of the prediction noise, H is the observation matrix, V k is the covariance matrix of the measurement noise, X k and X k-1 are the state vectors at time k and time k-1 respectively, Z k and Z k-1 are the measurement vectors at time k and time k-1 respectively. During the driving of the vehicle, since there will be a certain rotation angle θ (positive counterclockwise and negative clockwise) between the radar coordinate systems at time k and time k-1, a coordinate system rotation transformation is required to obtain the curve equation of time k-1 in the coordinate system of time k. Here x k-1 We take the state vector after coordinate transformation. Therefore, the state transition matrix A is the identity matrix. The observation matrix can be obtained from the observation vector as shown in equation (6):

[0109]

[0110] In the formula, P y is the longitudinal coordinate of all boundary points obtained by clustering the boundary points of each frame. Thus, the parameters necessary for Kalman filtering are obtained, and the specific process is as Figure 5 shown. In the figure is the prior estimate of the state vector at time k, P k|k-1 is the covariance matrix of the prior estimate at time k, P k-1 is the covariance matrix of the posterior estimate error at time k-1, Q is the covariance matrix of the state estimate noise, and R is the covariance matrix of the observation noise. The detailed steps of Kalman filtering are:

[0111] Judge whether the state vector at the previous moment (k-1) is empty. If it is empty, then judge whether the boundary points selected at the current moment (k) meet the threshold. If they do not meet, directly end. Otherwise, perform quadratic function fitting and output the features;

[0112] If the state vector at time (k-1) is not empty, then determine whether the boundary point selected at time (k) meets the threshold. If it does not meet the threshold, directly end; otherwise, proceed to step 4).

[0113] Perform the Kalman filter prediction process, as shown in equations (7) and (8):

[0114]

[0115] P k|k-1 = AP k-1 A T + Q ---(8)

[0116] Generate the observation matrix H and the state vector z at the current time using the boundary points selected at the current time k ;

[0117] The correction stage of the Kalman filter, the formulas are as shown in equations (9), (10) and (11):

[0118] K k = P k|k-1 H T (HP k|k-1 H T + R) -1 ---(9)

[0119]

[0120] P k = (I - K k H)P k|k-1 ---(11)

[0121] In the formula, K k is the Kalman gain, is the optimal estimated value obtained by the Kalman filter, P k is the covariance matrix of the posterior estimation error at time k, and I is the identity matrix

[0122] Finally, output the optimal estimated value obtained by the Kalman filter

[0123] On the other hand, as shown in Figure 6 this embodiment of the present invention also provides a road boundary detection system 1 based on a vehicle-mounted millimeter-wave radar, including:

[0124] A target point acquisition module 10, configured to acquire all static target points of the current frame output by the vehicle-mounted millimeter-wave radar

[0125] The preliminary screening module 20 is configured to screen out multiple stable static target points from all static coordinate points according to an initial screening range, and the initial screening range is determined according to the radar detection ability;

[0126] The alternative starting point acquisition module 30 is configured to use the stable static target points located within a preset screening range as alternative boundary starting points; specifically, with the radar central axis as the axis of symmetry, two search ranges are set within the field of view angle range of the vehicle-mounted millimeter-wave radar, and the two set search ranges are used as the preset screening range. Each search range takes the sector radius boundary corresponding to the field of view angle range of the vehicle-mounted millimeter-wave radar as one boundary, and the other boundary is parallel to the sector radius boundary;

[0127] The initial starting point determination module 40 is configured to sort the alternative boundary starting points in ascending order of the distance from the radar origin, and use the alternative boundary starting point with the smallest distance as the boundary starting point;

[0128] The search module 50 is configured to use the boundary starting point as the starting point to search for boundary points from the stable static target points within the preset search range, and use the found boundary points as new starting points to continue searching for the next boundary point until no next boundary point can be found;

[0129] The curve fitting module 60 is configured to perform quadratic curve fitting on all the boundary points obtained by searching using the least squares method to obtain the quadratic curve coefficients of the current frame;

[0130] And, the filtering processing module 70 is configured to perform filtering processing on the quadratic curve coefficients of the current frame and the quadratic curve coefficients of the previous frame using the Kalman filtering algorithm to obtain the target quadratic curve coefficients and the target road boundary curve.

[0131] The specific details of each module of the above road boundary detection system based on the vehicle-mounted millimeter-wave radar have been described in detail in the corresponding road boundary detection method based on the vehicle-mounted millimeter-wave radar, so they will not be elaborated here.

[0132] On the other hand, an embodiment of the present invention further provides an electronic device, including: a processor and a memory, and a computer-readable instruction is stored on the memory. When the computer-readable instruction is executed by the processor, the road boundary detection method based on the vehicle-mounted millimeter-wave radar as described in the above embodiment is implemented.

[0133] Specifically, the above memory and processor can be general memory and processor, and no specific limitation is made here. When the processor runs the computer-readable instruction stored in the memory, it can execute the road boundary detection method based on the vehicle-mounted millimeter-wave radar as described in the above embodiment.

[0134] In another aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the method for detecting road boundaries based on an in-vehicle millimeter-wave radar as described in the above embodiments.

[0135] Those of ordinary skill 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 relevant hardware through a program. This program can be stored in a computer-readable storage medium, and the storage medium can include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, an optical disk, etc.

[0136] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A road boundary detection method based on in-vehicle millimeter-wave radar, characterized in that, Comprising: Obtain all static target points of the current frame output by the vehicle-mounted millimeter-wave radar; Filter out multiple stable static target points from all the static target points according to an initial screening range, where the initial screening range is determined according to the radar detection ability; Take the stable static target points located within a preset screening range as candidate boundary starting points; wherein, with the radar central axis as the axis of symmetry, two search ranges are set within the field of view angle range of the vehicle-mounted millimeter-wave radar, and the two set search ranges are used as the preset screening range. Each search range takes the fan-shaped radius boundary corresponding to the field of view angle range of the vehicle-mounted millimeter-wave radar as one boundary, and the other boundary is parallel to the fan-shaped radius boundary; Sort the candidate boundary starting points in ascending order of the distance from the radar origin, and take the candidate boundary starting point with the smallest distance as the boundary starting point; Taking the boundary starting point as the starting point, search for boundary points from the stable static target points within a preset search range, and take the found boundary points as new starting points to continue searching for the next boundary point until no next boundary point can be found; Use the least squares method to perform quadratic curve fitting on all the found boundary points to obtain the quadratic curve coefficients of the current frame; Use the Kalman filter algorithm to filter the quadratic curve coefficients of the current frame and the quadratic curve coefficients of the previous frame to obtain the target quadratic curve coefficients and the target road boundary curve.

2. The road boundary detection method based on in-vehicle millimeter-wave radar according to claim 1, wherein Also comprising: If multiple boundary points are found within a preset search range, then calculate the distance between each boundary point and the radar origin; Sort all the boundary points in ascending order of the obtained distances, and take the boundary point corresponding to the minimum distance as the new starting point.

3. The road boundary detection method based on in-vehicle millimeter-wave radar according to claim 2, characterized in that, The distance calculation formula between the boundary point and the radar origin is expressed as: d 2 = αP x 2 + βP y 2 ---(1) Where: d is the distance between the calculated boundary point and the radar origin, α is the horizontal scale factor, β is the vertical scale factor, P x is the abscissa of the boundary point, and P y is the ordinate of the boundary point.

4. The road boundary detection method based on in-vehicle millimeter-wave radar according to claim 1, characterized in that Also comprising: If all the found boundary points meet the set quantity threshold and length threshold, then transfer to the step of using the least squares method to perform quadratic curve fitting on all the found boundary points to obtain the quadratic curve coefficients of the current frame; Otherwise, remove the current boundary starting point, and take the candidate boundary starting point that is the next one in the sorting after the current boundary starting point as the new boundary starting point.

5. The method for detecting road boundaries based on in-vehicle millimeter-wave radar according to claim 4, characterized in that, The step of removing the current boundary starting point and taking the candidate boundary starting point that is the next one in the sorting after the current boundary starting point as the new boundary starting point includes: Count the number of the remaining candidate boundary starting points; If the number of the remaining candidate boundary starting points is greater than zero, then take the candidate boundary starting point that is the next one in the sorting after the current boundary starting point as the new boundary starting point.

6. The road boundary detection method based on in-vehicle millimeter-wave radar according to claim 1, wherein In the radar coordinate system, obtain the initial screening range by setting a lateral threshold and a longitudinal threshold; wherein, the lateral threshold is obtained according to the road width and the radar detection ability in the real scene; the longitudinal threshold is determined according to the radar detection ability.

7. The method for detecting road boundaries based on in-vehicle millimeter-wave radar according to claim 1, wherein Using the least squares method to perform quadratic curve fitting on all the found boundary points to obtain the quadratic curve coefficients of the current frame, includes: Obtain all the boundary points found in the previous frame; Fuse all the boundary points obtained from the current frame search and all the boundary points obtained from the previous frame search; Use the least squares method to perform quadratic curve fitting on all the fused boundary points to obtain the quadratic curve coefficients of the current frame.

8. The road boundary detection method based on in-vehicle millimeter wave radar according to any one of claims 1-7, characterized in that, The greater the ordinate of the starting point, the greater the horizontal width and vertical width of the corresponding preset search range.

9. A road boundary detection system based on in-vehicle millimeter-wave radar, characterized in that, Include: A target point acquisition module configured to acquire all the static target points of the current frame output by the vehicle-mounted millimeter-wave radar; A preliminary screening module configured to screen out a plurality of stable static target points from all the static target points according to an initial screening range, where the initial screening range is determined according to the radar detection capability; An alternative starting point acquisition module configured to use the stable static target points located within a preset screening range as alternative boundary starting points; wherein, with the radar central axis as the axis of symmetry, two search ranges are set within the field of view angle range of the vehicle-mounted millimeter-wave radar, and the two set search ranges are used as the preset screening range. Each search range takes the sector radius boundary corresponding to the field of view angle range of the vehicle-mounted millimeter-wave radar as one boundary, and the other boundary is parallel to the sector radius boundary; An initial starting point determination module configured to sort the alternative boundary starting points in ascending order of the distance from the radar origin, and use the alternative boundary starting point with the smallest distance as the boundary starting point; A search module configured to start from the boundary starting point, search for boundary points from the stable static target points within a preset search range, and use the found boundary points as new starting points to continue searching for the next boundary point until no next boundary point can be found; A curve fitting module configured to use the least squares method to perform quadratic curve fitting on all the boundary points obtained from the search to obtain the quadratic curve coefficients of the current frame; And a filtering processing module configured to use the Kalman filtering algorithm to perform filtering processing on the quadratic curve coefficients of the current frame and the quadratic curve coefficients of the previous frame to obtain the target quadratic curve coefficients and the target road boundary curve.

10. An electronic device, characterized in that, Include: A processor and a memory, where computer-readable instructions are stored on the memory, and when the computer-readable instructions are executed by the processor, the road boundary detection method based on a vehicle-mounted millimeter-wave radar according to any one of claims 1 to 8 is implemented.

Images