Method and system for identifying, managing and controlling vehicles running at low speed on expressway

Abstract

The invention relates to the technical field of expressway management and control and multi-source data fusion, in particular to an expressway low-speed driving vehicle identification and management and control method and system so as to realize more accurate and effective management and control of abnormal low-speed driving vehicles and surrounding vehicles and guarantee expressway traffic safety. According to the technical scheme of the invention, the method comprises the following steps: S1, carrying out the data correction and alignment of expressway vehicle positioning data and ETC portal transaction data based on a position change relation, and extracting the vehicle driving speed; s2, calculating high-speed regional space-time operation characteristics in a multi-scale time window based on a macroscopic traffic flow model; s3, analyzing the real-time vehicle positioning data flow, capturing low-speed vehicles, establishing a suspicious vehicle cache table, tracking and recording, fusing and filtering continuous low-speed driving events, and finally, manually confirming the low-speed driving vehicles to generate a low-speed driving alarm; and S4, carrying out early warning reminding on the low-speed driving vehicle and surrounding vehicles, and carrying out single-point reminding on the low-speed vehicle.

Description

Technical Field

[0001] The present invention relates to the technical field of highway control and multi-source data fusion, and in particular to a method and system for identifying and controlling low-speed vehicles on a highway. Background Art

[0002] As an important part of the modern transportation system, highways are designed to provide an efficient and fast way of travel. However, in real life, "slow driving" on highways, that is, accidents caused by low-speed driving, are common and have posed a serious threat to road safety and traffic efficiency.

[0003] With the acceleration of highway informatization construction, various traffic data sources have been widely used on highways. However, due to the different working principles and working methods of detectors, each data source has its limitations. It is difficult to accurately identify low-speed vehicles based on a single data source. In recent years, some research methods have been based on vehicle terminals such as GPS / BDS to identify abnormal vehicle driving status. Not only does the reliance on and in-depth mining of single vehicle driving data put forward high requirements on the quality of vehicle data, but the vehicle data is not coupled with the road operation status, road linear status, etc., resulting in high false alarm and missed alarm rates for low-speed driving events, which makes it difficult to meet the subsequent active control needs of highways for accurately pushing reminder information for abnormal vehicles. On the other hand, highway ETC toll data is an important source for highway operation status monitoring, but it has been ignored in previous methods.

[0004] In summary, there are few existing methods for identifying low-speed vehicles on highways and they are highly dependent on the operating data of a single vehicle. How to fuse multi-source data at different time and space granularities on highways, effectively utilize the complementarity of different data in describing the status of vehicles and roads, and give early targeted warnings to abnormally low-speed vehicles on highways is an important task for highway traffic control and management. Summary of the invention

[0005] The purpose of the present invention is to overcome the deficiencies in the above-mentioned background technology and provide a method and system for identifying and controlling low-speed vehicles on highways, so as to achieve more accurate and effective control of abnormally low-speed vehicles and surrounding vehicles and ensure highway traffic safety.

[0006] The technical solution of the present invention is:

[0007] A method for identifying and controlling vehicles traveling at low speeds on a highway comprises the following steps:

[0008] S1: Based on the position change relationship, the highway vehicle positioning data and ETC gantry transaction data are corrected and aligned, and the vehicle speed is extracted;

[0009] S2: Calculate the spatiotemporal operation characteristics of high-speed sub-regions within a multi-scale time window based on a macro traffic flow model;

[0010] S3: Analyze the real-time vehicle positioning data stream, capture low-speed vehicles, establish a suspected vehicle cache table and track records, integrate and filter continuous low-speed driving events, and manually confirm low-speed driving vehicles to generate low-speed driving alarms;

[0011] S4: Provide early warning reminders to low-speed vehicles and their surrounding vehicles, and provide single-point reminders to low-speed vehicles.

[0012] In step S1: the vehicle positioning data includes the real-time latitude and longitude coordinates of the vehicle, vehicle information, and a timestamp; the ETC gantry transaction data includes a timestamp, gantry information, and vehicle information.

[0013] In step S1: the correction is the correction processing of repeated and abnormal data; the alignment is the alignment of vehicle positioning data and ETC gantry transaction data to ensure the consistency of timestamps.

[0014] In step S2: the multi-scale time window is a time unit of different time lengths; the high-speed regional spatiotemporal operation characteristics include the average speed of traffic flow in the road section.

[0015] In the step S3: analyzing the real-time vehicle positioning data stream includes screening the vehicles traveling within the required road section based on the real-time vehicle positioning data in combination with the static information data of the road network, and converting the original latitude and longitude coordinate data of the vehicle into the corresponding road stake number, while assisting with the ETC gantry transaction data to perform fixed-point and timed calibration on the vehicle positioning data with large errors, so as to accurately track the vehicles traveling on the road.

[0016] In step S3: the establishment of the suspected vehicle cache table includes: the average driving speed of the vehicle Average speed of traffic The absolute value of the difference is continuously higher than the speed difference threshold v y A suspected vehicle cache table is established for low-speed vehicles; for vehicles in the suspected vehicle cache table that have been in the range of t j If a vehicle is found to be driving at a low speed for more than 30 minutes, a low-speed driving or stopped driving alarm will be reported, and the alarm information will be continuously updated at a certain time interval; when the vehicle resumes normal driving, or the driving trajectory data of the vehicle is not collected within the threshold time ΔT, the vehicle will be deleted from the suspected vehicle cache table.

[0017] In step S3: the fusion and filtering include: for low-speed driving behaviors near road infrastructure such as longitudinal slopes, service areas, interchange hubs, etc., it is considered that they do not constitute low-speed driving alarms; the monitoring video detection data and the manually entered event data are integrated to filter the alarm reports around congestion and construction sections; for other low-speed driving alarms within one kilometer before and after the low-speed alarm vehicle, other low-speed driving alarms are integrated and filtered.

[0018] In step S4: the early warning reminder includes information board reminder, directional horn reminder, and flashing light reminder; the single point reminder includes vehicle terminal reminder and APP reminder.

[0019] The step S4 includes: directly issuing single-point message reminders to low-speed vehicles through the vehicle-mounted terminal and APP; issuing warning information to low-speed vehicles on the information board on the road section within 5 km in front of the low-speed vehicle; issuing warning information to the rear vehicles on the information board on the road section within 5 km behind the low-speed vehicle; issuing warning voice to vehicles within 5 km in front and behind the low-speed vehicle through the directional speaker; issuing flashing instructions to the flashing lights on the road section within 5 km in front and behind the low-speed vehicle.

[0020] A system for identifying and controlling vehicles traveling at low speeds on highways is characterized by comprising a memory and a processor, wherein the memory is used to store a computer program; and the processor is used to implement the method for identifying and controlling vehicles traveling at low speeds on highways when executing the computer program.

[0021] The beneficial effects of the present invention are:

[0022] (1) The present invention proposes a method for identifying and controlling low-speed vehicles on highways, which integrates multi-source heterogeneous data (highway vehicle positioning data, ETC gantry transaction data, surveillance video detection data, and manually entered event data), breaking the existing technology's excessive reliance on a single data source, improving the perception accuracy of road conditions, and avoiding the abnormal low-speed vehicle identification caused by road section congestion;

[0023] (2) The present invention further compares and verifies the abnormal recognition results with the special road alignment data set. The low-speed driving behavior generated in special areas such as longitudinal slopes, service areas, and interchange hubs should not be judged as abnormal driving behavior. This can overcome the shortcomings of the traditional vehicle abnormal driving behavior recognition process that only recognizes based on the current vehicle kinematics and ignores the effect of road alignment, making the recognition result more reliable.

[0024] (3) The present invention combines ETC gantry transaction data and vehicle positioning data to construct a suspected vehicle cache table, which can accurately track the driving trajectory of abnormal vehicles and analyze their abnormal driving behavior while reducing computing resource consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a flowchart of Example 1 of the present invention.

[0026] Figure 2 This is a flowchart of low-speed vehicle abnormality determination according to the first embodiment of the present invention.

[0027] Figure 3 It is a schematic diagram of a road control system page according to Example 1 of the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0029] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0030] Example 1

[0031] The implementation section of this embodiment is located on a certain expressway section in Zhejiang Province, which has special road lines such as ramp entrances and exits, hub intersections, and service areas.

[0032] This embodiment provides a method for identifying and controlling vehicles traveling at low speeds on a highway, comprising the following steps:

[0033] S1: Correct and align the highway vehicle positioning data and ETC gantry transaction data based on the position change relationship, and extract the vehicle speed;

[0034] The vehicle positioning data includes the real-time latitude and longitude coordinates of the vehicle, vehicle information, and timestamp;

[0035] The ETC gantry transaction data is a record of vehicles passing through the ETC gantry at different time points. The ETC gantry transaction data includes a timestamp, gantry information, and vehicle information.

[0036] The correction is the correction of repeated and abnormal data. Among them: for repeated data (including vehicle positioning data and ETC gantry data), the latest data is retained or repeated data is removed according to business needs; for abnormal data, abnormal vehicle positioning data is detected based on the position change relationship, and abnormal ETC gantry transaction data is detected based on the effective range of time and space, and finally the data is eliminated or corrected according to the abnormal type.

[0037] The alignment is the alignment of vehicle positioning data and ETC gantry transaction data, ensuring the consistency of timestamps and realizing the association between vehicle positioning data and ETC gantry transaction data.

[0038] The vehicle speed information is the average vehicle speed calculated based on the vehicle positioning data of adjacent points of the same vehicle. It can be regarded as the instantaneous speed of the vehicle at the last fixed point.

[0039] S2: Calculate the spatiotemporal operation characteristics of high-speed sub-regions within a multi-scale time window based on a macro traffic flow model;

[0040] The multi-scale time window is a time unit of different time lengths, which can be selected as 5 minutes, 1 hour, etc.

[0041] The spatial and temporal characteristics of highway operation in different regions include the average speed of traffic flow in the section. The road section is divided into sections at fixed intervals (such as 1 km).

[0042] Since the distribution of ETC gantries on highways is uneven, it is necessary to estimate the speed of highway sections based on ETC gantries transaction data and vehicle positioning data to obtain the average speed of traffic in the road section. Speed ​​estimation is achieved by building a macro traffic flow model, which is calibrated and solved using vehicle positioning data and ETC gantry transaction data.

[0043] The construction of the macro traffic flow model includes: using the Greenshields model as the basis for the fitting model, obtaining the fitting model parameters according to the historical traffic flow basic parameters (including traffic flow rate, vehicle speed, and road occupancy) of the corresponding gantry interval collected by the ETC gantry, and then establishing a macro traffic flow model based on the ETC gantry transaction data. In addition, when establishing a macro traffic flow model, it is necessary not only to pay attention to the historical real data collected by the ETC gantry, but also to consider the characteristics of traffic flow data changing over time, and input model factors such as the weather conditions on the day of data collection, so as to establish a higher-precision macro traffic flow model.

[0044] The macro traffic flow model is an existing model.

[0045] Calibration and solution of macro traffic flow model using vehicle positioning data and ETC gantry data include:

[0046] First, the vehicle positioning data is collected and sliced ​​according to the road section interval. After slicing, the average speed in the corresponding time is calculated. The time and space synchronization of the data must be strictly guaranteed during the data processing process. The average vehicle speed of each road section (road section interval) and time period is obtained, and the speed correction coefficient between the vehicle positioning data and the ETC gantry transaction data is established. The corrected speed is further matched with the pre-established macro traffic flow model to complete the model calibration of the fusion of vehicle positioning data and ETC gantry transaction data, thereby obtaining the speed estimate of each road section interval, that is, the average speed of the traffic flow.

[0047] S3: Analyze the real-time vehicle positioning data stream, capture low-speed vehicles, establish a suspected vehicle cache table and track records, integrate and filter continuous low-speed driving events, and manually confirm low-speed driving vehicles to generate low-speed driving alarms;

[0048] Analyzing the real-time vehicle positioning data stream includes: based on the real-time vehicle positioning data, combined with the static information data of the road network, screening the vehicles traveling within the required section (the highway is divided at fixed intervals of 1km, and usually a highway can form dozens of sections), and converting the original latitude and longitude coordinate data of the vehicle into the corresponding road stake number (with an accuracy of meters), and at the same time, assisted by the ETC gantry transaction data, performing fixed-point and timed calibration on the vehicle positioning data with large errors, so as to accurately track vehicles traveling on the road.

[0049] The establishment of the suspected vehicle cache table includes: the average driving speed of the vehicle Average speed of traffic The absolute value of the difference is continuously higher than the speed difference threshold v y A suspected vehicle cache table is established for low-speed vehicles; for vehicles in the suspected vehicle cache table that have been in the range of t j If a vehicle is found to be driving at a low speed for more than 30 minutes, a low-speed driving or stopped driving alarm will be reported, and the alarm information will be continuously updated at a certain time interval; when the vehicle resumes normal driving, or the driving trajectory data of the vehicle is not collected within the threshold time ΔT, the vehicle will be deleted from the suspected vehicle cache table.

[0050] The speed difference threshold v y It is obtained from actual road conditions and accident investigation and analysis, and is generally taken as 20-40km / h.

[0051] The driving trajectory data of the vehicles in the suspected vehicle cache table needs to be tracked and stored for a long time to facilitate later verification and analysis. The vehicle driving trajectory data is the gantry passage information during the vehicle driving process and the position and speed information at different time points.

[0052] Since the vehicle driving status is related to complex road conditions, it is necessary to fuse and filter low-speed driving events to screen out abnormal low-speed driving behaviors and issue an alarm.

[0053] Fusion and filtering include:

[0054] 1) Low-speed driving near road infrastructure such as longitudinal slopes, service areas, and interchange hubs is not considered to constitute a low-speed driving alarm;

[0055] 2) Integrate surveillance video detection data with manually entered event data to filter alarm reports around congested and construction sections;

[0056] 3) For other low-speed driving alarms within one kilometer before and after the low-speed alarm vehicle, fusion filtering is performed.

[0057] Manual confirmation of slow-moving vehicles means confirming slow-moving vehicles by manually checking real-time surveillance videos.

[0058] S4: Provide early warning reminders to low-speed vehicles and their surrounding vehicles, and provide single-point reminders to low-speed vehicles.

[0059] The early warning reminder includes information board reminder, directional horn reminder, and flashing light reminder. The single point reminder includes vehicle terminal reminder and APP reminder.

[0060] Specifically include:

[0061] 1) Send single-point message reminders to low-speed vehicles directly through the vehicle terminal and APP.

[0062] 2) Issue warning information to slow-moving vehicles on information boards within 5 km ahead of slow-moving vehicles;

[0063] 3) Issue warning information to the vehicles behind on the information board within 5km behind the low-speed vehicle;

[0064] 4) Use directional speakers to issue warning voice to vehicles within 5km in front and behind the slow-moving vehicle;

[0065] 5) Issue flashing instructions for flashing lights on roads within 5km in front and behind of low-speed vehicles.

[0066] When the vehicle is in a state of continuous low-speed driving, the road control system should automatically issue control commands to the roadside equipment within the newly entered section based on the above alarm content, and cancel the roadside equipment warning for the exit section.

[0067] According to the results of field tests, from August 21 to 29, 2024, there were 253 alarms for 5-minute slow driving. 247 of them were confirmed as slow driving events after manual verification by traffic police, with an identification accuracy rate of about 97.5%, achieving good application accuracy.

[0068] This embodiment provides a system for identifying and controlling vehicles traveling at low speeds on highways, including a memory and a processor, wherein the memory is used to store a computer program; and the processor is used to implement the method for identifying and controlling vehicles traveling at low speeds on highways when executing the computer program.

[0069] The present invention integrates vehicle positioning data at different time and space granularities with road ETC gantry transaction data, and predicts road conditions based on a macro traffic flow model. This can avoid the limitations of inaccurate abnormality identification caused by lack of road conditions based on single vehicle positioning data, thereby achieving more accurate and effective control of abnormally low-speed vehicles and surrounding vehicles, and ensuring highway traffic safety.

[0070] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for identifying and controlling vehicles traveling at low speeds on a highway, comprising the following steps: S1: Based on the position change relationship, the highway vehicle positioning data and ETC gantry transaction data are corrected and aligned, and the vehicle speed is extracted; S2: Calculate the spatiotemporal operation characteristics of high-speed sub-regions within a multi-scale time window based on a macro traffic flow model; S3: Analyze the real-time vehicle positioning data stream, capture low-speed vehicles, establish a suspected vehicle cache table and track records, merge and filter continuous low-speed driving events, and finally generate a low-speed driving alarm by manually confirming the low-speed driving vehicles; S4: Provide early warning reminders to low-speed vehicles and their surrounding vehicles, and provide single-point reminders to low-speed vehicles.

2. A method for identifying and controlling low-speed vehicles on a highway according to claim 1, characterized in that: In step S1: the vehicle positioning data includes the real-time latitude and longitude coordinates of the vehicle, vehicle information, and a timestamp; the ETC gantry transaction data includes a timestamp, gantry information, and vehicle information.

3. A method for identifying and controlling low-speed vehicles on a highway according to claim 1, characterized in that: In step S1: the correction is the correction processing of repeated and abnormal data; the alignment is the alignment of vehicle positioning data and ETC gantry transaction data to ensure the consistency of timestamps.

4. A method for identifying and controlling low-speed vehicles on a highway according to claim 1, characterized in that: In step S2: the multi-scale time window is a time unit of different time lengths; the high-speed regional spatiotemporal operation characteristics include the average speed of traffic flow in the road section.

5. The method for identifying and controlling low-speed vehicles on a highway according to claim 1, characterized in that: In the step S3: analyzing the real-time vehicle positioning data stream includes screening the vehicles traveling within the required road section based on the real-time vehicle positioning data in combination with the static information data of the road network, and converting the original latitude and longitude coordinate data of the vehicle into the corresponding road stake number, while assisting with the ETC gantry transaction data to perform fixed-point and timed calibration on the vehicle positioning data with large errors, so as to accurately track the vehicles traveling on the road.

6. A method for identifying and controlling low-speed vehicles on a highway according to claim 1, characterized in that: In step S3: the establishment of the suspected vehicle cache table includes: the average driving speed of the vehicle Average speed of traffic The absolute value of the difference is continuously higher than the speed difference threshold v y A suspected vehicle cache table is established for low-speed vehicles; for vehicles in the suspected vehicle cache table that have been continuously exceeded for t j If a vehicle is found to be driving at a low speed for more than 30 minutes, a low-speed driving or stopped driving alarm will be reported, and the alarm information will be continuously updated at a certain time interval; when the vehicle resumes normal driving, or the driving trajectory data of the vehicle is not collected within the threshold time ΔT, the vehicle will be deleted from the suspected vehicle cache table.

7. A method for identifying and controlling low-speed vehicles on a highway according to claim 1, characterized in that: In step S3: the fusion and filtering include: for low-speed driving behaviors near road infrastructure such as longitudinal slopes, service areas, interchange hubs, etc., it is considered that they do not constitute low-speed driving alarms; the monitoring video detection data and the manually entered event data are integrated to filter the alarm reports around congestion and construction sections; for other low-speed driving alarms within one kilometer before and after the low-speed alarm vehicle, other low-speed driving alarms are integrated and filtered.

8. The method for identifying and controlling low-speed vehicles on a highway according to claim 1, characterized in that: In step S4: the early warning reminder includes information board reminder, directional horn reminder, and flashing light reminder; the single point reminder includes vehicle terminal reminder and APP reminder.

9. A method for identifying and controlling low-speed vehicles on a highway according to claim 1, characterized in that: The step S4 includes: directly issuing single-point message reminders to low-speed vehicles through the vehicle-mounted terminal and APP; issuing warning information to low-speed vehicles on the information board on the road section within 5 km in front of the low-speed vehicle; issuing warning information to the rear vehicles on the information board on the road section within 5 km behind the low-speed vehicle; issuing warning voice to vehicles within 5 km in front and behind the low-speed vehicle through the directional speaker; issuing flashing instructions to the flashing lights on the road section within 5 km in front and behind the low-speed vehicle.

10. A system for identifying and controlling vehicles traveling at low speeds on a highway, characterized in that: It includes a memory and a processor, the memory is used to store a computer program; the processor is used to implement the method for identifying and controlling low-speed vehicles on highways as described in any one of claims 1 to 8 when executing the computer program.

Images