CN111540201B - Vehicle queuing length real-time estimation method and system based on roadside laser radar - Google Patents
Vehicle queuing length real-time estimation method and system based on roadside laser radar Download PDFInfo
- Publication number
- CN111540201B CN111540201B CN202010328012.2A CN202010328012A CN111540201B CN 111540201 B CN111540201 B CN 111540201B CN 202010328012 A CN202010328012 A CN 202010328012A CN 111540201 B CN111540201 B CN 111540201B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- point cloud
- cloud data
- laser radar
- dimensional point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/04—Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a vehicle queuing length real-time estimation method and a system based on a road side laser radar, which are used for acquiring all position points scanned by the road side laser radar on a vehicle on a road to obtain three-dimensional point cloud data to be processed; sequentially carrying out background filtering on the three-dimensional point cloud data to be processed; clustering the three-dimensional point cloud data after background filtering; carrying out target identification on the three-dimensional point cloud data after clustering processing, and identifying vehicles on the road; performing lane recognition based on the result after the target recognition; estimating the speed of each vehicle on the road based on the three-dimensional point cloud data of the adjacent frames of the laser radar; based on the speed of each vehicle on the road, the last vehicle on each lane is determined, and the queuing length on each lane is estimated. The vehicle queuing length is estimated in real time, a new traffic control measure is executed in real time according to the vehicle queuing length, the overall vehicle passing efficiency is optimized, and the problems of vehicle congestion, queuing spreading and the like in peak periods are solved.
Description
Technical Field
The disclosure relates to the technical field of traffic engineering, in particular to a method and a system for estimating vehicle queuing length in real time based on a roadside laser radar.
Background
The statements in this section merely provide background information related to the present disclosure and may not constitute prior art.
The information of the vehicle queue length is applied to various aspects in the traffic field, such as the performance evaluation of signal crossings, adaptive signal control, travel route selection and the like. Some of these applications, such as optimal signal control and trip routing, require real-time vehicle queue length information. The existing research shows that the vehicle queuing length can be obtained in an estimation mode or a direct detection method. However, the traditional estimation method has the problems that the vehicle queuing length cannot be reflected in real time and the like. In recent years, with the development of the related technology of "intelligent transportation", in the context of vehicle and road intelligence, many scholars begin to use the technology of internet of vehicles to perform related research on the queuing length of vehicles, but the related various researches are based on the assumption that all vehicles are in the "internet of vehicles", and now, the fact is that the application rate of the technology of the internet of vehicles in the current vehicles is low, and the technology of the internet of vehicles will be in a lower application rate for a certain time in the future.
In the course of implementing the present disclosure, the inventors found that the following technical problems exist in the prior art:
although the internet of vehicles technology can indeed provide accurate and valuable information for estimating the length of a vehicle queue, the related methods are limited due to low application rate of the vehicle network and the like. In order to solve the problems, a method which does not depend on the internet of vehicles and can estimate the queuing length of the vehicles in real time with high precision must be found.
Disclosure of Invention
In order to overcome the defects of the prior art, the method and the system for estimating the vehicle queuing length in real time based on the roadside laser radar are provided; by estimating the vehicle queuing length, a new traffic control measure can be taken in real time, the overall vehicle passing efficiency of a traffic network is optimized, and the problems of vehicle congestion, queuing spreading and the like in the peak period are solved.
In a first aspect, the present disclosure provides a roadside lidar-based vehicle queue length real-time estimation method;
the method for estimating the vehicle queuing length in real time based on the roadside laser radar comprises the following steps:
acquiring all position points scanned by a road side laser radar on a vehicle on a road to obtain three-dimensional point cloud data to be processed;
sequentially carrying out background filtering on the three-dimensional point cloud data to be processed; clustering the three-dimensional point cloud data after background filtering, and dividing all points belonging to the same object into one class;
carrying out target identification on the three-dimensional point cloud data after clustering processing, and identifying vehicles on the road;
performing lane recognition based on the result after the target recognition;
estimating the speed of each vehicle on the road based on the three-dimensional point cloud data of the adjacent frames of the laser radar;
based on the speed of each vehicle on the road, the last vehicle on each lane is determined, and the queuing length on each lane is estimated.
In a second aspect, the present disclosure provides a roadside lidar-based vehicle queue length real-time estimation apparatus;
vehicle queuing length real-time estimation device based on roadside lidar includes:
an acquisition module configured to: acquiring all position points scanned by a road side laser radar on a vehicle on a road to obtain three-dimensional point cloud data to be processed;
a background filtering module configured to: sequentially carrying out background filtering on the three-dimensional point cloud data to be processed;
a clustering module configured to: clustering the three-dimensional point cloud data after background filtering, and dividing all points belonging to the same object into one class;
a target identification module configured to: carrying out target identification on the three-dimensional point cloud data after clustering processing, and identifying vehicles on the road;
a lane identification module configured to: performing lane recognition based on the result after the target recognition;
a vehicle speed estimation module configured to: estimating the speed of each vehicle on the road based on the three-dimensional point cloud data of the adjacent frames of the laser radar;
a queue length estimation module configured to: based on the speed of each vehicle on the road, the last vehicle on each lane is determined, and the queuing length on each lane is estimated.
In a third aspect, the present disclosure also provides an electronic device comprising a memory and a processor, and computer instructions stored on the memory and executed on the processor, wherein when the computer instructions are executed by the processor, the method of the first aspect is performed.
In a fourth aspect, the present disclosure also provides a computer-readable storage medium for storing computer instructions which, when executed by a processor, perform the method of the first aspect.
In a fifth aspect, the present disclosure provides a roadside lidar-based vehicle queue length real-time estimation system;
a vehicle queuing length real-time estimation system based on a roadside laser radar;
vehicle queuing length real-time estimation system based on roadside lidar includes:
the system comprises a laser radar module, a communication module and a vehicle queuing length real-time estimation device based on the roadside laser radar as described in the second embodiment;
the laser radar module is used for scanning vehicles in a scanning range, acquiring three-dimensional point cloud data, and uploading the point cloud data to a vehicle queue length real-time estimation device based on a roadside laser radar; the vehicle queuing length real-time estimation device based on the roadside laser radar is used for receiving and processing three-dimensional point cloud data collected by the laser radar; transmitting the processing result to a traffic control terminal through a communication module;
and the traffic control terminal executes new traffic control measures based on the processing result of the road side laser radar vehicle queuing length real-time estimation device.
Compared with the prior art, the beneficial effect of this disclosure is:
the method has the advantages that the estimation of the vehicle queuing length is completed on the basis of processing and analyzing the point cloud data collected by the laser radar, the problem that the vehicle queuing length cannot be estimated in real time at present is solved, and certain traffic control measures can be taken on the basis of the estimated queuing length, so that the queuing length is reduced in real time, traffic jam is relieved, and the road passing efficiency is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure and are not to limit the disclosure.
FIG. 1 is a schematic diagram of a method for estimating a vehicle queue length in real time by using a roadside lidar according to a first embodiment of the disclosure;
FIG. 2 is a schematic diagram of a system for estimating a vehicle queue length in real time by using a roadside lidar according to a first embodiment of the disclosure;
FIG. 3 is a schematic diagram illustrating a point cloud data missing process according to a first embodiment of the disclosure;
FIG. 4 is a schematic diagram illustrating a point cloud data missing due to occlusion according to a first embodiment of the disclosure;
fig. 5 is a schematic diagram illustrating point cloud data missing due to packet loss in the first embodiment of the disclosure;
FIG. 6 illustrates a first embodiment of the present disclosure for identifying a vehicle at the end of a fleet of vehicles, case 1;
FIG. 7 illustrates a confirmation fleet end vehicle, case 2, according to a first embodiment of the present disclosure;
fig. 8 is a schematic diagram of a roadside lidar mounting arrangement according to a first embodiment of the disclosure.
Detailed Description
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. 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 disclosure belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present disclosure. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
The method comprises the steps of firstly, estimating the vehicle queuing length in real time based on a roadside laser radar;
as shown in fig. 1, the method for estimating the vehicle queue length in real time based on the roadside lidar includes:
s1: acquiring all position points scanned by a road side laser radar on a vehicle on a road to obtain three-dimensional point cloud data to be processed;
s2: sequentially carrying out background filtering on the three-dimensional point cloud data to be processed; clustering the three-dimensional point cloud data after background filtering, and dividing all points belonging to the same object into one class;
carrying out target identification on the three-dimensional point cloud data after clustering processing, and identifying vehicles on the road;
performing lane recognition based on the result after the target recognition;
s3: estimating the speed of each vehicle on the road based on the three-dimensional point cloud data of the adjacent frames of the laser radar;
based on the speed of each vehicle on the road, the last vehicle on each lane is determined, and the queuing length on each lane is estimated.
Further, the laser radar is arranged on a telegraph pole beside the road and is 3-5 meters away from the ground.
Further, the three-dimensional coordinates of each position point are three-dimensional coordinates x, y, and z of the laser radar located at the origin of the cartesian coordinate system.
Further, the data file obtained after the laser radar scans the object is further explained as follows: the file is in csv format, which contains numerous points and coordinates (x, y, z) corresponding to the points, and each frame during the operation time of the lidar has a data file that all scanned objects will be re-represented in the form of a point cloud.
As one or more embodiments, the specific steps of sequentially performing background filtering on the three-dimensional point cloud data to be processed include:
s201: acquiring three-dimensional point cloud data when no vehicle exists on a road in a set time period;
s202: dividing the vehicle-free three-dimensional point cloud data by utilizing cubes with the same size, namely performing rasterization processing, setting a point cloud density threshold value, and selecting cubes with point cloud densities larger than the point cloud density threshold value as background point cubes; storing points contained in all selected background point cubes into a matrix, wherein the matrix is regarded as a background matrix;
s203: and subtracting the background matrix from the three-dimensional point cloud data to be processed to obtain the three-dimensional point cloud data with the filtered background.
It will be appreciated that the beneficial effects of the background filtering are: and filtering out the point cloud (background point) data of other objects except for road users.
As one or more embodiments, the clustering process is performed on the three-dimensional point cloud data after background filtering, and all points belonging to the same object are classified into one type, and the specific steps include:
based on a DBSCAN algorithm, clustering processing is carried out on the three-dimensional point cloud data after background filtering, and all points belonging to the same object are clustered together.
It should be understood that the point clouds in the lidar data belong to unordered point clouds, i.e. points belonging to the same object are not clustered together. Clustering points belonging to an object together and naming the points as uniform IDs facilitates later data processing.
As one or more embodiments, the target recognition is performed on the three-dimensional point cloud data after the clustering process, and vehicles on the road are recognized; the method comprises the following specific steps:
extracting the characteristics of each clustering object from the three-dimensional point cloud data after clustering;
inputting the characteristics of each clustered object into a pre-trained random forest classifier, and outputting the category of the current object; and obtaining the length of each vehicle body on the road according to the category of the current object and the clustering result.
Further, the extracting the feature of each cluster object includes:
and extracting the length, the height, the ratio of the length to the height of each cluster object, the distance between the current cluster object and the laser radar, the number of points included in the current cluster object and the outline of the current cluster object.
Further, the training process of the pre-trained random forest model comprises:
and extracting the characteristics of the known object class labels, and inputting the characteristics of the known object class labels into the random forest model for training to obtain the trained random forest model.
It should be understood that after point cloud clustering, there are various road users (cars, bicycles, electric vehicles, pedestrians, etc.) on the road, and these objects need to be classified in order to better estimate the vehicle queue length. By selecting the characteristics of 6 objects (target length, target height, difference between target height and target length, distance from the lidar, number of points, and target profile), a target classifier (which is formed by machine learning training using an RF (random forest) classifier) is established, and then target classification is performed.
As one or more embodiments, the lane recognition based on the result after the target recognition specifically includes:
and setting a point cloud density threshold value, and regarding the area of the three-dimensional point cloud data with the point cloud density larger than the point cloud density threshold value as a lane.
After the target classification step is completed, users only have vehicles on the road, and it is clear that the estimated vehicle queuing length is specific to a certain lane, so that lanes in the scanning range of the laser radar need to be distinguished and identified. In the lane recognition step, it is assumed that the lane change is not performed when the vehicle approaches the intersection, that is, the vehicle travels straight, so the density of the point cloud on each lane is higher than that of the boundary line between the lanes. Based on the method, similar to a method for searching background points, the lane is identified by comparing the density of the point cloud.
As one or more embodiments, the estimating the speed of each vehicle on the road based on the three-dimensional point cloud data of the adjacent frames of the laser radar specifically comprises:
tracking the same vehicle in a plurality of frame data scanned by the laser radar, selecting a point closest to the laser radar in the tracked vehicle point cloud, and calculating the speed of the vehicle by using the coordinate change of the same point in adjacent frames.
It should be understood that, in selecting a point closest to the laser radar in the tracked vehicle point cloud, the speed of the vehicle can be obtained by using the coordinate change of the same point in adjacent frames, and the specific formula is as follows:
where V represents the speed of the vehicle and F represents the lidar rotational frequency in units: HZ, XiThe abscissa, Y, representing a point in the vehicle point cloud in the ith frame that is closest to the lidariThe ordinate, Z, of a point in the vehicle point cloud in the ith frame that is closest to the lidariRepresenting the vertical coordinate, X, of a point in the vehicle point cloud closest to the lidar in frame ii-1The abscissa, Y, representing a point in the vehicle point cloud in the i-1 th frame closest to the lidari-1The ordinate, Z, of a point in the vehicle point cloud in the i-1 th frame that is closest to the laser radari-1And the vertical coordinate represents a point closest to the laser radar in the vehicle point cloud in the (i-1) th frame.
Furthermore, the same vehicle is tracked in a plurality of frame data scanned by the laser radar, and a global nearest neighbor algorithm GNN is adopted.
It will be appreciated that the information of vehicle speed is critical to finding the end vehicle of the fleet, since the end vehicle may be at a low or even stationary speed, and the end vehicle of the fleet may be determined from the information of vehicle speed. Therefore, it is necessary to continuously track the same vehicle between different frames (the present disclosure uses global nearest neighbor algorithm GNN in vehicle tracking), and after the continuous tracking of a vehicle is achieved, the speed of the vehicle can be roughly estimated according to the distance traveled by the vehicle in a certain time (adjacent frames).
As one or more embodiments, the determining a last vehicle on each lane based on the speed of each vehicle on the road, and further estimating the queuing length on each lane, specifically includes:
determining a tail vehicle on each lane;
determining a tail vehicle body length on each lane;
and calculating the queuing length on each lane according to the length of the tail vehicle, the length of the tail vehicle and the lengths of other vehicles on the current lane.
Further, the determining the tail vehicle on each lane specifically includes:
when it comes tonAnd when the vehicle appears in the ith laser radar scanning image and also appears in the (i + 1) th laser radar scanning image, calculating the speed V of the nth vehicle in the ith frame and the speed V ' of the (i + 1) th frame, and if V ' < V and V ' < 5km/h, considering that the vehicle is at the end of the vehicle fleet, and otherwise, considering that the (n-1) th vehicle is at the end of the vehicle fleet.
When the nth vehicle appears in the ith frame of laser radar scanning image and disappears in the (i + 1) th frame of laser radar scanning image, acquiring the distance between the (n-1) th vehicle and the (n-2) th vehicle, and indirectly estimating the distance between the nth vehicle and the (n-1) th vehicle; assuming that the distance between the nth vehicle and the (n-1) th vehicle is equal to the distance between the (n-1) th vehicle and the (n-2) th vehicle;
and calculating the speed V of the nth vehicle at the ith frame and the speed V ' of the nth vehicle at the (i + 1) th frame, and considering that the nth vehicle is at the end of the motorcade when V ' < V and V ' < 5km/h, and otherwise considering that the (n-1) th vehicle is at the end of the motorcade.
When the nth vehicle appears in the ith frame of laser radar scanning image and disappears in the (i + 1) th frame of laser radar scanning image, the nth vehicle is considered to be blocked or data loss is caused by data loss.
After the previous operations such as background filtering, clustering and the like, the point clouds related to each vehicle on the lane are classified into one type, so that the length of each vehicle body and the distance between the vehicles can be known, and the distance between the vehicles refers to the distance between the tail of the vehicle of the previous vehicle and the head of the vehicle of the next vehicle.
The reason why which vehicle is the last vehicle is determined is that the farther the laser radar is, the less the point cloud is, the incomplete detection of the vehicle is, and the length of the vehicle body cannot be accurately determined. Assuming that each body length (except the last) is x1, x2 … xn-1, plus the determined end vehicle body length xn, the fleet length can be calculated.
Assuming that n vehicles are shared on the lane, when the nth vehicle appears in the ith frame and the (i + 1) th frame disappears, the nth vehicle may be considered to be at the end of the vehicle fleet, and the distance between the nth vehicle and the (n-1) th vehicle may be obtained by using a back-stepping method, i.e. the distance between the (n-1) th vehicle and the (n-1) th vehicle is indirectly estimated from the distance between the (n-1) th vehicle and the (n-2) th vehicle, so that the speeds (respectively denoted as V and V ') in the ith frame and the (i + 1) th frame may be calculated, and the nth vehicle may be considered to be at the end of the vehicle fleet when V ' < V and V ' < 5km/h, as shown in fig. 5, otherwise, the nth vehicle may be considered to be at the end of the vehicle fleet.
It should be understood that determining the last vehicle in the fleet is a key factor in estimating the length of the vehicle queue, as the length of the vehicle queue can also be determined when the last vehicle in the fleet is determined.
It should be understood that the position of the trailing vehicle on each lane is determined, taking into account the following two cases:
(1) as shown in fig. 3, a large truck may shade small vehicles in its adjacent lanes;
(2) as shown in fig. 4, the connection between the laser radar and the computer performing data processing is not stable enough, and data loss occurs. Both of these conditions result in a fan-shaped defect in the point cloud, which in turn prevents certain vehicles in the fleet from being seen.
Further, the method for determining the length of the vehicle body of the vehicle at the tail of the fleet comprises the following specific steps:
and when the measured length of the vehicle body at the tail of the motorcade is more than 6m, the measured length is used as the length of the vehicle body at the tail of the motorcade. If the measured length of the last vehicle of the fleet is less than 6m, the default length of the last vehicle of the fleet is 6 m.
The reason for determining the length of the vehicle body at the end of the fleet is that when the road is blocked and the vehicles are queued up too long, the last vehicle in the fleet may not be fully detected (the point cloud density is lower as the distance from the lidar is greater, i.e., the point cloud does not present the full view of the vehicle), as shown in fig. 7.
Therefore, the measured vehicle data at the end of the fleet can be chosen as follows:
(1) when the speed of the vehicle is below 5km/h, the vehicle can be considered to be at the tail of the motorcade;
(2) it is known that the length of the small car is 6m or less, and when the measured length of the car body is greater than 6m, the measured length is regarded as the length of the car body. If the measured length is less than 6m, the default length is 6m, which is summarized as follows:
according to the method and the system for estimating the vehicle queuing length in real time by using the roadside lidar, the vehicle information on the road is acquired in real time, the point cloud data with high precision is generated, a series of processing is performed on the point cloud data, the vehicle queuing length is estimated, and the traffic control terminal dynamically executes new traffic control measures according to the point cloud data, so that the road traffic jam is relieved, and the road traffic efficiency is improved. At present, no method and means for effectively acquiring the vehicle queuing length in real time exist, and the embodiment of the invention can make up for the application blank.
The embodiment provides a vehicle queuing length real-time estimation device based on a roadside laser radar;
vehicle queuing length real-time estimation device based on roadside lidar includes:
an acquisition module configured to: acquiring all position points scanned by a road side laser radar on a vehicle on a road to obtain three-dimensional point cloud data to be processed;
a background filtering module configured to: sequentially carrying out background filtering on the three-dimensional point cloud data to be processed;
a clustering module configured to: clustering the three-dimensional point cloud data after background filtering, and dividing all points belonging to the same object into one class;
a target identification module configured to: carrying out target identification on the three-dimensional point cloud data after clustering processing, and identifying vehicles on the road;
a lane identification module configured to: performing lane recognition based on the result after the target recognition;
a vehicle speed estimation module configured to: estimating the speed of each vehicle on the road based on the three-dimensional point cloud data of the adjacent frames of the laser radar;
a queue length estimation module configured to: based on the speed of each vehicle on the road, the last vehicle on each lane is determined, and the queuing length on each lane is estimated.
In a third embodiment, the present embodiment further provides an electronic device, which includes a memory, a processor, and computer instructions stored in the memory and executed on the processor, where the computer instructions, when executed by the processor, implement the method in the first embodiment.
In a fourth embodiment, the present embodiment further provides a computer-readable storage medium for storing computer instructions, and the computer instructions, when executed by a processor, implement the method of the first embodiment.
Fifth, the present embodiment provides a vehicle queue length real-time estimation system based on a roadside lidar; as shown in fig. 2, the vehicle queuing length real-time estimation system based on the roadside lidar;
vehicle queuing length real-time estimation system based on roadside lidar includes:
the system comprises a laser radar module, a communication module and a vehicle queuing length real-time estimation device based on the roadside laser radar as described in the second embodiment;
the laser radar module is used for scanning vehicles in a scanning range, acquiring three-dimensional point cloud data, and uploading the point cloud data to a vehicle queue length real-time estimation device based on a roadside laser radar; the vehicle queuing length real-time estimation device based on the roadside laser radar is used for receiving and processing three-dimensional point cloud data collected by the laser radar; transmitting the processing result to a traffic control terminal through a communication module;
and the traffic control terminal executes new traffic control measures based on the processing result of the road side laser radar vehicle queuing length real-time estimation device.
The lidar module includes: the laser radar, the processor and the data transmission line are connected in sequence.
Further, as shown in fig. 8, laser radar (rotary laser radar, solid laser radar and the like) is installed on a roadside telegraph pole or a signal pole, is 3-5 meters away from the ground, can be installed in modes such as a lift truck and a ladder, and is selected according to actual conditions. The lower part of the pole, approximately 1m from the ground, is equipped with a processor with data processing algorithm, which is connected with the data transmission line and covered with a steel box to protect the pole from the surrounding environment. Laser radar and treater all need the power supply, so the mode of the inside line of walking of accessible pole satisfies the power supply demand. In order to transmit the result processed by the processor, an optical fiber is connected to the processor.
And the traffic control terminal receives the result of the drive test laser radar module after collecting and processing the point cloud data, analyzes the result (the data analysis unit works), and executes a new traffic control measure in real time. In the embodiment, the signal lamp is taken as an example, the time length of the traffic light at the intersection (the work of the control unit) can be reconfigured in real time, so that the queuing length of the vehicles is reduced, and the purposes of relieving traffic jam and reducing parking time are achieved.
And the information communication module is used for transmitting the information of the laser radar module and the traffic control terminal.
The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure, and various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure.
Claims (14)
1. The method for estimating the vehicle queuing length in real time based on the roadside laser radar is characterized by comprising the following steps:
acquiring all position points scanned by a road side laser radar on a vehicle on a road to obtain three-dimensional point cloud data to be processed;
sequentially carrying out background filtering on the three-dimensional point cloud data to be processed; clustering the three-dimensional point cloud data after background filtering, and dividing all points belonging to the same object into one class;
acquiring three-dimensional point cloud data when no vehicle exists on a road in a set time period;
dividing the vehicle-free three-dimensional point cloud data by utilizing cubes with the same size, namely performing rasterization processing, setting a point cloud density threshold value, and selecting cubes with point cloud densities larger than the point cloud density threshold value as background point cubes; storing points contained in all selected background point cubes into a matrix, wherein the matrix is regarded as a background matrix;
subtracting the background matrix from the three-dimensional point cloud data to be processed to obtain three-dimensional point cloud data with the filtered background;
carrying out target identification on the three-dimensional point cloud data after clustering processing, and identifying vehicles on the road;
performing lane recognition based on the result after the target recognition;
estimating the speed of each vehicle on the road based on the three-dimensional point cloud data of the adjacent frames of the laser radar;
based on the speed of each vehicle on the road, the last vehicle on each lane is determined, and the queuing length on each lane is estimated.
2. The method as claimed in claim 1, wherein the clustering process is performed on the background filtered three-dimensional point cloud data to classify all points belonging to the same object into one class, and the specific steps include:
based on a DBSCAN algorithm, clustering processing is carried out on the three-dimensional point cloud data after background filtering, and all points belonging to the same object are clustered together.
3. The method as claimed in claim 1, wherein the clustering process is performed to identify the target of the clustered three-dimensional point cloud data, and the vehicles on the road are identified; the method comprises the following specific steps:
extracting the characteristics of each clustering object from the three-dimensional point cloud data after clustering;
inputting the characteristics of each clustered object into a pre-trained random forest classifier, and outputting the category of the current object; and obtaining the length of each vehicle body on the road according to the category of the current object and the clustering result.
4. The method as claimed in claim 3, wherein said extracting the feature of each cluster object comprises:
and extracting the length, the height, the ratio of the length to the height of each cluster object, the distance between the current cluster object and the laser radar, the number of points included in the current cluster object and the outline of the current cluster object.
5. A method as claimed in claim 3, wherein the training process for the pre-trained random forest model comprises:
and extracting the characteristics of the known object class labels, and inputting the characteristics of the known object class labels into the random forest model for training to obtain the trained random forest model.
6. The method as claimed in claim 1, wherein the lane recognition is performed based on the result of the object recognition, and the steps include:
and setting a point cloud density threshold value, and regarding the area of the three-dimensional point cloud data with the point cloud density larger than the point cloud density threshold value as a lane.
7. The method of claim 1, wherein estimating the speed of each vehicle on the road based on the three-dimensional point cloud data of adjacent frames of lidar comprises:
tracking the same vehicle in a plurality of frame data scanned by the laser radar, selecting a point closest to the laser radar in the tracked vehicle point cloud, and calculating the speed of the vehicle by using the coordinate change of the same point in adjacent frames.
8. The method of claim 1, wherein said determining the last vehicle in each lane based on the speed of each vehicle on the roadway to estimate the queue length in each lane comprises:
determining a tail vehicle on each lane;
determining a tail vehicle body length on each lane;
and calculating the queuing length on each lane according to the length of the tail vehicle, the length of the tail vehicle and the lengths of other vehicles on the current lane.
9. The method of claim 8, wherein said determining the end vehicle in each lane comprises:
when the nth vehicle appears in the ith frame of laser radar scanning image and also appears in the (i + 1) th frame of laser radar scanning image, calculating the speed V of the nth vehicle in the ith frame and the speed sum V of the (i + 1) th frame′If V is′<V and V′<The vehicle is considered to be at the tail of the motorcade at 5km/h, otherwise, the n-1 th vehicle is considered to be at the tail of the motorcade;
when the nth vehicle appears in the ith frame of laser radar scanning image and disappears in the (i + 1) th frame of laser radar scanning image, acquiring the distance between the (n-1) th vehicle and the (n-2) th vehicle, and indirectly estimating the distance between the nth vehicle and the (n-1) th vehicle; assuming that the distance between the nth vehicle and the (n-1) th vehicle is equal to the distance between the (n-1) th vehicle and the (n-2) th vehicle; calculating the speed V of the nth vehicle in the ith frame and the speed V of the nth vehicle in the (i + 1) th frame′When V is′<V and V′<And 5km/h considers that the nth vehicle is at the tail of the motorcade, and conversely, considers that the (n-1) th vehicle is at the tail of the motorcade.
10. The method of claim 8, wherein determining the length of the last vehicle body of the fleet comprises:
when the measured length of the vehicle body of the vehicle at the tail of the motorcade is more than 6m, the measured length is used as the length of the vehicle body of the vehicle at the tail of the motorcade; if the measured length of the last vehicle of the fleet is less than 6m, the default length of the last vehicle of the fleet is 6 m.
11. Vehicle queuing length real-time estimation device based on roadside laser radar, characterized by includes:
an acquisition module configured to: acquiring all position points scanned by a road side laser radar on a vehicle on a road to obtain three-dimensional point cloud data to be processed;
a background filtering module configured to: sequentially carrying out background filtering on the three-dimensional point cloud data to be processed;
acquiring three-dimensional point cloud data when no vehicle exists on a road in a set time period;
dividing the vehicle-free three-dimensional point cloud data by utilizing cubes with the same size, namely performing rasterization processing, setting a point cloud density threshold value, and selecting cubes with point cloud densities larger than the point cloud density threshold value as background point cubes; storing points contained in all selected background point cubes into a matrix, wherein the matrix is regarded as a background matrix;
subtracting the background matrix from the three-dimensional point cloud data to be processed to obtain three-dimensional point cloud data with the filtered background;
a clustering module configured to: clustering the three-dimensional point cloud data after background filtering, and dividing all points belonging to the same object into one class;
a target identification module configured to: carrying out target identification on the three-dimensional point cloud data after clustering processing, and identifying vehicles on the road;
a lane identification module configured to: performing lane recognition based on the result after the target recognition;
a vehicle speed estimation module configured to: estimating the speed of each vehicle on the road based on the three-dimensional point cloud data of the adjacent frames of the laser radar;
a queue length estimation module configured to: based on the speed of each vehicle on the road, the last vehicle on each lane is determined, and the queuing length on each lane is estimated.
12. An electronic device comprising a memory and a processor, and computer instructions stored on the memory and executable on the processor, the computer instructions when executed by the processor performing the method of any of claims 1-10.
13. A computer-readable storage medium storing computer instructions which, when executed by a processor, perform the method of any one of claims 1 to 10.
14. Vehicle queuing length real-time estimation system based on roadside laser radar, characterized by includes:
a lidar module, a communication module, and the roadside lidar-based vehicle queue length real-time estimation device of claim 11;
the laser radar module is used for scanning vehicles in a scanning range, acquiring three-dimensional point cloud data, and uploading the point cloud data to a vehicle queue length real-time estimation device based on a roadside laser radar; the vehicle queuing length real-time estimation device based on the roadside laser radar is used for receiving and processing three-dimensional point cloud data collected by the laser radar; transmitting the processing result to a traffic control terminal through a communication module;
and the traffic control terminal executes new traffic control measures based on the processing result of the road side laser radar vehicle queuing length real-time estimation device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010328012.2A CN111540201B (en) | 2020-04-23 | 2020-04-23 | Vehicle queuing length real-time estimation method and system based on roadside laser radar |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010328012.2A CN111540201B (en) | 2020-04-23 | 2020-04-23 | Vehicle queuing length real-time estimation method and system based on roadside laser radar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111540201A CN111540201A (en) | 2020-08-14 |
| CN111540201B true CN111540201B (en) | 2021-03-30 |
Family
ID=71977157
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010328012.2A Active CN111540201B (en) | 2020-04-23 | 2020-04-23 | Vehicle queuing length real-time estimation method and system based on roadside laser radar |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111540201B (en) |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112085950B (en) * | 2020-08-17 | 2021-10-29 | 西安电子科技大学 | Method and system for estimating traffic state discrimination index, storage medium and application |
| CN111932943B (en) * | 2020-10-15 | 2021-05-14 | 深圳市速腾聚创科技有限公司 | Dynamic target detection method and device, storage medium and roadbed monitoring equipment |
| CN113176547B (en) * | 2020-10-20 | 2022-03-22 | 苏州思卡信息系统有限公司 | Method for filtering background of roadside radar in real time based on Bessel modeling |
| CN113176546B (en) * | 2020-10-20 | 2022-04-01 | 苏州思卡信息系统有限公司 | Method for filtering background of road side radar in real time based on NURBS modeling |
| CN112816959B (en) * | 2021-01-14 | 2022-05-31 | 森思泰克河北科技有限公司 | Clustering method, device, equipment and storage medium for vehicles |
| CN113343835B (en) * | 2021-06-02 | 2022-04-15 | 合肥泰瑞数创科技有限公司 | Object identification method and system suitable for emergency rescue and storage medium |
| CN113313654B (en) * | 2021-06-23 | 2023-12-15 | 上海西井科技股份有限公司 | Laser point cloud filtering denoising method, system, equipment and storage medium |
| CN113447953B (en) * | 2021-06-29 | 2022-08-26 | 山东高速建设管理集团有限公司 | Background filtering method based on road traffic point cloud data |
| CN113345237A (en) * | 2021-06-29 | 2021-09-03 | 山东高速建设管理集团有限公司 | Lane-changing identification and prediction method, system, equipment and storage medium for extracting vehicle track by using roadside laser radar data |
| CN113903179B (en) * | 2021-09-30 | 2022-07-26 | 山东大学 | Using method of multi-beam laser radar background filtering device based on point cloud density superposition distribution |
| CN114093165A (en) * | 2021-11-17 | 2022-02-25 | 山东大学 | Roadside laser radar-based vehicle-pedestrian conflict automatic identification method |
| CN114812435B (en) * | 2022-04-29 | 2023-10-20 | 苏州思卡信息系统有限公司 | Vehicle three-dimensional point cloud data filtering method |
| CN115188195A (en) * | 2022-07-21 | 2022-10-14 | 同济大学 | Method and system for extracting vehicle track of urban omnidirectional intersection in real time |
| CN115512542B (en) * | 2022-09-21 | 2023-08-18 | 山东高速集团有限公司 | Track restoration method and system considering shielding based on roadside laser radar |
| CN118312807B (en) * | 2024-06-07 | 2024-08-20 | 山东大学 | Laser radar point cloud real-time clustering method and system and intelligent vehicle |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3337197B2 (en) * | 1997-04-04 | 2002-10-21 | 富士重工業株式会社 | Outside monitoring device |
| JP2006302228A (en) * | 2005-04-19 | 2006-11-02 | Sigma Denki Kogyo Kk | Signal control method at intersection and traffic measuring method for each direction at intersection |
| JP5296626B2 (en) * | 2009-07-24 | 2013-09-25 | 株式会社ナビタイムジャパン | Traffic jam status display system, terminal device, traffic jam status display server, traffic jam status display device, and traffic jam status display method |
| CN105809956A (en) * | 2014-12-31 | 2016-07-27 | 大唐电信科技股份有限公司 | Method and device for obtaining queuing length of vehicles |
| CN106441150A (en) * | 2016-09-22 | 2017-02-22 | 深圳市速腾聚创科技有限公司 | Vehicle three-dimensional parameter measuring method and device |
| CN107564285A (en) * | 2017-08-29 | 2018-01-09 | 南京慧尔视智能科技有限公司 | Vehicle queue length detection method and system based on microwave |
| CN107705563A (en) * | 2017-11-28 | 2018-02-16 | 苏州大学 | Continuous vehicle speed detection method based on laser radar |
| CN107886725A (en) * | 2017-11-16 | 2018-04-06 | 山东交通学院 | Queue length computational methods and device based on bayonet socket data |
| CN207367369U (en) * | 2017-08-29 | 2018-05-15 | 南京慧尔视智能科技有限公司 | Vehicle queue length detecting system based on microwave |
| CN108415011A (en) * | 2018-02-08 | 2018-08-17 | 长安大学 | One kind realizing vehicle queue detection method based on multi-target tracking radar |
| CN108765974A (en) * | 2018-06-15 | 2018-11-06 | 江西路通科技有限公司 | Traffic condition monitoring equipment, monitoring method and system |
| CN109785632A (en) * | 2019-03-14 | 2019-05-21 | 济南浪潮高新科技投资发展有限公司 | A kind of magnitude of traffic flow statistical method and device |
| CN109858460A (en) * | 2019-02-20 | 2019-06-07 | 重庆邮电大学 | A kind of method for detecting lane lines based on three-dimensional laser radar |
| CN110111590A (en) * | 2019-06-04 | 2019-08-09 | 南京慧尔视智能科技有限公司 | A kind of vehicle dynamic queue length detection method |
| CN110211388A (en) * | 2019-05-27 | 2019-09-06 | 武汉万集信息技术有限公司 | Multilane free-flow vehicle matching process and system based on 3D laser radar |
| CN110379168A (en) * | 2019-06-24 | 2019-10-25 | 东南大学 | A kind of vehicular traffic information acquisition method based on Mask R-CNN |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5906773B2 (en) * | 2012-02-01 | 2016-04-20 | 富士通株式会社 | Sensor device, distance measurement method, and distance measurement program |
| CN103985251B (en) * | 2014-04-21 | 2016-03-02 | 东南大学 | A kind of method and system of vehicle queue length measuring and calculating |
| CN107591002B (en) * | 2017-09-21 | 2020-06-02 | 电子科技大学 | Real-time estimation method for highway traffic parameters based on distributed optical fiber |
-
2020
- 2020-04-23 CN CN202010328012.2A patent/CN111540201B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3337197B2 (en) * | 1997-04-04 | 2002-10-21 | 富士重工業株式会社 | Outside monitoring device |
| JP2006302228A (en) * | 2005-04-19 | 2006-11-02 | Sigma Denki Kogyo Kk | Signal control method at intersection and traffic measuring method for each direction at intersection |
| JP5296626B2 (en) * | 2009-07-24 | 2013-09-25 | 株式会社ナビタイムジャパン | Traffic jam status display system, terminal device, traffic jam status display server, traffic jam status display device, and traffic jam status display method |
| CN105809956A (en) * | 2014-12-31 | 2016-07-27 | 大唐电信科技股份有限公司 | Method and device for obtaining queuing length of vehicles |
| CN106441150A (en) * | 2016-09-22 | 2017-02-22 | 深圳市速腾聚创科技有限公司 | Vehicle three-dimensional parameter measuring method and device |
| CN207367369U (en) * | 2017-08-29 | 2018-05-15 | 南京慧尔视智能科技有限公司 | Vehicle queue length detecting system based on microwave |
| CN107564285A (en) * | 2017-08-29 | 2018-01-09 | 南京慧尔视智能科技有限公司 | Vehicle queue length detection method and system based on microwave |
| CN107886725A (en) * | 2017-11-16 | 2018-04-06 | 山东交通学院 | Queue length computational methods and device based on bayonet socket data |
| CN107705563A (en) * | 2017-11-28 | 2018-02-16 | 苏州大学 | Continuous vehicle speed detection method based on laser radar |
| CN108415011A (en) * | 2018-02-08 | 2018-08-17 | 长安大学 | One kind realizing vehicle queue detection method based on multi-target tracking radar |
| CN108765974A (en) * | 2018-06-15 | 2018-11-06 | 江西路通科技有限公司 | Traffic condition monitoring equipment, monitoring method and system |
| CN109858460A (en) * | 2019-02-20 | 2019-06-07 | 重庆邮电大学 | A kind of method for detecting lane lines based on three-dimensional laser radar |
| CN109785632A (en) * | 2019-03-14 | 2019-05-21 | 济南浪潮高新科技投资发展有限公司 | A kind of magnitude of traffic flow statistical method and device |
| CN110211388A (en) * | 2019-05-27 | 2019-09-06 | 武汉万集信息技术有限公司 | Multilane free-flow vehicle matching process and system based on 3D laser radar |
| CN110111590A (en) * | 2019-06-04 | 2019-08-09 | 南京慧尔视智能科技有限公司 | A kind of vehicle dynamic queue length detection method |
| CN110379168A (en) * | 2019-06-24 | 2019-10-25 | 东南大学 | A kind of vehicular traffic information acquisition method based on Mask R-CNN |
Non-Patent Citations (3)
| Title |
|---|
| An approach to detect and delineate street curbs from MLS 3D point cloud data;Borja Rodríguez-Cuenca;《Automation in Construction》;20150115;全文 * |
| 基于图像处理的车辆排队长度鲁棒性检测算法;李卫斌，辛乐;《计算机测量与控制》;20110831;全文 * |
| 路段车辆排队估计研究;祁宏生;《中国优秀硕士学位论文全文数据库（工程科技Ⅱ辑）》;20081015;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111540201A (en) | 2020-08-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111540201B (en) | Vehicle queuing length real-time estimation method and system based on roadside laser radar | |
| US20210302585A1 (en) | Smart navigation method and system based on topological map | |
| WO2018068653A1 (en) | Point cloud data processing method and apparatus, and storage medium | |
| CN113345237A (en) | Lane-changing identification and prediction method, system, equipment and storage medium for extracting vehicle track by using roadside laser radar data | |
| CN113421432B (en) | Traffic restriction information detection method and device, electronic equipment and storage medium | |
| CN110458050B (en) | Vehicle cut-in detection method and device based on vehicle-mounted video | |
| US12093045B2 (en) | Method and system for operating a mobile robot | |
| CN111027430A (en) | Traffic scene complexity calculation method for intelligent evaluation of unmanned vehicles | |
| CN112487905A (en) | Method and system for predicting danger level of pedestrian around vehicle | |
| CN113378751A (en) | Traffic target identification method based on DBSCAN algorithm | |
| CN114879192A (en) | Decision tree vehicle type classification method based on road side millimeter wave radar and electronic equipment | |
| CN114049610B (en) | Active discovery method for motor vehicle reversing and reverse driving illegal behaviors on expressway | |
| CN117875820B (en) | Intelligent distribution road monitoring system and method based on artificial intelligence | |
| CN114926984A (en) | Real-time traffic conflict collection and road safety evaluation method | |
| CN116704750B (en) | Traffic state identification method based on clustering algorithm, electronic equipment and medium | |
| CN116913081A (en) | Vehicle queuing length detection method based on roadside laser radar | |
| CN116087988A (en) | Multi-line laser radar obstacle detection method | |
| CN115797409A (en) | 3D target object positioning method and device and electronic equipment | |
| CN113284338B (en) | Method for calculating influence of motor vehicle emergency avoidance no-lamp control pedestrian crossing on traffic flow | |
| CN112258881B (en) | Vehicle management method based on intelligent traffic | |
| CN115421160A (en) | Road edge detection method, device, equipment, vehicle and storage medium | |
| Kaur et al. | RFID based Intelligent Transport System with RSU Communication for Emergency Vehicles in Urbanization | |
| CN113674329A (en) | Vehicle driving behavior detection method and system | |
| CN113128847A (en) | Entrance ramp real-time risk early warning system and method based on laser radar | |
| CN112802333A (en) | AI video analysis-based highway network safety situation analysis system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200930 Address after: 250002, No. 12550, East 2nd Ring Road, Shizhong District, Shandong, Ji'nan Applicant after: SHANDONG University Applicant after: Shandong high speed Group Co.,Ltd. Address before: 250002, No. 12550, East 2nd Ring Road, Shizhong District, Shandong, Ji'nan Applicant before: SHANDONG University Applicant before: QILU TRANSPORTATION DEVELOPMENT GROUP Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |