A Three-Radar Measurement Method for Vehicle Outer Contour Information

Abstract

The present invention provides a three - radar measurement method for vehicle outer contour information, including: radar calibration, calibrating the length - measuring lidar, the first width - and - height - measuring lidar, and the second width - and - height - measuring lidar, and initializing the parameters of the length - measuring lidar, the first width - and - height - measuring lidar, and the second width - and - height - measuring lidar; performing length - measuring target detection and width - and - height - measuring target detection, and for vehicles that meet the requirements, calculating the current length, height, and width of the vehicle; continuously tracking the target, recording the length, height, and width of the vehicles that meet the requirements, and through continuous multi - frame tracking, calculating the final length, height, and width of the vehicle, which are the contour data of the vehicle; the three - radar measurement method for vehicle outer contour information provided by the present invention uses three lidars to measure the vehicle outer contour; filters through background data, greatly improving the data validity; and uses multi - frame data for target tracking to further ensure the data validity.

Description

Technical Field

[0001] The present invention belongs to the technical field of radar detection, and particularly relates to a three-radar measurement method for vehicle outer contour information. Background Art

[0002] In the prior art, the dynamic overloading and over-limit detection devices at each toll station can only detect the axle load mass, and cannot identify some geometric parameters of the vehicle, such as the size of the vehicle. These geometric parameters are of great significance for traffic safety and effective control of over-limit operation. Some toll stations directly use a tape measure to conduct contact measurement on the vehicle. Such a measurement method has low measurement accuracy, poor measurement efficiency, and increases the workload and labor time of the operators. In some prior art research on using lidar to detect size, the data frame transmission situation in the actual detection environment, as well as the problems of target detection filtering and target tracking using data frames, are not considered.

[0003] For example, the Chinese invention patent with the publication number CN102679889B discloses a "measurement method and device for vehicle size". Among them, the measurement method for vehicle size includes: scanning a plane perpendicular to the vehicle driving direction with a lidar to measure the width and height of the vehicle; and scanning a plane parallel to the vehicle driving direction and perpendicular to the horizontal plane with a lidar to measure the length of the vehicle.

[0004] Chinese invention patent CN108592801A;

[0005] Chinese invention patent CN104655249A;

[0006] Chinese invention patent CN110728747A;

[0007] It can be seen that the prior art mainly measures the width and height of the vehicle by scanning a plane perpendicular to the vehicle driving direction with a lidar, and measures the length of the vehicle by scanning a plane parallel to the vehicle driving direction and perpendicular to the horizontal plane with a lidar. Its measurement model lacks support for data frames and cannot solve the problems of target detection and tracking through data frames. Summary of the Invention

[0008] The purpose of the present invention is to provide a three-radar measurement method for vehicle outer contour information, aiming to solve the problem that the prior art cannot use frame difference data to track the target when measuring the vehicle outer contour with two lidars.

[0009] To achieve the above technical purpose and reach the above technical effect, the present invention is realized through the following technical solutions:

[0010] The present invention provides a three-radar measurement method for vehicle outer contour information, including the following steps:

[0011] Step S1: Radar calibration. Calibrate the length-measuring lidar, the first width-height measuring lidar, and the second width-height measuring lidar. Initialize the ground clearance H1, angular resolution ω1, radar point count P1, sampling period T1, and the angle α between the first scan line and the ground of the length-measuring lidar; the ground clearance H2, angular resolution ω2, radar point count P2, sampling period T2, and the angle β between the first scan line and the ground of the first width-height measuring lidar; the ground clearance H3, angular resolution ω3, radar point count P3, sampling period T3, and the angle γ between the first scan line and the ground of the second width-height measuring lidar; the distance D between the first width-height measuring lidar and the second width-height measuring lidar L ; the distances D from the length-measuring lidar to the centers of the first width-height measuring lidar and the second width-height measuring lidar d ;

[0012] Take the position of the length-measuring lidar relative to the ground as the first coordinate zero point, and the position of the first width-height measuring lidar relative to the ground as the second coordinate zero point; record the initial length and height data {D 1(0)} of the length-measuring lidar, and the initial first width-height data {W 1(0)} of the first width-height measuring lidar, and the initial second width-height data {E 1(0)} of the second width-height measuring lidar;

[0013] Step S2: Width-height target detection. Take the initial first width-height data {W 1(0)} as the width-height background data, and obtain one frame of real-time width-height detection data {W 1(0+a*T2)} at each sampling period T2; compare the real-time width-height detection data with the width-height background data, and calculate the change rate μ of the width-height measuring lidar according to the following formula 22 :

[0014] μ 22 =(W 1(0+a*T2) –W 1(0) ) / W 1(0) ;

[0015] In the formula, W 1(0+a*T2) is the radar measurement value of the first width-height measuring lidar at the a-th frame of data (the a-th sampling period); W 1(0) is the initial width-height data of the first width-height measuring lidar;

[0016] When the change rate μ 22 of the first width-height measuring lidar at the a-th frame of data is less than 2%, update the width-height background data, and take the current real-time width-height detection data {W 1(0+a*T2)} as the new width-height background data; when the change rate μ 22 of the first width-height measuring lidar at the j-th frame of data is greater than or equal to 2%, it is determined that a vehicle has entered the detection range;

[0017] Calculate the second abscissa group of the vehicle measured by the first width and height measuring lidar relative to the second coordinate zero point according to the following formula:

[0018] X 2j =(W 1(0+j*T2) -W 1(0+a*T2) ) * cos(ω2 * P2 + β);

[0019] In the formula, X 2j is the second abscissa of the vehicle measured by the first width and height measuring lidar relative to the second coordinate zero point; W 1(0+j*T2) is the radar measurement value of the first width and height measuring lidar at the j-th frame of data (the j-th sampling period); W 1(0+a*T2) is the width and height background data of the first width and height measuring lidar;

[0020] Calculate the second ordinate group of the vehicle measured by the first width and height measuring lidar relative to the second coordinate zero point according to the following formula:

[0021] Y 2j =H2 - (W 1(0+j*T2) -W 1(0+a*T2) ) * sin(ω2 * P2 + β);

[0022] In the formula, Y 2j is the second ordinate of the vehicle measured by the first length measuring lidar relative to the second coordinate zero point; W 1(0+j*T2) is the radar measurement value of the first width and height measuring lidar at the j-th frame of data (the j-th sampling period); W 1(0+a*T2) is the width and height background data of the first width and height measuring lidar;

[0023] Calculate the second abscissa group of the vehicle measured by the second width and height measuring lidar relative to the second coordinate zero point according to the following formula:

[0024] X 3j =(E 1(0+j*T3) -E 1(0+a*T3) ) * cos(ω3 * P3 + γ) - D L ;

[0025] In the formula, X 3j is the second abscissa of the vehicle measured by the second width and height measuring lidar relative to the second coordinate zero point; E 1(0+j*T3) is the radar measurement value of the second width and height measuring lidar at the j-th frame of data (the j-th sampling period), E 1(0+a*T3) is the width and height background data of the second width and height measuring lidar, D L is the distance between the first width and height measuring lidar and the second width and height measuring lidar;

[0026] Calculate the second ordinate group of the vehicle measured by the second width and height measuring lidar relative to the second coordinate zero point according to the following formula:

[0027] Y 3j =H3-(E 1(0+j*T3) -E 1(0+a*T3) )*sin(ω3*P3+γ);

[0028] In the formula, Y 3j is the second vertical coordinate of the vehicle measured by the second length measuring lidar relative to the second coordinate zero point; E 1(0+j*T3) is the radar measurement value of the second width and height measuring lidar at the j-th frame of data (the j-th sampling period), and E 1(0+a*T2) is the width and height background data of the second width and height measuring lidar;

[0029] The length W of the vehicle at the j-th frame of data j is the difference between the maximum value and the minimum value of the second abscissa group X 2j and X 3j :

[0030] W j =max{X 21 ,X 22 ,…,X 2j ,X 31 ,X 32 ,…,X 3j}-min{X 21 ,X 22 ,…X 2j ,X 31 ,X 32 ,…,X 3j};

[0031] The second height K of the vehicle at the j-th frame of data j is the maximum value of the second vertical coordinate group Y 2j and Y 3j :

[0032] K j =max{Y 21 ,Y 22 ,…,Y 2j ,Y 31 ,Y 32 …,Y 3j};

[0033] Step S3: Length measurement target detection. Use the initial length and height data {D 1(0)} as the length and height background data, and obtain one frame of real-time length and height detection data {D 1(0+a*T1)} at each sampling period T1; Compare the real-time length and height detection data with the length and height background data, and calculate the length change rate μ of the length measuring lidar according to the following formula 11 :

[0034] μ11 =(D 1(0+a*T1) –D 1(0) ) / D 1(0) ;

[0035] Where D 1(0+a*T1) is the radar measurement value of the length measurement lidar at the a-th frame of data; D 1(0) is the initial height and length data of the length measurement lidar;

[0036] When the change rate μ 11 of the length measurement lidar at the a-th frame of data is less than 2%, update the height and length background data, and use the current real-time height and length detection data D 1(0+a*T1) as the new height and length background data; when the length change rate μ 11 of the length measurement lidar at the i-th frame of data is greater than or equal to 2%, it is determined that a vehicle has entered the detection range, and the first abscissa group of the vehicle measured by the length measurement lidar relative to the first coordinate zero point is calculated according to the following formula:

[0037] X 1i =D d -(D 1(0+i*T1) -D 1(0+a*T1) )*cos(ω1*P1 + α);

[0038] Where X 1i is the first abscissa of the vehicle measured by the length measurement lidar relative to the first coordinate zero point; D 1(0+i*T1) is the radar measurement value of the vehicle measured by the length measurement lidar at the i-th frame of data (the i-th sampling period); D 1(0+a*T1) is the height and length background data of the length measurement lidar; D d is the distance between the length measurement lidar and the centers of the first width and height measurement lidar and the second width and height measurement lidar;

[0039] The length L i of the vehicle at the i-th frame of data 1i is the maximum value in the first abscissa group X

[0040] L i = max{X 11 , X 12 , X 13 , …, X 1n};

[0041] The first ordinate group of the vehicle measured by the length measurement lidar relative to the first coordinate zero point is calculated according to the following formula:

[0042] Y 1i = H1 - (D 1(0+i*T1) - D 1(0+a*T1) )*sin(ω1*P1 + α),

[0043] Wherein, Y 1i is the first vertical coordinate of the vehicle measured by the length-measuring lidar relative to the first coordinate zero point; D 1(0+i*T1) is the radar measurement value of the vehicle measured by the length-measuring lidar at the i-th frame of data; D 1(0+a*T1) is the height and length background data of the length-measuring lidar;

[0044] The first height M of the vehicle at the i-th frame of data i is the maximum value in the first vertical coordinate group Y 1i :

[0045] M i = max{Y 11 , Y 12 , Y 13 , …, Y 1i};

[0046] Step S4: Width and height target tracking. Set the width threshold δ, and judge the width W of the vehicle at the j-th frame of data j ; If the width W of the vehicle at the j-th frame of data j is less than the width threshold δ, then discard the target and return to step S2; If the width W of the vehicle at the j-th frame of data j is greater than or equal to the width threshold ε, then start tracking from the j-th frame of data. When the change rate μ 22 of the width and height measuring lidar is continuously greater than or equal to 2% for r consecutive frames of data, record r consecutive frames of data until the change rate μ 22 of the width and height measuring lidar is less than 2%, indicating that the vehicle has left, the recording ends, and the vehicle width W = max{W j , W j+1 , …, W j+r}, the second height K of the vehicle = max{K j , K j+1 , …, K j+r}; the vehicle length L = max{L i , L i+1 , …, L i+r}, the first height M of the vehicle = max{M i , M i+1 , …, M i+r};

[0047] Step S5: Vehicle information matching. Set the matching threshold ζ. If the difference |M - K| between the first height M of the vehicle and the second height K of the vehicle is greater than ζ, the matching fails and return to step S2; If the difference |M - K| between the first height M of the vehicle and the second height K of the vehicle is less than or equal to ζ, the matching is successful, the second height K of the vehicle is used as the final height of the vehicle, the vehicle length L is used as the final length of the vehicle, and the vehicle width W is used as the final width of the vehicle, and the process ends.

[0048] Advantages of the present invention:

[0049] The three - radar measurement method for vehicle outer contour information provided by the present invention uses three lidar to measure the vehicle outer contour; filters through background data, greatly improving the data validity; and uses multi - frame data for target tracking to further ensure the data validity. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic structural diagram of the arrangement of the first width - and - height - measuring lidar and the second width - and - height - measuring lidar according to the present invention;

[0051] Figure 2 It is a schematic diagram of the spacing calibration between the length - measuring lidar and the width - and - height - measuring lidar according to the present invention;

[0052] Figure 3 It is a schematic diagram of the lidar calibration of the length - measuring lidar according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below through specific embodiments with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0054] In an embodiment of the present invention, a three - radar measurement method for vehicle outer contour information is provided. As Figure 1 、 Figure 3 shown, a length - measuring lidar 200 is arranged on the length - measuring gantry 101, a first width - and - height - measuring lidar 300 and a second width - and - height - measuring lidar 400 are arranged on the width - and - height - measuring gantry 100. The length - measuring lidar 200 is used to measure the length of the vehicle 500 along the driving direction of the vehicle 500, and the first width - and - height - measuring lidar 300 and the second width - and - height - measuring lidar 400 are used to measure the width and height of the vehicle 500 in a direction perpendicular to the driving direction of the vehicle 500; specifically, it includes the following steps:

[0055] Step S1: Lidar calibration. Calibrate the length - measuring lidar 200, the first width - and - height - measuring lidar 300 and the second width - and - height - measuring lidar 400, and initialize the ground - clearance height H1, angular resolution ω1, radar point count P1, sampling period T1 of the length - measuring lidar 200, and the angle α between the first scan line and the ground, the ground - clearance height H2, angular resolution ω2, radar point count P2, sampling period T2 of the first width - and - height - measuring lidar 300, and the angle β between the first scan line and the ground; the ground - clearance height H3, angular resolution ω3, radar point count P3, sampling period T3 of the second width - and - height - measuring lidar 400, and the angle γ between the first scan line and the ground; the spacing D between the first width - and - height - measuring lidar 300 and the second width - and - height - measuring lidar 400 L; The distance D between the length measuring lidar 200 and the centers of the first width and height measuring lidar 300 and the second width and height measuring lidar 400 d (as Figure 2 shown);

[0056] The position of the length measuring lidar 200 relative to the ground is used as the first coordinate zero point, and the position of the first width and height measuring lidar 300 relative to the ground is used as the second coordinate zero point; Record the initial length and height data {D 1(0)} of the length measuring lidar 200, and the first width and height initial data {W 1(0)} of the first width and height measuring lidar 300, and the second width and height initial data {E 1(0)} of the second width and height measuring lidar 400;

[0057] Step S2: Width and height target detection. Use the first width and height initial data {W 1(0)} as the width and height background data, and obtain a frame of real-time width and height detection data {W 1(0+a*T2)} in each sampling period T2; Compare the real-time width and height detection data with the width and height background data, and calculate the change rate μ 22 of the width and height measuring lidar according to the following formula:

[0058] μ 22 = (W 1(0+a*T2) – W 1(0) ) / W 1(0) ;

[0059] In the formula, W 1(0+a*T2) is the radar measurement value of the first width and height measuring lidar at the a-th frame of data (the a-th sampling period); W 1(0) is the width and height initial data of the first width and height measuring lidar;

[0060] When the change rate μ 22 of the first width and height measuring lidar at the a-th frame of data is less than 2%, update the width and height background data, and use the current real-time width and height detection data {W 1(0+a*T2)} as the new width and height background data; When the degree change rate μ 22 of the first width and height measuring lidar at the j-th frame of data is greater than or equal to 2%, it is determined that a vehicle has entered the detection range;

[0061] Calculate the second abscissa group of the vehicle measured by the first width and height measuring lidar relative to the second coordinate zero point according to the following formula:

[0062] X 2j = (W 1(0+j*T2) - W 1(0+a*T2) ) * cos(ω2 * P2 + β);

[0063] In the formula, X 2jThe second abscissa of the vehicle measured by the first width and height measuring lidar relative to the second coordinate zero point; W 1(0+j*T2) The radar measurement value of the first width and height measuring lidar at the j-th frame of data (the j-th sampling period); W 1(0+a*T2) The width and height background data of the first width and height measuring lidar;

[0064] Calculate the second ordinate group of the vehicle measured by the first width and height measuring lidar relative to the second coordinate zero point according to the following formula:

[0065] Y 2j = H2 - (W 1(0+j*T2) - W 1(0+a*T2) ) * sin(ω2 * P2 + β);

[0066] In the formula, Y 2j The second ordinate of the vehicle measured by the first length measuring lidar relative to the second coordinate zero point; W 1(0+j*T2) The radar measurement value of the first width and height measuring lidar at the j-th frame of data (the j-th sampling period); W 1(0+a*T2) The width and height background data of the first width and height measuring lidar;

[0067] Calculate the second abscissa group of the vehicle measured by the second width and height measuring lidar relative to the second coordinate zero point according to the following formula:

[0068] X 3j =(E 1(0+j*T3) - E 1(0+a*T3) ) * cos(ω3 * P3 + γ) - D L ;

[0069] In the formula, X 3j The second abscissa of the vehicle measured by the second width and height measuring lidar relative to the second coordinate zero point; E 1(0+j*T3) The radar measurement value of the second width and height measuring lidar at the j-th frame of data (the j-th sampling period), E 1(0+a*T3) The width and height background data of the second width and height measuring lidar, D L The distance between the first width and height measuring lidar and the second width and height measuring lidar;

[0070] Calculate the second ordinate group of the vehicle measured by the second width and height measuring lidar relative to the second coordinate zero point according to the following formula:

[0071] Y 3j = H3 - (E 1(0+j*T3) - E 1(0+a*T3) ) * sin(ω3 * P3 + γ);

[0072] In the formula, Y 3j The second ordinate of the vehicle measured by the second length measuring lidar relative to the second coordinate zero point; E 1(0+j*T3)is the radar measurement value of the second width and height lidar at the j-th frame of data (the j-th sampling period), E 1(0+a*T3) is the width and height background data of the second width and height lidar;

[0073] The length W of the vehicle at the j-th frame of data j is the second abscissa group X 2j and X 3j is the difference between the maximum and minimum values in:

[0074] W j = max{X 21 , X 22 , …, X 2j , X 31 , X 32 , …, X 3j}- min{X 21 , X 22 , … X 2j , X 31 , X 32 , …, X 3j};

[0075] The second height K of the vehicle at the j-th frame of data j is the second ordinate group Y 2j and Y 3j is the maximum value in:

[0076] K j = max{Y 21 , Y 22 , …, Y 2j , Y 31 , Y 32 …, Y 3j};

[0077] Step S3: Length measurement target detection. Use the initial length and height data {D 1(0)} as the length and height background data, and obtain one frame of real-time length and height detection data {D 1(0+a*T1)} at each sampling period T1; Compare the real-time length and height detection data with the length and height background data, and calculate the length change rate μ 11 of the length measurement lidar according to the following formula:

[0078] μ 11 = (D 1(0+a*T1) – D 1(0) ) / D 1(0) ;

[0079] In the formula, D 1(0+a*T1) is the radar measurement value of the length measurement lidar at the a-th frame of data; D 1(0) is the initial length and height data of the length measurement lidar;

[0080] When the change rate μ of the length measurement lidar at the a-th frame of data 11 is less than 2%, update the background data of vehicle height and length, and use the current real-time vehicle height and length detection data D 1(0+a*T1) as the new background data of vehicle height and length; when the length change rate μ of the length measurement lidar at the i-th frame of data 11 is greater than or equal to 2%, it is determined that a vehicle has entered the detection range, and calculate the first abscissa group of the vehicle measured by the length measurement lidar relative to the first coordinate zero point according to the following formula:

[0081] X 1i =D d -(D 1(0+i*T1) -D 1(0+a*T1) )*cos(ω1*P1 + α);

[0082] In the formula, X 1i is the first abscissa of the vehicle measured by the length measurement lidar relative to the first coordinate zero point; D 1(0+i*T1) is the radar measurement value of the vehicle measured by the length measurement lidar at the i-th frame of data (the i-th sampling period); D 1(0+a*T1) is the background data of vehicle height and length of the length measurement lidar; D d is the distance from the length measurement lidar to the center of the first width and height measurement lidar and the second width and height measurement lidar;

[0083] The length L of the vehicle at the i-th frame of data i is the maximum value in the first abscissa group X 1i :

[0084] L i =max{X 11 ,X 12 ,X 13 ,…,X 1n};

[0085] Calculate the first ordinate group of the vehicle measured by the length measurement lidar relative to the first coordinate zero point according to the following formula:

[0086] Y 1i =H1-(D 1(0+i*T1) -D 1(0+a*T1) )*sin(ω1*P1 + α),

[0087] In the formula, Y 1i is the first ordinate of the vehicle measured by the length measurement lidar relative to the first coordinate zero point; D 1(0+i*T1) is the radar measurement value of the vehicle measured by the length measurement lidar at the i-th frame of data; D 1(0+a*T1) is the background data of vehicle height and length of the length measurement lidar;

[0088] The first height M of the vehicle at the i-th frame of data iis the maximum value of the first set of vertical coordinates Y 1i :

[0089] M i = max{Y 11 , Y 12 , Y 13 , …, Y 1i};

[0090] Step S4: Width and height target tracking. Set the width threshold δ, and judge the width W of the vehicle at the j-th frame of data j ; If the width W of the vehicle at the j-th frame of data j is less than the width threshold δ, then discard the target and return to Step S2; If the width W of the vehicle at the j-th frame of data j is greater than or equal to the width threshold ε, then start tracking from the j-th frame of data. When the change rate μ 22 of the width and height lidar is continuously greater than or equal to 2% for r consecutive frames of data, record r consecutive frames of data until the change rate μ 22 of the width and height lidar is less than 2%, indicating that the vehicle has left, and the recording ends. The vehicle width W = max{W j , W j+1 , …, W j+r}, the vehicle's second height K = max{K j , K j+1 , …, K j+r}; The vehicle length L = max{L i , L i+1 , …, L i+r}, and the vehicle's first height M = max{M i , M i+1 , …, M i+r};

[0091] Step S5: Vehicle information matching. Set the matching threshold ζ. If the difference |M - K| between the vehicle's first height M and the vehicle's second height K is greater than ζ, then the matching fails and return to Step S2; If the difference |M - K| between the vehicle's first height M and the vehicle's second height K is less than or equal to ζ, then the matching is successful. The vehicle's second height K is used as the vehicle's final height, the vehicle length L is used as the vehicle's final length, and the vehicle width W is used as the vehicle's final width, and the process ends.

[0092] References to "some embodiments", "an embodiment", or "embodiments" in this specification refer to a particular feature, structure, or property described in connection with the embodiments being included in at least one embodiment. Thus, the appearances of the phrases "in some embodiments", "in an embodiment", or "in embodiments" in various places throughout this specification are not necessarily referring to the same embodiment. Additionally, the particular features, structures, or properties may be combined in any suitable manner in one or more embodiments. Accordingly, a particular feature, structure, or property shown or described in connection with one embodiment may be combined, in whole or in part, with the features, structures, or properties of one or more other embodiments without limitation, so long as the combination is not illogical or non - working. Further, the elements in the drawings of this application are for illustrative purposes only and are not drawn to scale.

[0093] Having thus described several aspects of at least one embodiment of the invention, it will be appreciated that various changes, modifications, and improvements will readily occur to those skilled in the art. Such changes, modifications, and improvements are intended to be within the spirit and scope of the invention.

Claims

1. A three-radar measurement method for vehicle outline information, characterized in that , it includes the following steps: Step S1: Radar calibration. Calibrate the length-measuring lidar, the first width-and-height-measuring lidar, and the second width-and-height-measuring lidar. Initialize the ground clearance H1, angular resolution ω1, radar point count P1, sampling period T1, and the angle α between the first scan line and the ground of the length-measuring lidar; the ground clearance H2, angular resolution ω2, radar point count P2, sampling period T2, and the angle β between the first scan line and the ground of the first width-and-height-measuring lidar; the ground clearance H3, angular resolution ω3, radar point count P3, sampling period T3, and the angle γ between the first scan line and the ground of the second width-and-height-measuring lidar; the distance D between the first width-and-height-measuring lidar and the second width-and-height-measuring lidar L ; the distances D from the length-measuring lidar to the centers of the first width-and-height-measuring lidar and the second width-and-height-measuring lidar d ; The position of the length-measuring lidar relative to the ground is used as the first coordinate zero point, and the position of the first width-and-height-measuring lidar relative to the ground is used as the second coordinate zero point; record the initial length and height data {D 1(0)} of the length-measuring lidar, and the first width-and-height initial data {W 1(0)} of the first width-and-height-measuring lidar, and the second width-and-height initial data {E 1(0)} of the second width-and-height-measuring lidar; Step S2: Width and height target detection. Use the first width and height initial data {W 1(0)} as the width and height background data, and obtain one frame of real-time width and height detection data {W 1(0+a*T2)} in each sampling period T2; compare the real-time width and height detection data with the width and height background data, and calculate the change rate μ 22 of the width and height lidar according to the following formula: μ 22 =(W 1(0+a*T2) –W 1(0) ) / W 1(0) ; Where, W 1(0+a*T2) is the radar measurement value of the first width and height measurement lidar at the a-th frame of data; W 1(0) is the initial width and height data of the first width and height measurement lidar; When the change rate μ of the first width and height detection lidar for the a-th frame of data 22 is less than 2%, update the width and height background data, and use the current real-time width and height detection data {W 1(0+a*T2)} as the new width and height background data; when the change rate μ of the first width and height detection lidar for the j-th frame of data 22 is greater than or equal to 2%, it is determined that a vehicle has entered the detection range; Calculate the second abscissa group of the vehicle measured by the first width and height lidar relative to the second coordinate zero point according to the following formula: X 2j =(W 1(0+j*T2) -W 1(0+a*T2) ) * cos(ω2 * P2 + β); where X 2j is the second abscissa of the vehicle measured by the first width and height measuring lidar relative to the second coordinate zero point; W 1(0+j*T2) is the radar measurement value of the first width and height measuring lidar at the j-th frame of data; W 1(0+a*T2) is the width and height background data of the first width and height measuring lidar; Calculate the second ordinate group of the vehicle measured by the first width and height lidar relative to the second coordinate zero point according to the following formula: Y 2j =H2 - (W 1(0+j*T2) - W 1(0+a*T2) ) * sin(ω2 * P2 + β); Where Y 2j is the second vertical coordinate of the vehicle measured by the first width and height measurement lidar relative to the second coordinate zero point; W 1(0+j*T2) is the radar measurement value of the first width and height measurement lidar at the j-th frame of data; W 1(0+a*T2) is the width and height background data of the first width and height measurement lidar; Calculate the second abscissa group of the vehicle measured by the second width and height lidar relative to the second coordinate zero point according to the following formula: X 3j =(E 1(0+j*T3) -E 1(0+a*T3) )*cos(ω3*P3+γ)-D L ; Where X 3j is the second abscissa of the vehicle measured by the second width and height lidar relative to the second coordinate zero point; E 1(0+j*T3) is the radar measurement value of the second width and height lidar at the j-th frame of data, E 1(0+a*T3) is the width and height background data of the second width and height lidar, D L is the distance between the first width and height lidar and the second width and height lidar; Calculate the second ordinate group of the vehicle measured by the second width and height lidar relative to the second coordinate zero point according to the following formula: Y 3j =H3-(E 1(0+j*T3) -E 1(0+a*T3) )*sin(ω3*P3+γ); Where Y 3j is the second ordinate of the vehicle measured by the second width and height lidar relative to the second coordinate zero point; E 1(0+j*T3) is the radar measurement value of the second width and height lidar at the j-th frame of data, and E 1(0+a*T3) is the width and height background data of the second width and height lidar; The width W of the vehicle at the j-th frame of data j is the second abscissa group X 2j and the difference between the maximum value and the minimum value in X 3j is: W j = max{X 21 , X 22 , …, X 2j , X 31 , X 32 , …, X 3j} - min{X 21 , X 22 , … X 2j , X 31 , X 32 , …, X 3j}; The second height K of the vehicle at the j-th frame data j is the second ordinate group Y 2j and Y 3j is the maximum value in: K j = max{Y 21 , Y 22 , …, Y 2j , Y 31 , Y 32 …, Y 3j}; Step S3: Length measurement target detection. Use the initial height and length data {D 1(0)} as the height and length background data, and obtain one frame of real-time height and length detection data {D 1(0+a*T1)} in each sampling period T1; compare the real-time height and length detection data with the height and length background data, and calculate the length change rate μ of the length measurement lidar according to the following formula 11 : μ 11 =(D 1(0+a*T1) –D 1(0) ) / D 1(0) ; where D 1(0+a*T1) is the radar measurement value of the length-measuring lidar at the a-th frame of data; D 1(0) is the initial height and length data of the length-measuring lidar. The change rate μ of the length measurement lidar at the a-th frame of data 11 When it is less than 2%, update the background data of vehicle height and length, and use the current real-time vehicle height and length detection data D 1(0+a*T1) as the new background data of vehicle height and length; when the length change rate μ of the length measurement lidar at the i-th frame of data 11 is greater than or equal to 2%, it is determined that a vehicle has entered the detection range, and the first abscissa group of the vehicle measured by the length measurement lidar relative to the first coordinate zero point is calculated according to the following formula: X 1i =D d -(D 1(0+i*T1) -D 1(0+a*T1) )*cos(ω1*P1+α); Wherein, X 1i is the first abscissa of the vehicle measured by the length-measuring lidar relative to the first coordinate zero point; D 1(0+i*T1) is the radar measurement value of the vehicle measured by the length-measuring lidar at the i-th frame of data; D 1(0+a*T1) is the height and length background data of the length-measuring lidar; D d is the distance between the length-measuring lidar and the centers of the first width and height measuring lidar and the second width and height measuring lidar; The length L of the vehicle at the i-th frame of data i is the maximum value in 1i the first abscissa group X: L i = max{X 11 , X 12 , X 13 , …, X 1n}; Calculate the first ordinate group of the vehicle measured by the length-measuring lidar relative to the first coordinate zero point according to the following formula: Y 1i =H1-(D 1(0+i*T1) -D 1(0+a*T1) )*sin(ω1*P1+α), Where Y 1i is the first vertical coordinate of the vehicle measured by the length-measuring lidar relative to the first coordinate zero point; D 1(0+i*T1) is the radar measurement value of the vehicle measured by the length-measuring lidar at the i-th frame of data; D 1(0+a*T1) is the height and length background data of the length-measuring lidar; The first height M of the vehicle at the i-th frame of data i is the maximum value in the first ordinate group Y 1i : M i = max{Y 11 , Y 12 , Y 13 , …, Y 1i}; Step S4: Width and height measurement for target tracking. Set the width threshold δ and judge the width W of the vehicle at the j-th frame of data. j If the width W of the vehicle at the j-th frame of data j is less than the width threshold δ, discard the target and return to step S2. The width W of the vehicle at the j-th frame data j is greater than or equal to the width threshold ε, then start tracking from the j-th frame data, and the change rate μ of the width and height measuring lidar 22 is continuously greater than or equal to 2% for r consecutive frames of data until the change rate μ of the width and height measuring lidar 22 is less than 2%, indicating that the vehicle has left and the recording ends. The vehicle width W = max{W j , W j+1 , …, W j+r}, the second height K of the vehicle = max{K j , K j+1 , …, K j+r}; the vehicle length L = max{L i , L i+1 , …, L i+r , the first height M of the vehicle = max{M i , M i+1 , …, M i+r}; Step S5: Vehicle information matching. Set the matching threshold ζ. If the difference |M - K| between the first vehicle height M and the second vehicle height K is greater than ζ, the matching fails and return to step S2; if the difference |M - K| between the first vehicle height M and the second vehicle height K is less than or equal to ζ, the matching is successful. The second vehicle height K is used as the final vehicle height, the vehicle length L is used as the final vehicle length, and the vehicle width W is used as the final vehicle width, and the process ends.

Images