High-precision radar speed measurement method in tunnel for intelligent transportation

Abstract

The application relates to the technical field of radar measurement, in particular to a high-precision radar speed measurement method in a tunnel for intelligent traffic. The method matches initial point cloud data of different radars by using a matching method, determines information weight values corresponding to each radar in combination with point cloud quality, and further determines effective point cloud data. The point cloud speed and the Doppler speed are further weighted and fused to obtain movement information of each vehicle area. The interference degree of the vehicle area at each moment is determined, and then the influence degree of each historical moment can be determined in combination with information between the historical moment and the real-time moment. The predicted movement information is determined in combination with the influence degree, and the movement information of the preceding vehicle is used for adjustment. The application eliminates the influence of the multipath effect in the tunnel, and realizes effective speed measurement and accurate prediction of each vehicle.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of radar measurement, in particular to a high-precision radar speed measurement method in a tunnel for intelligent transportation. BACKGROUND

[0002] In the tunnel, millimeter wave radar is usually used to measure the speed of vehicles, and the speed measurement results are uploaded to the cloud through Internet of Things technology to realize intelligent operation and management of tunnel transportation. Inside the tunnel, due to the complex tunnel structure and vehicle environment, the millimeter wave signal may be affected by multipath effect, resulting in signal reflection and refraction errors, reducing the measurement accuracy. For example, when the millimeter wave radar measures the speed in the tunnel, the concrete and metal structure of the tunnel wall will reflect the radar signal, so that the electromagnetic wave of the same target returns to the receiving device through multiple paths such as direct reflection, which affects the accuracy of the measured speed.

[0003] The prior art usually distinguishes the measured target from the distance, speed, azimuth and pitch angle of the radar to measure the speed of the moving target, but the above method has low accuracy in complex environments. For example, when the traffic in the tunnel is large and there are large vehicles, the phenomenon of multi-path interference may be aggravated, which is not conducive to accurate speed measurement of moving vehicles. SUMMARY

[0004] In order to solve the technical problem that the prior art cannot effectively analyze the data collected by the radar due to the complex vehicle environment in the tunnel, and cannot accurately measure the speed of the vehicle in the tunnel, the purpose of the present application is to provide a high-precision radar speed measurement method in a tunnel for intelligent transportation, and the technical solution is as follows:

[0005] The present application provides a high-precision radar speed measurement method in a tunnel for intelligent transportation, which comprises:

[0006] A plurality of radars are installed at the monitoring position in the tunnel to obtain initial point cloud data obtained by each radar at the collection time; the information weight of each initial point cloud data is obtained according to the matching degree between the initial point clouds and the point cloud quality of each initial point cloud; and all the initial point cloud data are registered by using the information weight to obtain effective point cloud data;

[0007] The vehicle area in the effective point cloud data is obtained, the point cloud speed of each vehicle area at each time is obtained by the motion change of the vehicle area at adjacent time, and the moving information of each vehicle area at real time is obtained by weighting analysis of the point cloud density, weighting fusion of the point cloud speed and the Doppler speed.

[0008] For each vehicle area, the interference degree of each vehicle area is obtained according to the Doppler velocity stability of the vehicle area at the corresponding time; the influence degree of each historical time is obtained according to the movement information difference between each historical time and the real time, the interference degree of the vehicle area at the historical time, and the distance between the historical time and the real time, the movement information of the historical time is analyzed by using the influence degree, the predicted movement information of each vehicle area is obtained, and the predicted movement information of each vehicle area is adjusted by using the movement information of the front vehicle at the real time.

[0009] Further, the matching degree acquisition method comprises:

[0010] For any point cloud in any initial point cloud data, the point cloud is matched with other point clouds in other initial point cloud data within a preset neighborhood range, and it is judged whether the point cloud is a stable feature point according to the matching result; the number ratio of stable feature points in the initial point cloud data is taken as the matching degree of each initial point cloud data.

[0011] Further, the method for judging whether the point cloud is a stable feature point according to the matching result comprises:

[0012] Any one initial point cloud data is taken as target initial point cloud data, any one point cloud in the target initial point cloud data is taken as target point cloud, and other point clouds of other initial point cloud data in the preset neighborhood range of the target point cloud are taken as matching point clouds; the cosine similarity of the FPFH vector between the target point cloud and the matching point cloud is obtained, the Euclidean distance between the target point cloud and the matching point cloud is combined, and the matching index between the target point cloud and the matching point cloud is obtained according to the difference of the Doppler velocity; the matching point cloud of the target point cloud is taken as the matching point cloud of the target point cloud when the matching index is greater than a preset threshold; if there are matching point clouds of the target point cloud in all other initial point cloud data, the target point cloud is a stable feature point.

[0013] Further, the point cloud quality acquisition method comprises:

[0014] The product of the point cloud density and the average value of the echo intensity is taken as the point cloud quality of an initial point cloud data.

[0015] Further, the vehicle area acquisition method comprises:

[0016] The point clouds in the effective point cloud data are clustered by using the DBSCAN clustering algorithm, and each clustering cluster obtained is a vehicle area.

[0017] Further, the interference degree acquisition method comprises:

[0018] Obtaining the Doppler velocity range between all point clouds in each vehicle region, normalizing the ratio of the Doppler velocity range and the number of point clouds in the vehicle region to obtain the interference degree.

[0019] Further, the obtaining of the movement information of each vehicle region at the real-time moment comprises:

[0020] Normalizing the point cloud density of the effective point cloud data to obtain a point cloud density weight; taking the point cloud density weight as the weight of the point cloud velocity, taking the result of subtracting the point cloud density weight from the positive integer 1 as the weight of the Doppler velocity, and weighting and fusing the point cloud velocity and the Doppler velocity to obtain the movement speed in the movement information of each vehicle region at the real-time moment.

[0021] The point cloud movement direction of the vehicle region between adjacent moments is used as the movement direction in the movement information.

[0022] Further, the obtaining method of the influence degree comprises:

[0023] For each historical moment, a direction similarity in the movement direction between the historical moment and the real-time moment is obtained, and a movement speed difference between the movement information is obtained; and according to the movement speed difference, a time sequence distance between the historical moment and the real-time moment, and an interference degree of the vehicle region at the historical moment, an untrustworthiness of the historical moment is obtained; and the ratio of the direction similarity and the untrustworthiness is taken as the influence degree.

[0024] Further, the obtaining method of the predicted movement information comprises:

[0025] For each historical moment, the movement direction of the vehicle region at each historical moment is converted to a unit vector, and the unit vectors of all historical moments are weighted and fused by using the influence degree to obtain a predicted movement direction in the predicted movement information.

[0026] The movement speed in the movement information at the real-time moment is adjusted according to the movement speed increment of the real-time moment relative to the previous moment to obtain a predicted movement speed in the predicted movement information.

[0027] Further, the adjusting of the predicted movement information of each vehicle region by using the movement information of the front vehicle at the real-time moment comprises:

[0028] If there is no front vehicle in the vehicle region at the real-time moment, the predicted movement information of the vehicle region is not adjusted.

[0029] If there is a front vehicle in the vehicle area at the real-time moment, and the distance is less than the safety distance specified by the tunnel, then the predicted moving speed in the predicted moving information is reduced according to the distance between the vehicle area at the real-time moment and the front vehicle, and the difference between the predicted moving speed of the vehicle area and the moving speed of the front vehicle.

[0030] The present application has the following beneficial effects:

[0031] The present application obtains initial point cloud data collected by different radars at each monitoring position of the tunnel. Considering the problem of reduced reference of initial point cloud data of a certain radar caused by multipath effect, the present application matches the initial point cloud data of different radars, and then determines the information weight value corresponding to each radar in combination with point cloud quality, and further determines effective point cloud data. In order to further realize accurate speed measurement of vehicles, after obtaining the effective point cloud data, the point cloud speed and the Doppler speed are further weighted and fused to obtain the moving information of each vehicle area, which reduces the influence of multipath effect and reflects the real and effective moving information. The present application further considers the demand of intelligent traffic for data prediction, first determines the interference degree of the vehicle area at each moment, and then determines the influence degree of each historical moment in combination with the information between the historical moment and the real-time moment. The influence degree represents the reference degree of information at the historical moment to the real-time moment, and then the predicted moving information can be determined in combination with the influence degree, and the moving information of the front vehicle can be further adjusted to realize effective speed measurement and accurate prediction of each vehicle. BRIEF DESCRIPTION OF DRAWINGS

[0032] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, a brief introduction will be given to the drawings needed in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor based on these drawings.

[0033] Figure 1 A flow chart of a high-precision radar speed measurement method for intelligent traffic in a tunnel is provided for an embodiment of the present application.

[0034] Figure 2 A tunnel radar deployment schematic diagram is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0035] To further illustrate the technical means and effects of the present invention to achieve the intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effects of a high-precision radar speed measurement method in a tunnel for smart transportation proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable form.

[0036] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0037] The following describes in detail a specific scheme of a high-precision radar speed measurement method in a tunnel for smart transportation provided by the present invention with reference to the accompanying drawings.

[0038] See also Figure 1 , which shows a flow chart of a high-precision radar speed measurement method in a tunnel for smart transportation provided by one embodiment of the present invention, the method comprising:

[0039] Step S1: Multiple radars are installed at monitoring locations in the tunnel, and the initial point cloud data obtained by each radar at the time of acquisition is obtained; based on the degree of matching between the initial point clouds and the point cloud quality of each initial point cloud, the information weight of each initial point cloud data is obtained; and all the initial point cloud data are aligned using the information weight to obtain valid point cloud data.

[0040] In the embodiment of the present invention, in order to reduce the interference of multipath effect, a multi-radar information fusion method is used to improve the information reference. Therefore, multiple radars are installed at each monitoring position in the tunnel, and each radar can obtain the corresponding initial point cloud data at each sampling time. That is, the initial point cloud data is the obtained point cloud set. Figure 2Fig. 1 shows a schematic diagram of a tunnel radar deployment according to an embodiment of the present application, which shows a tunnel radar deployment schematic diagram provided by an embodiment of the present application, the embodiment of the present application sets multiple monitoring positions at a fixed interval in the tunnel, each monitoring position is a monitoring section, three millimeter wave radars are deployed on the monitoring section to obtain initial point cloud data. In the embodiment of the present application, the fixed interval between the monitoring sections is set to 500 meters, and the millimeter wave radars are deployed directly above each monitoring section and at a position deviated by 15° from the directly above. It should be noted that the embodiment of the present application considers that radars of different frequency bands have different sensitivities to interference, and selects multiple frequency band radars for joint monitoring, which can reduce the risk of detection failure caused by interference of a single frequency band, and makes the radars of the same section correspond to different frequency bands, for example, in the embodiment of the present application, 4D millimeter wave radars in 76-79GHz, 24.05-24.25GHz and 92-96GHz are selected respectively. In the embodiment of the present application, the data acquisition frequency is set to 30 Hz, and after the data acquisition of each radar is completed, the coordinates of the initial point cloud data can be calibrated, and the initial point cloud data can be mapped to the same coordinate system for data analysis.

[0041] It should be noted that after obtaining the initial point cloud data in the embodiment of the present application, in order to avoid the interference of non-vehicle information, the points with Doppler velocity of 0 are screened out, and the initial point cloud data after preprocessing is a point cloud set containing only vehicle information point cloud.

[0042] In the tunnel, the detection range of a single radar may be incomplete due to large vehicle shielding, or distorted due to multi-path interference caused by metal interference, tunnel wall reflection and other factors, therefore, the embodiment of the present application combines the initial point cloud data collected by multiple radars to obtain effective point cloud data with strong reference and less interference.

[0043] Due to the complexity of the environment in the tunnel, the traffic flow is large and there is large vehicle shielding, which has different degrees of influence on a single radar at different times, and it is impossible to use a fixed weight to calibrate the information effectiveness of each radar. Therefore, the embodiment of the present application obtains the data information weight of each initial point cloud in real time and dynamically based on the matching degree between the initial point clouds and the point cloud quality of each initial point cloud. That is, for an initial point cloud data, the more matched it is with other initial point cloud data and the higher the point cloud quality is, the smaller the interference on the corresponding radar is, the lighter the data loss is, the greater the reference of the initial point cloud data information is, and the greater the information weight is.

[0044] After the information weight is obtained, the initial point cloud data collected by all radars at each moment can be registered to obtain the effective point cloud data at the moment. In the embodiment of the present application, the initial point cloud data collected by the radars at the same monitoring section can be registered by using the information weight by using the ICP algorithm, and the specific content is a technical means familiar to those skilled in the art, which is not described here.

[0045] Preferably, in the embodiment of the present application, the matching degree acquisition method comprises:

[0046] For any point cloud in any initial point cloud data, the point cloud is matched with other point clouds in other initial point cloud data within a preset neighborhood range, and whether the point cloud is a stable feature point is determined according to the matching result. For an initial point cloud, the more stable feature points, the less the initial point cloud data is affected by the multipath effect, and the less the data loss effect is, so the proportion of the number of stable feature points in the initial point cloud data can be used as the matching degree of each initial point cloud data.

[0047] Further, in the embodiment of the present application, in the matching process between point clouds, the more similar the Doppler velocities between two point clouds are, and the closer the spatial distances are, the higher the matching degree between the two points should be, so whether the point cloud is a stable feature point is determined according to the matching result, comprising:

[0048] Any one initial point cloud data is taken as target initial point cloud data, any one point cloud in the target initial point cloud data is taken as target point cloud, and other point clouds in other initial point cloud data within a preset neighborhood range of the target point cloud are taken as matching point clouds.

[0049] The cosine similarity of the Fast Point Feature Histogram (FPFH) vector between the target point cloud and the matching point cloud is obtained, and the matching index between the target point cloud and the matching point cloud is obtained by combining the Euclidean distance between the target point cloud and the matching point cloud and the difference in Doppler velocity. That is, the smaller the Euclidean distance, the smaller the difference in Doppler velocity, and the larger the cosine similarity between the target point cloud and the matching point cloud, the more matched the two point clouds are, and the larger the matching index is.

[0050] The matching point cloud of the target point cloud is taken as the matching point cloud whose matching index is greater than a preset threshold; if there are matching point clouds of the target point cloud in all other initial point cloud data, the target point cloud is a stable feature point. Similarly, all stable feature points can be determined in each initial point cloud.

[0051] In the embodiment of the present application, the radius of the preset neighborhood range is set to 0.3 meters in the real coordinate system, that is, the range with a radius of 0.3 meters centered on the target point cloud is the neighborhood range. After the matching index is normalized, the threshold is set to 0.7. The normalization method adopts the range standardization, and the numerical value is limited between 0 and 1.

[0052] As an example, in the embodiment of the present application, the matching index is obtained by: taking the absolute value of the difference between the Doppler velocities of the point clouds as the difference in Doppler velocity, multiplying the Euclidean distance and the difference in Doppler velocity to obtain the difference index. The matching index is obtained by multiplying the difference index after negative correlation mapping and the cosine similarity. It should be noted that the negative correlation mapping method is a basic mathematical means familiar to those skilled in the art, and specific methods such as inverse form and function mapping can be adopted. In the embodiment of the present application, the inverse form is adopted, and in order to avoid the denominator being 0, the inverse of the difference index added by 0.01 is taken as the negative correlation mapping result.

[0053] Preferably, in the embodiment of the present application, the product of the point cloud density and the average value of the echo intensity is taken as the point cloud quality of an initial point cloud data.

[0054] In the embodiment of the present application, after the point cloud quality and the matching degree are quantified, the product of the two is taken as the information weight.

[0055] Step S2: obtaining the vehicle region in the effective point cloud data, obtaining the point cloud velocity of each vehicle region at each time through the motion change of the vehicle region at adjacent times, and obtaining the moving information of each vehicle region at the real time through weighted analysis of the point cloud density, weighted fusion of the point cloud velocity and the Doppler velocity.

[0056] The effective point cloud data obtained in step S1 is the point cloud data with the influence of multipath effect reduced, but in order to further obtain the effective motion information of the vehicle, the embodiment of the present application does not directly take the Doppler velocity reflected by the effective point cloud data as the real velocity information. The Doppler velocity refers to the radial velocity of the target object relative to the observation device measured by applying the Doppler effect, and its core principle is based on the frequency change caused by the relative motion of the wave source and the receiver. Therefore, the two kinds of motion information need to be fused and verified to obtain effective moving information.

[0057] Firstly, the embodiment of the present application obtains the vehicle region in the effective point cloud data. Because the vehicle has a fixed shape, only the vehicle moves in the tunnel, and other moving objects such as pedestrians do not appear, so the point cloud obtained after preprocessing is only the vehicle point cloud, and therefore the vehicle region can be effectively screened in the effective point cloud data according to the distance between the point clouds. The vehicle regions at different times are matched, and the location and point cloud information of each vehicle region at different times are determined. The matching process is similar to the point cloud data matching method in step S1: for a target vehicle region at a real-time time, the target vehicle region is matched with each vehicle region at a historical time, and the vehicle region at the historical time to be matched is a to-be-matched vehicle region. The target point cloud in the target vehicle region is matched with the to-be-matched point cloud in the to-be-matched vehicle region to calculate the matching index, and if the matching index of a certain matching point cloud is greater than a preset threshold, it is determined that the matching is successful. Change the target point cloud, and the proportion of the matching point cloud in the target vehicle region is used as the matching degree between the target vehicle region and the to-be-matched vehicle region. The vehicle region with a matching degree greater than 0.8 at the historical time is selected as the same vehicle region of the target vehicle region.

[0058] It should be noted that, considering that the radar has a monitoring range, during the vehicle region matching process between different times, the historical time range is set to the ratio of the monitoring section interval to the maximum historical vehicle speed.

[0059] At this point, the point cloud speed of each vehicle region at each time can be obtained by the movement change of the vehicle region at adjacent times. For the real-time time, the centroid point of the vehicle region is selected as the reference point, the moving distance of the reference point between the real-time time and the previous time is obtained, and the ratio of the moving distance to the time interval is the point cloud speed. The point cloud speed is the speed information determined by the point cloud movement. For the point cloud, the greater the point cloud density, the greater the reference of the point cloud speed; on the contrary, the smaller the point cloud density, the smaller the reference of the point cloud speed, and more reference Doppler speed is needed. Therefore, the point cloud density can be used for weight analysis, the point cloud speed and the Doppler speed are weighted and fused to obtain the movement information of each vehicle region at the real-time time.

[0060] It should be noted that, because the effective point cloud data is registered point cloud data, the Doppler speed of the vehicle region is the average of the Doppler speed components of all points in the region in the common coordinate system.

[0061] Preferably, in the embodiment of the present application, the vehicle region acquisition method comprises:

[0062] The point clouds in the effective point cloud data are clustered by using a density-based clustering algorithm (Density-Based Spatial Clustering of Applications with Noise, DBSCAN), and each clustering cluster is a vehicle area. In the embodiment of the application, the neighborhood radius in the DBSCAN clustering algorithm is set to 0.3 meters in the real coordinate system, and the minimum sample number is set to 5.

[0063] Preferably, in the embodiment of the application, the movement information of each vehicle area at the real-time time is obtained, including:

[0064] The point cloud density of the effective point cloud data is normalized to obtain a point cloud density weight; the point cloud density weight is used as the weight of the point cloud velocity, and an integer 1 minus the result of the point cloud density weight is used as the weight of the Doppler velocity; the point cloud velocity and the Doppler velocity are weighted and fused to obtain the movement speed in the movement information of each vehicle area at the real-time time. The point cloud movement direction between adjacent time instants of the vehicle area is used as the movement direction in the movement information.

[0065] It should be noted that the normalization algorithm in the embodiment of the application can use range standardization, that is, in the respective dimensions, the maximum and minimum values are used for normalization, and details are not repeated.

[0066] Step S3: For each vehicle area, the interference degree of each vehicle area is obtained according to the Doppler velocity stability of the vehicle area at the corresponding time; the influence degree of each historical time is obtained according to the movement information difference between each historical time and the real-time time, the interference degree of the vehicle area at the historical time, and the distance between the historical time and the real-time time; the movement information of the historical time is weighted and analyzed by using the influence degree, the predicted movement information of each vehicle area is obtained, and the predicted movement information of each vehicle area is adjusted by using the movement information of the preceding vehicle at the real-time time.

[0067] The intelligent traffic not only needs to effectively collect the motion information of the vehicle in the tunnel, but also needs to effectively predict the motion trend at the future time, so as to avoid the communication time delay problem and assist the terminal to make a management response in time. The information prediction can be combined with the historical time information for statistical integration, and then the motion trend at the future time is analyzed. However, for the historical time, the influence degree of the motion information at the real time is different, which is specifically affected by the point cloud signal quality, the moving information difference between the two times, and the time distance and other factors. Therefore, in the embodiment of the present application, the interference degree of each vehicle area is obtained according to the Doppler velocity stability of the vehicle area at the corresponding time; the influence degree of each historical time is obtained according to the moving information difference between each historical time and the real time, the interference degree of the vehicle area at the historical time, and the distance between the historical time and the real time. That is, the influence degree represents the information reference degree of the moving information of the historical time compared with the present time, so that the moving information of each vehicle area can be obtained by using the influence degree to analyze the moving information of the historical time. Further considering that the distance between the vehicle area and the front vehicle at the real time will affect the moving speed of the vehicle at the future time, the predicted moving information of each vehicle area can be adjusted by combining the moving information of the front vehicle at the real time. Thus, accurate and effective prediction information is obtained.

[0068] Preferably, the interference degree obtaining method in the embodiment of the present application comprises:

[0069] The Doppler velocity range of all point clouds in each vehicle area is obtained, and the ratio of the Doppler velocity range to the number of point clouds in the vehicle area is normalized to obtain the interference degree. That is, the more uneven the Doppler velocity distribution of the point cloud data in the vehicle area at a time, the greater the fluctuation of the point cloud detection result of the same vehicle, and the greater the interference degree; the smaller the number of point clouds, the poorer the quality of the point clouds, and the greater the interference degree.

[0070] Preferably, in the embodiment of the present application, the influence degree obtaining method comprises:

[0071] For each historical time, the direction similarity in the moving direction between the historical time and the real time of the vehicle area is obtained; and the moving speed difference between the moving information, the untrustworthiness of the historical time is obtained according to the moving speed difference, the time sequence distance between the historical time and the real time, and the interference degree of the vehicle area at the historical time. That is, the greater the time sequence distance, the greater the interference degree, and the greater the moving speed difference, the smaller the information reference of the historical time, and the greater the untrustworthiness. The ratio of the direction similarity to the untrustworthiness is taken as the influence degree.

[0072] In the embodiment of the present application, similar to the method for obtaining the matching index, the influence degree is obtained by multiplying the moving speed difference, the time sequence distance between the historical time and the real time, and the interference degree of the vehicle area at the historical time, and multiplying the direction similarity. The specific negative correlation mapping method is not described again.

[0073] It should be noted that the direction similarity can be obtained by using the method for obtaining the cosine similarity, converting the moving directions of the vehicle area at the historical time and the real time into unit vectors, and calculating the cosine similarity between the unit vectors. The moving speed difference is the absolute value of the difference between the moving speeds.

[0074] Preferably, in the embodiment of the present application, the method for obtaining the predicted moving information comprises:

[0075] For each historical time, the moving direction of the vehicle area at each historical time is converted into a unit vector, and the unit vectors of all historical times are weighted and fused by using the influence degree to obtain the predicted moving direction in the predicted moving information. That is, the direction of the weighted and fused vector is taken as the predicted moving direction.

[0076] The moving speed in the moving information at the real time is adjusted according to the moving speed increment of the real time relative to the previous time to obtain the predicted moving speed in the predicted moving information. That is, the sum of the moving speed at the real time and the moving speed increment is taken as the predicted moving speed. The moving speed increment is the difference between the moving speed at the real time and the moving speed at the previous time.

[0077] Preferably, in the embodiment of the present application, the predicted moving information of each vehicle area is adjusted by using the moving information of the front vehicle at the real time, comprising:

[0078] If there is no front vehicle in the vehicle area at the real time, the predicted moving information of the vehicle area is not adjusted.

[0079] If there is a front vehicle in the vehicle area at the real time, and the distance is less than the safety distance specified by the tunnel, the predicted moving speed in the predicted moving information is reduced according to the distance between the vehicle area and the front vehicle at the real time, and the difference between the moving speeds in the moving information of the vehicle area and the front vehicle. In the embodiment of the present application, the excess distance is obtained by subtracting the distance between the vehicle area and the front vehicle from the safety distance specified by the tunnel. The difference between the predicted moving speed of the vehicle area and the moving speed of the front vehicle is obtained, the difference is multiplied by the excess distance to obtain a danger degree after normalization. The difference between the positive integer 1 and the danger degree is taken as an adjustment coefficient, and the product of the adjustment coefficient and the predicted speed is taken as the adjusted predicted moving speed.

[0080] To sum up, the embodiment of the present application matches the initial point cloud data of different radars by using a matching method, determines the information weight value corresponding to each radar in combination with the point cloud quality, and further determines the effective point cloud data. The point cloud velocity is further weighted and fused with the Doppler velocity to obtain the moving information of each vehicle area. The interference degree of the vehicle area at each time is determined, and the influence degree of each historical time can be determined in combination with the information between the historical time and the real time. The predicted moving information is determined in combination with the influence degree, and the moving information of the preceding vehicle is used for adjustment. The present application eliminates the influence of the multipath effect in the tunnel, and realizes the effective speed measurement and accurate prediction of each vehicle.

[0081] It should be noted that the above-mentioned order of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or can be advantageous.

[0082] Each of the embodiments in the specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each embodiment mainly describes the difference from other embodiments.

Claims

1. A high-precision radar speed measurement method in a tunnel for smart transportation, characterized in that: The method comprises: Multiple radars are installed at monitoring locations within the tunnel to obtain initial point cloud data acquired by each radar at the time of acquisition. Based on the degree of matching between the initial point clouds and the point cloud quality of each initial point cloud, an information weight for each initial point cloud data is obtained. All initial point cloud data are registered using the information weight to obtain valid point cloud data. Obtain the vehicle area in the valid point cloud data, obtain the point cloud velocity of each vehicle area at each moment through the movement changes of the vehicle area at adjacent moments, perform weight analysis using the point cloud density, and perform weighted fusion of the point cloud velocity and Doppler velocity to obtain the movement information of each vehicle area at real time; For each vehicle area, the interference degree of each vehicle area is obtained based on the Doppler velocity stability of the vehicle area at the corresponding moment; the influence degree of each historical moment is obtained based on the difference in movement information between each historical moment and the real-time moment, the interference degree of the vehicle area at the historical moment, and the distance between the historical moment and the real-time moment. The movement information of the historical moment is weightedly analyzed using the said influence degree to obtain the predicted movement information of each vehicle area, and the predicted movement information of each vehicle area is adjusted using the movement information of the preceding vehicle at the real-time moment.

2. The high-precision radar speed measurement method in a tunnel for smart transportation according to claim 1 is characterized in that: The method for obtaining the matching degree includes: For any point cloud in any initial point cloud data, the point cloud is matched with other point clouds in other initial point cloud data within a preset neighborhood range, and whether the point cloud is a stable feature point is determined based on the matching result; the proportion of stable feature points in the initial point cloud data is used as the matching degree of each initial point cloud data.

3. The high-precision radar speed measurement method in a tunnel for smart transportation according to claim 2 is characterized in that: The step of determining whether the point cloud is a stable feature point according to the matching result includes: Any initial point cloud data is used as the target initial point cloud data, any point cloud in the target initial point cloud data is used as the target point cloud, and other point clouds of other initial point cloud data within the preset neighborhood of the target point cloud are used as point clouds to be matched; the cosine similarity of the FPFH vector between the target point cloud and the point cloud to be matched is obtained, and the matching index between the target point cloud and the point cloud to be matched is obtained by combining the Euclidean distance between the target point cloud and the point cloud to be matched, as well as the difference in Doppler velocity; the point cloud to be matched with a matching index greater than a preset threshold is used as the matching point cloud of the target point cloud; if there is a matching point cloud of the target point cloud in all other initial point cloud data, the target point cloud is a stable feature point.

4. The high-precision radar speed measurement method in a tunnel for smart transportation according to claim 1 is characterized in that: The method for obtaining the point cloud quality includes: The product of point cloud density and average echo intensity is taken as the point cloud quality of an initial point cloud data.

5. The high-precision radar speed measurement method in a tunnel for smart transportation according to claim 1 is characterized in that: The method for obtaining the vehicle area includes: The DBSCAN clustering algorithm is used to cluster the point clouds in the valid point cloud data, and each cluster obtained is a vehicle area.

6. The high-precision radar speed measurement method in a tunnel for smart transportation according to claim 1 is characterized in that: The method for obtaining the interference degree includes: The Doppler velocity range difference between all point clouds in each vehicle area is obtained, and the ratio of the Doppler velocity range difference to the number of point clouds in the vehicle area is normalized to obtain the interference degree.

7. The high-precision radar speed measurement method in a tunnel for smart transportation according to claim 1, characterized in that: The obtaining of the real-time movement information of each vehicle area includes: Normalizing the point cloud density of the valid point cloud data to obtain a point cloud density weight; using the point cloud density weight as the weight of the point cloud velocity; subtracting the point cloud density weight from the positive integer 1 as the weight of the Doppler velocity; performing weighted fusion on the point cloud velocity and the Doppler velocity to obtain the moving velocity of each vehicle area in the moving information at the real time; The point cloud movement direction of the vehicle area between adjacent moments is used as the movement direction in the movement information.

8. The high-precision radar speed measurement method in a tunnel for smart transportation according to claim 1 is characterized in that: The method for obtaining the impact degree includes: For each historical moment, the directional similarity in the movement direction of the vehicle area between the historical moment and the real-time moment is obtained; as well as the movement speed difference between the movement information, the unreliability of the historical moment is obtained based on the movement speed difference, the temporal distance between the historical moment and the real-time moment, and the degree of interference of the vehicle area at the historical moment; the ratio of the directional similarity to the unreliability is used as the degree of influence.

9. The high-precision radar speed measurement method in a tunnel for smart transportation according to claim 1, characterized in that: The method for obtaining predicted movement information includes: For each historical moment, convert the movement direction of the vehicle area at each historical moment into a unit vector, and use the influence degree to weight the unit vectors of all historical moments to obtain the predicted movement direction in the predicted movement information; The moving speed in the movement information at the real time is adjusted according to the moving speed increment at the real time relative to the previous time to obtain the predicted moving speed in the predicted movement information.

10. The high-precision radar speed measurement method in a tunnel for smart transportation according to claim 1, characterized in that: The method of adjusting the predicted movement information of each vehicle area using the real-time movement information of the preceding vehicle includes: If there is no preceding vehicle in the vehicle area at the real time, the predicted movement information of the vehicle area will not be adjusted; If there is a preceding vehicle in the vehicle area at the real time moment and the distance is less than the safety distance specified by the tunnel, the predicted moving speed in the predicted movement information is reduced based on the distance between the vehicle area and the preceding vehicle at the real time moment and the difference between the predicted moving speed of the vehicle area and the moving speed of the preceding vehicle.

Images