A vehicle outer contour measurement method and device based on laser radar

Abstract

The application provides a vehicle contour measurement method and device based on laser radar. The method responds to a vehicle to be measured driving into a detection interval, collects first and second contour parameters of the vehicle by using first and second contour radars, collects first positioning parameters of the vehicle by using a first positioning radar, and marks a vehicle type of the vehicle to be measured according to the first, second and first positioning parameters. In the case of the vehicle to be measured being an extra-long vehicle, the second positioning parameters of the vehicle are collected by using a second positioning radar in response to the vehicle to be measured driving to a registration position. A split vehicle signal is obtained, and a contour fusion parameter set and a positioning fusion parameter set are generated. Three-dimensional point cloud data are generated based on the contour fusion parameter set and the positioning fusion parameter set, and the contour parameters of the vehicle to be measured are calculated according to the three-dimensional point cloud data. The above method can measure the contour size of an extra-long vehicle while taking into account the contour size measurement of a conventional vehicle.

Description

Technical Field

[0001] This application relates to the field of lidar technology, and in particular to a lidar method and apparatus for measuring vehicle outline. Background Technology

[0002] In the transportation industry, to reduce traffic accidents such as scratches and collisions caused by vehicles being too large on the road, it is necessary to measure the external dimensions of oversized transport vehicles. Two width and height radars are installed at the ends of the first crossbar, and a length-measuring radar is installed in the middle of the second crossbar. By using the width and height radars in conjunction with the length-measuring radar to locate the front position of the vehicle as it passes the width and height radars, the vehicle's length, width, and height can be calculated, ultimately determining the vehicle's external dimensions.

[0003] Because of the limited installation height of the length-measuring radar, the length of extra-long vehicles may exceed its measurement range, leading to inaccurate calculations of the vehicle's external dimensions. While increasing the installation distance between the first and second crossbars can enable the measurement of extra-long vehicles, a longer installation distance can prevent the radar from detecting smaller vehicles in time, thus reducing the accuracy of external dimension measurements for smaller vehicles. Therefore, the above method cannot simultaneously guarantee the accuracy of external dimension measurements for both small and extra-long vehicles. Summary of the Invention

[0004] This application provides a method and apparatus for measuring the outline of a vehicle based on lidar, which can calculate the outline dimensions of ultra-long vehicles while also taking into account the outline dimensions of conventional vehicles.

[0005] In a first aspect, this application provides a vehicle outline measurement method based on lidar, comprising: in response to a vehicle under test entering a detection zone, acquiring a first outline parameter of the vehicle under test using a first outline radar, acquiring a second outline parameter of the vehicle under test using a second outline radar, and acquiring a first positioning parameter of the vehicle under test using a first positioning radar; the detection zone includes a first crossbar, a second crossbar, and a third crossbar arranged in parallel in sequence; the first outline radar is disposed at one end of the first crossbar, the second outline radar is disposed at the other end of the first crossbar, and the first positioning radar is disposed at the middle position of the second crossbar;

[0006] The vehicle type of the vehicle under test is marked according to the first shape parameter, the second shape parameter and the first positioning parameter, and the vehicle type includes extra-long vehicles and regular vehicles;

[0007] When the vehicle type of the vehicle under test is an extra-long vehicle, in response to the vehicle under test driving to the registration position, the second positioning parameters of the vehicle under test are collected by the second positioning radar. The registration position is set between the first crossbar and the second crossbar, and the second positioning radar is set at the middle position of the third crossbar.

[0008] The vehicle separation signal is acquired, and a set of shape fusion parameters is generated based on the first shape parameter and the second shape parameter, and a set of positioning fusion parameters is generated based on the first positioning parameter and the second positioning parameter. The vehicle separation signal is used to characterize the rear of the vehicle under test passing the first crossbar.

[0009] Three-dimensional point cloud data is generated based on the shape fusion parameter set and the positioning fusion parameter set, and the outline parameters of the vehicle under test are calculated based on the three-dimensional point cloud data. The outline parameters include length, width and height.

[0010] Optionally, if the vehicle length type of the vehicle under test is a regular length vehicle, in response to the rear of the vehicle under test passing the first crossbar, a set of shape fusion parameters is generated based on the first shape parameters and the second shape parameters.

[0011] Three-dimensional point cloud data is generated based on the shape fusion parameter set and the first positioning parameters, and the outline parameters of the vehicle under test are calculated based on the three-dimensional point cloud data. The outline parameters include length, width and height.

[0012] Optionally, the first shape parameter includes first shape data and a first shape timestamp, wherein the first shape timestamp is used to characterize the acquisition time of the first shape data;

[0013] The second shape parameter includes second shape data and a second shape timestamp, wherein the second shape timestamp is used to characterize the acquisition time of the second shape data;

[0014] The first positioning parameter includes first positioning data and a first positioning timestamp. The first positioning data is the front position information of the vehicle under test collected by the first positioning radar, and the first positioning timestamp is used to characterize the time when the first positioning data was collected.

[0015] The second positioning parameter includes second positioning data and a second positioning timestamp. The second positioning data is the front position information of the vehicle under test collected by the second positioning radar, and the second positioning timestamp is used to characterize the time when the second positioning data was collected.

[0016] Optionally, the vehicle type of the vehicle under test is marked according to the first shape parameter, the second shape parameter and the first positioning parameter, including: generating a vehicle separation signal according to the first shape parameter and the second shape parameter, wherein the vehicle separation signal includes a vehicle separation timestamp, and the vehicle separation timestamp is used to characterize the time when the rear of the vehicle under test passes the first crossbar.

[0017] The registration timestamp is obtained based on the first positioning parameter, and the registration timestamp is used to characterize the time information corresponding to when the first positioning data is the registration position;

[0018] If the vehicle assignment timestamp is earlier than or equal to the registration timestamp, then the vehicle under test is marked as a regular vehicle.

[0019] If the vehicle assignment timestamp is later than the registration timestamp, the vehicle under test is marked as an oversized vehicle.

[0020] Optionally, generating a vehicle separation signal based on the first shape parameter and the second shape parameter includes: in response to the first shape radar or the second shape radar detecting the rear of the vehicle under test, generating the vehicle separation timestamp, wherein the vehicle separation timestamp is the acquisition time of the first shape parameter or the second shape parameter;

[0021] The vehicle allocation signal is generated based on the vehicle allocation timestamp.

[0022] Optionally, generating a shape fusion parameter set based on the first shape parameter and the second shape parameter includes: obtaining a first shape parameter set, the first shape parameter set including a first number of first shape data and a first number of first shape timestamps, the first shape data and the first shape timestamps having a correlation relationship;

[0023] Obtain a second set of shape parameters, which includes a second number of second shape data and a second number of second shape timestamps, and the second shape data and the second shape timestamps are related.

[0024] Sort the first number of first shape timestamps and the second number of second shape timestamps in chronological order from early to late to generate a first time series;

[0025] The first set of shape parameters and the second set of shape parameters are fused according to the first time series to generate the shape fusion parameter set.

[0026] Optionally, the vehicle outline measurement method based on lidar responds to the second positioning radar detecting that the vehicle under test has traveled to the position of the second crossbar, and stops the first positioning radar from collecting the first positioning parameters of the vehicle under test. The step of generating a positioning fusion parameter set based on the first positioning parameters and the second positioning parameters includes: obtaining a first positioning parameter set, the first positioning parameter set including a third number of first positioning data and a third number of first positioning timestamps, the first positioning data and the first positioning timestamps having a correlation relationship.

[0027] Obtain a second set of positioning parameters, which includes a fourth number of second positioning data and a fourth number of second positioning timestamps, and the second positioning data and the second positioning timestamps are related.

[0028] Sort the third number of first location timestamps and the fourth number of second location timestamps in chronological order from early to late to generate a second time series;

[0029] The first positioning parameter set and the second positioning parameter set are fused according to the second time series to generate a positioning fusion parameter set.

[0030] Optionally, generating 3D point cloud data based on the shape fusion parameter set and the positioning fusion parameter set includes: traversing the shape fusion parameter set to obtain key timestamps and key shape data, wherein the key timestamps and the key shape data are correlated.

[0031] A first marker timestamp and a second marker timestamp are obtained from the positioning fusion parameters. The first marker timestamp is used to characterize the moment in the second time series with the shortest time interval from the reference timestamp, and the second marker timestamp is used to characterize the moment in the second time series with the second shortest time interval from the reference timestamp.

[0032] Acquire first marker positioning data and second marker positioning data, wherein the first marker positioning data is the vehicle front position information associated with the first marker timestamp, and the second marker positioning data is the vehicle front position information associated with the second marker timestamp;

[0033] Based on the key timestamp, the first marker timestamp, the second marker timestamp, the first marker positioning data, and the second marker positioning data, calculate the key positioning data corresponding to the key timestamp;

[0034] Establish the association between the key timestamp, the key positioning data, and the key shape data to generate three-dimensional point cloud data.

[0035] Optionally, based on the key timestamp, the first marker timestamp, the second marker timestamp, the first marker positioning data, and the second marker positioning data, the key positioning data corresponding to the key timestamp is calculated, including: calculating the difference between the key timestamp and the first marker timestamp to obtain a first time difference;

[0036] Calculate the difference between the second marked timestamp and the first marked timestamp to obtain the second time difference;

[0037] Divide the first time difference by the second time difference to calculate the time coefficient;

[0038] Calculate the difference between the second marker positioning data and the first marker positioning data to obtain the positioning data difference;

[0039] The time coefficient is multiplied by the difference in positioning data, and then added to the first marked positioning data to calculate the key positioning data.

[0040] Secondly, this application provides a vehicle outline measurement device based on lidar. The vehicle outline measurement device includes: a data acquisition module configured to: in response to a vehicle under test entering a detection zone, acquire a first outline parameter of the vehicle under test using a first outline radar, acquire a second outline parameter of the vehicle under test using a second outline radar, and acquire a first positioning parameter of the vehicle under test using a first positioning radar; the detection zone includes a first crossbar, a second crossbar, and a third crossbar arranged in parallel in sequence; the first outline radar is disposed at one end of the first crossbar, the second outline radar is disposed at the other end of the first crossbar, and the first positioning radar is disposed at the middle position of the second crossbar;

[0041] The type detection module is configured to: mark the vehicle type of the vehicle under test according to the first shape parameter, the second shape parameter and the first positioning parameter, wherein the vehicle type includes extra-long vehicles and regular vehicles;

[0042] The auxiliary positioning module is configured to: when the vehicle type of the vehicle under test is an extra-long vehicle, in response to the vehicle under test driving to the registration position, use a second positioning radar to collect the second positioning parameters of the vehicle under test, the registration position being set between the first crossbar and the second crossbar, and the second positioning radar being set at the middle position of the third crossbar;

[0043] The data fusion module is configured to: acquire vehicle separation signals, generate a set of shape fusion parameters based on the first shape parameters and the second shape parameters, and generate a set of positioning fusion parameters based on the first positioning parameters and the second positioning parameters, wherein the vehicle separation signals are used to characterize the rear of the vehicle under test passing the first crossbar;

[0044] The outline calculation module is configured to generate three-dimensional point cloud data based on the shape fusion parameter set and the positioning fusion parameter set, and calculate the outline parameters of the vehicle under test based on the three-dimensional point cloud data, wherein the outline parameters include length, width and height.

[0045] As can be seen from the above technical solution, this application provides a vehicle outline measurement method and device based on lidar. The method, in response to a vehicle entering a detection zone, uses a first outline radar and a second outline radar to collect the vehicle's first and second outline parameters, and uses a first positioning radar to collect the vehicle's first positioning parameters. The vehicle type is then labeled based on the first outline parameters, second outline parameters, and first positioning parameters. In the case of an extra-long vehicle, in response to the vehicle reaching the registration position, the second positioning radar collects the vehicle's second positioning parameters. Vehicle separation signals are acquired, and an outline fusion parameter set and a positioning fusion parameter set are generated. Three-dimensional point cloud data is generated based on the outline fusion parameter set and the positioning fusion parameter set, and the outline parameters of the vehicle are calculated based on the three-dimensional point cloud data. This method can measure the outline dimensions of extra-long vehicles while also measuring the outline dimensions of conventional vehicles. Attached Figure Description

[0046] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 Example diagram of a dual-bar three-radar setup for measuring the external dimensions of a vehicle;

[0048] Figure 2 This is a schematic diagram illustrating the steps of the vehicle outline measurement method according to an embodiment of this application;

[0049] Figure 3 This is a schematic diagram of the lidar installation according to an embodiment of this application;

[0050] Figure 4 This is a schematic diagram of the registration position in an embodiment of this application;

[0051] Figure 5 This is a schematic diagram illustrating the steps of generating a shape fusion parameter set according to an embodiment of this application;

[0052] Figure 6 This is a schematic diagram illustrating the steps of generating a location fusion parameter set according to an embodiment of this application;

[0053] Figure 7 This is a schematic diagram illustrating the steps of generating three-dimensional point cloud data according to an embodiment of this application;

[0054] Figure 8 This is a schematic diagram of the vehicle outline measuring device according to an embodiment of this application. Detailed Implementation

[0055] The embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described below do not represent all embodiments consistent with this application. They are merely examples of systems and methods consistent with some aspects of this application as detailed in the claims.

[0056] In the transportation industry, to reduce traffic accidents such as scratches and collisions caused by vehicles being too large on the road, it is necessary to measure the external dimensions of oversized transport vehicles. Two width and height radars are installed at the ends of the first crossbar, and a length-measuring radar is installed in the middle of the second crossbar. By using the width and height radars in conjunction with the length-measuring radar to locate the front position of the vehicle as it passes the width and height radars, the vehicle's length, width, and height can be calculated, ultimately determining the vehicle's external dimensions.

[0057] Figure 1 Example diagram of a dual-bar three-radar setup for measuring vehicle dimensions, as shown below. Figure 1 As shown, width and height radar 1 and width and height radar 2 are installed at both ends of the first crossbar to measure the width and height of the vehicle under test. The length-measuring radar is installed in the middle of the second crossbar to measure the length of the vehicle under test. By using the two width and height radars and the one length-measuring radar in conjunction, the system can detect the vehicle entering the detection zone (i.e., the front of the vehicle enters the first crossbar) and continue until the vehicle leaves the detection zone (i.e., the rear of the vehicle leaves the first crossbar). The length-measuring radar is then used to locate the position of the front of the vehicle, and the vehicle length is calculated as the difference between the distance between the two crossbars and the position of the front of the vehicle when it leaves the detection zone.

[0058] Due to the limited installation height of the length-measuring radar, in some embodiments, when measuring the length of extra-long vehicles using the above method, the length of the extra-long vehicle may exceed the measurement range of the radar, leading to inaccurate calculated vehicle dimensions. While increasing the installation distance between the first and second crossbars can achieve the measurement of extra-long vehicles, a longer installation distance may prevent the radar from detecting smaller vehicles in time, thus reducing the accuracy of measuring the dimensions of smaller vehicles. Therefore, the above method cannot simultaneously guarantee the measurement accuracy of the dimensions of both small and extra-long vehicles.

[0059] To address the challenge of simultaneously measuring the external dimensions of ultra-long vehicles and conventional vehicles, this application provides a LiDAR-based method for vehicle external dimension measurement. (See attached document.) Figure 2 The vehicle outline measurement method provided in this application includes:

[0060] Step S100: In response to the vehicle under test entering the detection zone, the first shape parameters of the vehicle under test are collected using the first shape radar, the second shape parameters of the vehicle under test are collected using the second shape radar, and the first positioning parameters of the vehicle under test are collected using the first positioning radar.

[0061] Figure 3 This is a schematic diagram of the lidar installation according to an embodiment of this application, as shown below. Figure 3 As shown, the detection range includes a first horizontal bar, a second horizontal bar, and a third horizontal bar arranged in parallel. A first shape radar is located at one end of the first horizontal bar, a second shape radar is located at the other end of the first horizontal bar, a first positioning radar is located in the middle of the second horizontal bar, and a second positioning radar is located in the middle of the third horizontal bar. For example, the distance between the first and second horizontal bars is 25 meters, and the distance between the second and third horizontal bars is 100 meters.

[0062] It should be noted that both the shape radar and the positioning radar can emit radar signals. When these signals encounter a target object, they are reflected. The shape radar and the positioning radar receive the reflected signals and calculate the target's speed relative to the radar based on the frequency difference between the emitted and received signals. Furthermore, they determine the target distance by measuring the round-trip time of the signal, thereby constructing the vehicle's three-dimensional shape, detecting the vehicle's boundary positions in the length and width directions, and determining the vehicle's speed as it approaches or moves away from the radar.

[0063] In some embodiments, the first shape parameters acquired by the first shape radar include first shape data and a first shape timestamp, wherein the first shape data and the first shape timestamp are correlated, and the first shape timestamp is used to characterize the acquisition time of the first shape data. By recording the acquisition time of the first shape data, a basis can be provided for subsequent calculation of the vehicle shape.

[0064] Similarly, the second shape parameter includes second shape data and a second shape timestamp, with the second shape timestamp representing the acquisition time of the second shape data. The first positioning parameter includes first positioning data and a first positioning timestamp. The first positioning data is the front position information of the vehicle under test acquired by the first positioning radar, and the first positioning timestamp represents the acquisition time of the first positioning data. The second positioning parameter includes second positioning data and a second positioning timestamp. The second positioning data is the front position information of the vehicle under test acquired by the second positioning radar, and the second positioning timestamp represents the acquisition time of the second positioning data.

[0065] Step S200: Mark the vehicle type of the vehicle under test according to the first external shape parameter, the second external shape parameter and the first positioning parameter.

[0066] In some embodiments, different radars are used for measuring the outline of different vehicle types, including extra-long vehicles and conventional vehicles. During the vehicle type labeling process, a vehicle identification signal is first generated based on a first outline parameter and a second outline parameter. This vehicle identification signal includes a vehicle identification timestamp, which characterizes the moment when the rear of the vehicle under test passes the first crossbar.

[0067] The vehicle separation signal indicates that the rear of the vehicle under test passes the first crossbar. In response to the first or second profile radar detecting the rear of the vehicle under test, the acquisition time of the first or second profile parameter is set as the vehicle separation timestamp, and a vehicle separation signal is generated based on the timestamp. Next, a registration timestamp is obtained based on the first positioning parameter. The registration timestamp represents the time information corresponding to the registration position when the first positioning data is used. If the vehicle separation timestamp is earlier than or equal to the registration timestamp, the vehicle under test is marked as a regular vehicle; if the vehicle separation timestamp is later than the registration timestamp, the vehicle under test is marked as an extra-long vehicle.

[0068] like Figure 4 As shown, a registration position is set between the first and second horizontal bars. The time corresponding to when the vehicle reaches the registration position is the registration timestamp. If the vehicle separation timestamp is earlier than or equal to the registration timestamp, it means that when the front of the vehicle under test reaches the registration position, the vehicle has already passed the first horizontal bar, and the vehicle under test is marked as a regular vehicle. If the vehicle separation timestamp is later than the registration timestamp, it means that when the front of the vehicle under test reaches the registration position, the vehicle has not yet passed the first horizontal bar, and the vehicle under test is marked as an oversized vehicle.

[0069] It should be noted that the distance between the registration position and the second crossbar is the registration interval. Within the registration interval, the first positioning radar and the second positioning radar need to perform data registration. Because when locating the front position of an extra-long vehicle, the first positioning parameters collected by the first positioning radar are needed to calculate the front position before the front of the vehicle passes the second crossbar, and the second positioning parameters are needed to calculate the front position after the front of the vehicle passes the second crossbar, therefore, the first and second positioning parameters need to be registered within the registration interval to improve the accuracy of vehicle outline measurement.

[0070] Step S300: Determine whether the vehicle type of the vehicle to be tested is an oversized vehicle.

[0071] Because the embodiments of this application use a first shape radar, a second shape radar, a first positioning radar, and a second positioning radar to collect data on the shape and front position of the vehicle when measuring the outline of an extra-long vehicle, while only the first shape radar, the second shape radar, and the first positioning radar are used to collect data on the shape and front position of a regular vehicle when measuring the outline of the vehicle, it is necessary to determine whether the vehicle to be measured is an extra-long vehicle, and then use different methods to perform the measurement of the vehicle outline.

[0072] Step S3011: When the vehicle type of the vehicle under test is an extra-long vehicle, in response to the vehicle under test driving to the registration position, the second positioning parameters of the vehicle under test are collected using the second positioning radar.

[0073] For example, vehicle B is 30 meters long. When the rear of vehicle B passes the first pole, the front of vehicle B has already exceeded the monitoring range of the first positioning radar. Therefore, it is necessary to use the first positioning radar to locate the front of vehicle B when the front of vehicle B just enters the first crossbar. After the front of vehicle B enters the registration position, the second positioning radar is used to locate the front of vehicle B. The shape of vehicle B is measured by combining the first and second shape radars, and then the outline of vehicle B is measured.

[0074] Step S3012: Obtain the vehicle separation signal, generate a shape fusion parameter set based on the first shape parameter and the second shape parameter, and generate a positioning fusion parameter set based on the first positioning parameter and the second positioning parameter.

[0075] In some embodiments, a first shape parameter is used to describe the shape information of one side of the vehicle, and a second shape parameter is used to describe the shape information of the other side of the vehicle. A shape fusion parameter set is generated based on the first shape parameter and the second shape parameter to obtain the overall shape information of the vehicle.

[0076] like Figure 5 As shown, a first set of shape parameters and a second set of shape parameters are obtained. The first set of shape parameters includes a first number of first shape data and a first number of first shape timestamps, and the first shape data and the first shape timestamps are correlated. The second set of shape parameters includes a second number of second shape data and a second number of second shape timestamps, and the second shape data and the second shape timestamps are correlated. The first number of first shape timestamps and the second number of second shape timestamps are then sorted in ascending order to generate a first time series. Finally, the first set of shape parameters and the second set of shape parameters are fused based on the first time series to generate a shape fusion parameter set.

[0077] For example, the first and second shape-sensing radars scan the shape of a vehicle passing directly beneath them at a frequency of 100 lines per second, with each scan line of each radar corresponding to a scan time t. x The first shape timestamps of the first shape radar are t 11 t 12 t 13 …t 1n The second shape timestamps of the second shape radar are t 21 t 22 t 23 …t 2n The first and second shape timestamps are sorted in chronological order from earliest to latest to generate a first time series. The scanning data from the two shape scanning radars are then fused according to this first time series to form a shape fusion parameter set with time stamps. The shape fusion parameter set includes the recorded timestamps: T1, T2, T3…T… n And the shape fusion data corresponding to the recorded time: d1, d2, d3…d n .

[0078] In some embodiments, in response to the second positioning radar detecting that the vehicle under test has traveled to the position of the second crossbar, the first positioning radar stops collecting the first positioning parameters of the vehicle under test, and generates a positioning fusion parameter set based on the first positioning parameters and the second positioning parameters.

[0079] like Figure 6 As shown, in the process of generating the positioning fusion parameter set, a first positioning parameter set and a second positioning parameter set are obtained. The first positioning parameter set includes a third number of first positioning data points and a third number of first positioning timestamps, which are correlated. The second positioning parameter set includes a fourth number of second positioning data points and a fourth number of second positioning timestamps, which are also correlated. The third number of first positioning timestamps and the fourth number of second positioning timestamps are then sorted in ascending order to generate a second time series. Finally, the first and second positioning parameter sets are fused based on the second time series to generate the positioning fusion parameter set.

[0080] For example, the vehicle under test is an extra-long vehicle. When the vehicle enters the detection zone, the first positioning radar records the position of the vehicle's front end (pos1) at different times. When the vehicle's front end reaches the registration position, the second positioning radar is activated to record the position of the front end (pos2) at different times. Based on the fact that the front end position of the same vehicle is the same at the same time, the second positioning radar aligns with the first positioning radar within the registration distance between the registration position and the second crossbar position, achieving data registration between the two positioning radars. After data registration, in response to the vehicle's measured position reaching below the second crossbar, the first positioning radar stops collecting the first positioning parameters of the vehicle under test. The first and second positioning parameter sets are then fused to generate a positioning fusion parameter set. The positioning fusion parameter set includes the recording times: t1, t2, t3…t n And the corresponding front positions of the vehicles: p1, p2, p3…p n .

[0081] Step S3013: Generate three-dimensional point cloud data based on the shape fusion parameter set and the positioning fusion parameter set, and calculate the outline parameters of the vehicle under test based on the three-dimensional point cloud data.

[0082] In some embodiments, the outline parameters include length, width, and height. For example... Figure 7 As shown, in the process of generating 3D point cloud data based on the shape fusion parameter set and the positioning fusion parameter set, the shape fusion parameter set is first traversed to obtain key timestamps and key shape data. Then, the first and second marked timestamps are obtained from the positioning fusion parameters. The key timestamps are related to the key shape data. The first marked timestamp is used to characterize the moment with the shortest time interval from the reference timestamp in the second time series, and the second marked timestamp is used to characterize the moment with the second shortest time interval from the reference timestamp in the second time series.

[0083] Acquire first and second marker positioning data, and calculate key positioning data corresponding to the key timestamp based on the key timestamp, the first marker timestamp, the second marker timestamp, the first marker positioning data, and the second marker positioning data. The first marker positioning data is the vehicle front position information associated with the first marker timestamp, and the second marker positioning data is the vehicle front position information associated with the second marker timestamp. Finally, establish the association between the key timestamp, key positioning data, and key shape data to generate 3D point cloud data.

[0084] For example, with the key timestamp T1, the two time points closest to T1 are obtained from the second time series and used as the first marker timestamp t1 and the second marker timestamp t2. The first marker positioning data p1 corresponding to the first marker timestamp t1 is obtained, the second marker positioning data p2 corresponding to the second marker timestamp t2 is obtained, and the key positioning data P1 corresponding to the key timestamp T1 is calculated.

[0085] In some embodiments, the key positioning data corresponding to the key timestamp is calculated using the following formula: First, the difference between the key timestamp and the first marker timestamp is calculated to obtain the first time difference; then, the difference between the second marker timestamp and the first marker timestamp is calculated to obtain the second time difference; then, the first time difference and the second time difference are divided to calculate the time coefficient; then, the difference between the second marker positioning data and the first marker positioning data is calculated to obtain the positioning data difference; finally, the time coefficient is multiplied by the positioning data difference and added to the first marker positioning data to calculate the key positioning data. The formula for calculating the key positioning data is as follows:

[0086]

[0087] Among them, P i For key location data, T i For key timestamps, t i1 For the first marker timestamp, t i2 p is the second marker timestamp. i1 For the first marker location data, p i2 The second marker is used to locate the data.

[0088] Step S3021: When the vehicle length type of the vehicle under test is a regular length vehicle, in response to the rear of the vehicle under test passing the first crossbar, a set of shape fusion parameters is generated based on the first shape parameter and the second shape parameter.

[0089] For example, vehicle A is 6.5 meters long. When the rear of vehicle A passes the first crossbar, the first positioning radar can continuously detect the position of vehicle A. The second positioning radar is far away from the front of vehicle A. Therefore, by using the first and second shape radars to measure the shape of vehicle A and using the first positioning radar to locate the front of vehicle A, the outline of vehicle A can be measured without using the second positioning radar.

[0090] A set of shape fusion parameters is generated based on the first and second shape parameters. The process of generating the shape fusion parameter set for ultra-long vehicles is similar and will not be repeated here.

[0091] Step S3022: Generate three-dimensional point cloud data based on the shape fusion parameter set and the first positioning parameters, and calculate the outline parameters of the vehicle under test based on the three-dimensional point cloud data.

[0092] In some embodiments, a three-dimensional spatial range is set according to the approximate position of the vehicle in space to extract points of the vehicle's outline. The points in the point cloud data are divided into different clusters using a clustering algorithm. The vehicle's outline points are identified based on the characteristics of the clusters, and outline parameters, including length, width, and height, are calculated based on the outline points.

[0093] As can be seen from the above technical solution, this application provides a vehicle outline measurement method and device based on lidar. The method, in response to a vehicle entering a detection zone, uses a first outline radar and a second outline radar to collect the vehicle's first and second outline parameters, and uses a first positioning radar to collect the vehicle's first positioning parameters. The vehicle type is then labeled based on the first outline parameters, second outline parameters, and first positioning parameters. In the case of an extra-long vehicle, in response to the vehicle reaching the registration position, the second positioning radar collects the vehicle's second positioning parameters. Vehicle separation signals are acquired, and an outline fusion parameter set and a positioning fusion parameter set are generated. Three-dimensional point cloud data is generated based on the outline fusion parameter set and the positioning fusion parameter set, and the outline parameters of the vehicle are calculated based on the three-dimensional point cloud data. This method can measure the outline dimensions of extra-long vehicles while also measuring the outline dimensions of conventional vehicles.

[0094] Based on the vehicle outline measurement method based on LiDAR provided in the above embodiments, some embodiments of this application also provide a vehicle outline measurement device based on LiDAR, see [link to relevant documentation]. Figure 8 The vehicle outline measuring device is used to perform the vehicle outline measuring method described in the first embodiment, and the vehicle outline measuring device includes:

[0095] The data acquisition module 801 is configured to: in response to the vehicle under test entering the detection zone, acquire the first shape parameters of the vehicle under test using a first shape radar, acquire the second shape parameters of the vehicle under test using a second shape radar, and acquire the first positioning parameters of the vehicle under test using a first positioning radar; the detection zone includes a first crossbar, a second crossbar, and a third crossbar arranged in parallel in sequence; the first shape radar is located at one end of the first crossbar, the second shape radar is located at the other end of the first crossbar, and the first positioning radar is located at the middle position of the second crossbar.

[0096] The type detection module 802 is configured to: mark the vehicle type of the vehicle under test according to the first shape parameter, the second shape parameter and the first positioning parameter, the vehicle type including extra-long vehicles and regular vehicles.

[0097] The auxiliary positioning module 803 is configured to: when the vehicle type of the vehicle under test is an extra-long vehicle, in response to the vehicle under test driving to the registration position, use the second positioning radar to collect the second positioning parameters of the vehicle under test. The registration position is set between the first and second crossbars, and the second positioning radar is set at the middle position of the third crossbar.

[0098] The data fusion module 804 is configured to: acquire vehicle separation signals, generate a set of shape fusion parameters based on the first shape parameters and the second shape parameters, and generate a set of positioning fusion parameters based on the first positioning parameters and the second positioning parameters. The vehicle separation signals are used to characterize the rear of the vehicle under test passing the first crossbar.

[0099] The outline calculation module 805 is configured to generate three-dimensional point cloud data based on the shape fusion parameter set and the positioning fusion parameter set, and calculate the outline parameters of the vehicle under test based on the three-dimensional point cloud data. The outline parameters include length, width and height.

[0100] The effects of the above system in the execution of the method can be found in the description of the method above, and will not be repeated here.

[0101] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein; the specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the claims.

[0102] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope; the scope of the invention is limited only by the appended claims.

Claims

1. A method for measuring vehicle outline based on lidar, characterized in that, include: In response to the vehicle under test entering the detection zone, the system uses a first shape radar to collect the first shape parameters of the vehicle under test, a second shape radar to collect the second shape parameters of the vehicle under test, and a first positioning radar to collect the first positioning parameters of the vehicle under test. The detection zone includes a first crossbar, a second crossbar, and a third crossbar arranged in parallel in sequence. The first shape radar is located at one end of the first crossbar, the second shape radar is located at the other end of the first crossbar, and the first positioning radar is located at the middle position of the second crossbar. The vehicle type of the vehicle under test is marked according to the first shape parameter, the second shape parameter and the first positioning parameter, and the vehicle type includes extra-long vehicles and regular vehicles; When the vehicle type of the vehicle under test is an extra-long vehicle, in response to the vehicle under test driving to the registration position, the second positioning parameters of the vehicle under test are collected by the second positioning radar. The registration position is set between the first crossbar and the second crossbar, and the second positioning radar is set at the middle position of the third crossbar. The vehicle separation signal is acquired, and a set of shape fusion parameters is generated based on the first shape parameter and the second shape parameter, and a set of positioning fusion parameters is generated based on the first positioning parameter and the second positioning parameter. The vehicle separation signal is used to characterize the rear of the vehicle under test passing the first crossbar. Three-dimensional point cloud data is generated based on the shape fusion parameter set and the positioning fusion parameter set, and the outline parameters of the vehicle under test are calculated based on the three-dimensional point cloud data. The outline parameters include length, width and height.

2. The vehicle outline measurement method based on lidar according to claim 1, characterized in that, Also includes: When the vehicle length type of the vehicle under test is a regular length vehicle, in response to the rear of the vehicle under test passing the first crossbar, a set of shape fusion parameters is generated based on the first shape parameters and the second shape parameters. Three-dimensional point cloud data is generated based on the shape fusion parameter set and the first positioning parameters, and the outline parameters of the vehicle under test are calculated based on the three-dimensional point cloud data. The outline parameters include length, width and height.

3. The vehicle outline measurement method based on lidar according to claim 2, characterized in that, The first shape parameter includes first shape data and a first shape timestamp, wherein the first shape timestamp is used to characterize the acquisition time of the first shape data; The second shape parameter includes second shape data and a second shape timestamp, wherein the second shape timestamp is used to characterize the acquisition time of the second shape data; The first positioning parameter includes first positioning data and a first positioning timestamp. The first positioning data is the front position information of the vehicle under test collected by the first positioning radar, and the first positioning timestamp is used to characterize the time when the first positioning data was collected. The second positioning parameter includes second positioning data and a second positioning timestamp. The second positioning data is the front position information of the vehicle under test collected by the second positioning radar, and the second positioning timestamp is used to characterize the time when the second positioning data was collected.

4. The vehicle outline measurement method based on lidar according to claim 3, characterized in that, The vehicle type of the vehicle under test is identified based on the first external shape parameter, the second external shape parameter, and the first positioning parameter, including: A vehicle separation signal is generated based on the first shape parameter and the second shape parameter. The vehicle separation signal includes a vehicle separation timestamp, which is used to characterize the time when the rear of the vehicle under test passes the first crossbar. The registration timestamp is obtained based on the first positioning parameter, and the registration timestamp is used to characterize the time information corresponding to when the first positioning data is the registration position; If the vehicle assignment timestamp is earlier than or equal to the registration timestamp, then the vehicle under test is marked as a regular vehicle. If the vehicle assignment timestamp is later than the registration timestamp, the vehicle under test is marked as an oversized vehicle.

5. The vehicle outline measurement method based on lidar according to claim 4, characterized in that, The step of generating a vehicle assignment signal based on the first and second shape parameters includes: In response to the first or second shape radar detecting the rear of the vehicle under test, the vehicle-specific timestamp is generated, wherein the vehicle-specific timestamp is the time when the first or second shape parameter is collected; The vehicle allocation signal is generated based on the vehicle allocation timestamp.

6. The vehicle outline measurement method based on lidar according to claim 1, characterized in that, The step of generating a shape blending parameter set based on the first shape parameter and the second shape parameter includes: Obtain a first set of shape parameters, the first set of shape parameters including a first number of first shape data and a first number of first shape timestamps, the first shape data and the first shape timestamps are related; Obtain a second set of shape parameters, which includes a second number of second shape data and a second number of second shape timestamps, and the second shape data and the second shape timestamps are related. Sort the first number of first shape timestamps and the second number of second shape timestamps in chronological order from early to late to generate a first time series; The first set of shape parameters and the second set of shape parameters are fused according to the first time series to generate the shape fusion parameter set.

7. The vehicle outline measurement method based on lidar according to claim 6, characterized in that, In response to the second positioning radar detecting that the vehicle under test has moved to the position of the second crossbar, the first positioning radar stops collecting the first positioning parameters of the vehicle under test, and the step of generating a positioning fusion parameter set based on the first positioning parameters and the second positioning parameters includes: Obtain a first set of positioning parameters, which includes a third number of first positioning data and a third number of first positioning timestamps, and the first positioning data and the first positioning timestamps are related. Obtain a second set of positioning parameters, which includes a fourth number of second positioning data and a fourth number of second positioning timestamps, and the second positioning data and the second positioning timestamps are related. Sort the third number of first location timestamps and the fourth number of second location timestamps in chronological order from early to late to generate a second time series; The first positioning parameter set and the second positioning parameter set are fused according to the second time series to generate a positioning fusion parameter set.

8. The vehicle outline measurement method based on lidar according to claim 7, characterized in that, The generation of 3D point cloud data based on the shape fusion parameter set and the positioning fusion parameter set includes: The set of shape fusion parameters is traversed to obtain key timestamps and key shape data, wherein the key timestamps and key shape data are correlated. A first marker timestamp and a second marker timestamp are obtained from the positioning fusion parameters. The first marker timestamp is used to characterize the moment in the second time series with the shortest time interval from the reference timestamp, and the second marker timestamp is used to characterize the moment in the second time series with the second shortest time interval from the reference timestamp. Acquire first marker positioning data and second marker positioning data, wherein the first marker positioning data is the vehicle front position information associated with the first marker timestamp, and the second marker positioning data is the vehicle front position information associated with the second marker timestamp; Based on the key timestamp, the first marker timestamp, the second marker timestamp, the first marker positioning data, and the second marker positioning data, calculate the key positioning data corresponding to the key timestamp; Establish the association between the key timestamp, the key positioning data, and the key shape data to generate three-dimensional point cloud data.

9. The vehicle outline measurement method based on lidar according to claim 8, characterized in that, The step of calculating the key positioning data corresponding to the key timestamp based on the key timestamp, the first marker timestamp, the second marker timestamp, the first marker positioning data, and the second marker positioning data includes: Calculate the difference between the key timestamp and the first marked timestamp to obtain the first time difference; Calculate the difference between the second marked timestamp and the first marked timestamp to obtain the second time difference; Divide the first time difference by the second time difference to calculate the time coefficient; Calculate the difference between the second marker positioning data and the first marker positioning data to obtain the positioning data difference; The time coefficient is multiplied by the difference in positioning data, and then added to the first marked positioning data to calculate the key positioning data.

10. A vehicle outline measurement device based on lidar, characterized in that, include: The data acquisition module is configured to: in response to a vehicle under test entering the detection zone, acquire a first shape parameter of the vehicle under test using a first shape radar, acquire a second shape parameter of the vehicle under test using a second shape radar, and acquire a first positioning parameter of the vehicle under test using a first positioning radar; the detection zone includes a first crossbar, a second crossbar, and a third crossbar arranged in parallel in sequence; the first shape radar is located at one end of the first crossbar, the second shape radar is located at the other end of the first crossbar, and the first positioning radar is located at the middle position of the second crossbar; The type detection module is configured to: mark the vehicle type of the vehicle under test according to the first shape parameter, the second shape parameter and the first positioning parameter, wherein the vehicle type includes extra-long vehicles and regular vehicles; The auxiliary positioning module is configured to: when the vehicle type of the vehicle under test is an extra-long vehicle, in response to the vehicle under test driving to the registration position, use a second positioning radar to collect the second positioning parameters of the vehicle under test, the registration position being set between the first crossbar and the second crossbar, and the second positioning radar being set at the middle position of the third crossbar; The data fusion module is configured to: acquire vehicle separation signals, generate a set of shape fusion parameters based on the first shape parameters and the second shape parameters, and generate a set of positioning fusion parameters based on the first positioning parameters and the second positioning parameters, wherein the vehicle separation signals are used to characterize the rear of the vehicle under test passing the first crossbar; The outline calculation module is configured to generate three-dimensional point cloud data based on the shape fusion parameter set and the positioning fusion parameter set, and calculate the outline parameters of the vehicle under test based on the three-dimensional point cloud data, wherein the outline parameters include length, width and height.

Images