CN117037464A - Expressway fatigue driving early warning management method based on multisource data fusion analysis - Google Patents
Expressway fatigue driving early warning management method based on multisource data fusion analysis Download PDFInfo
- Publication number
- CN117037464A CN117037464A CN202310534050.7A CN202310534050A CN117037464A CN 117037464 A CN117037464 A CN 117037464A CN 202310534050 A CN202310534050 A CN 202310534050A CN 117037464 A CN117037464 A CN 117037464A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- fatigue driving
- data
- portal
- expressway
- 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.)
- Pending
Links
- 238000007726 management method Methods 0.000 title claims abstract description 45
- 238000004458 analytical method Methods 0.000 title claims abstract description 39
- 230000004927 fusion Effects 0.000 title claims abstract description 22
- 238000000034 method Methods 0.000 claims abstract description 30
- 230000006399 behavior Effects 0.000 claims description 51
- 238000010586 diagram Methods 0.000 claims description 14
- 206010039203 Road traffic accident Diseases 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 7
- 238000011144 upstream manufacturing Methods 0.000 claims description 7
- 238000011835 investigation Methods 0.000 claims description 6
- 238000011160 research Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000007774 longterm Effects 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 claims description 2
- 230000002265 prevention Effects 0.000 abstract 1
- 230000008859 change Effects 0.000 description 3
- 238000013499 data model Methods 0.000 description 3
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 2
- 239000008280 blood Substances 0.000 description 2
- 210000004369 blood Anatomy 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 239000001301 oxygen Substances 0.000 description 2
- 229910052760 oxygen Inorganic materials 0.000 description 2
- 238000011002 quantification Methods 0.000 description 2
- 240000007651 Rubus glaucus Species 0.000 description 1
- 235000011034 Rubus glaucus Nutrition 0.000 description 1
- 235000009122 Rubus idaeus Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 239000007789 gas Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003864 performance function Effects 0.000 description 1
- 208000020016 psychiatric disease Diseases 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000000284 resting effect Effects 0.000 description 1
- 231100000430 skin reaction Toxicity 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
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
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- 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
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a highway fatigue driving early warning management method and system based on multisource data fusion analysis, wherein the method comprises the following steps: acquiring traffic flow data; obtaining highway geographic information; fusing high-speed portal information and road gate information to identify potential fatigue driving vehicles; the potential fatigue driving vehicle stays at a high speed for discriminant analysis; and judging the fatigue driving vehicle, and realizing the prevention and management of the fatigue driving vehicle. The invention mainly identifies the potential fatigue driving vehicle and judges whether the vehicle stays in the service area after entering the expressway, and prompts the potential fatigue driving vehicle through the high-speed enterprise management system and the traffic police management system after confirming the potential fatigue driving vehicle, thereby improving the management capability of the expressway fatigue driving vehicle.
Description
Technical Field
The invention relates to the field of traffic management, in particular to a highway fatigue driving early warning management method and system based on multi-source data fusion analysis.
Background
After a driver continuously drives for a long time, physiological and psychological disorders occur, and unsafe factors such as inattention, reduced judgment ability, and misactions or improper operation exist. According to the method of road traffic safety, the driver is allowed to drive for more than 8 hours each day or continuously drive for 4 hours without stopping and resting for more than 20 minutes, and the fatigue driving is judged to be the fatigue driving, the design speed of the expressway is basically more than or equal to 80km/h, the running speed of the vehicle is faster, and the fatigue driving is extremely easy to cause traffic safety accidents. At present, the recognition and analysis of the fatigue driving of the expressway vehicles are mainly carried out in the city at home and abroad: and (3) installing or wearing a method for acquiring fatigue driving related parameters, a snapshot image extraction and identification method, and portal or bayonet passing time to judge that the fatigue driving behavior exists in the vehicle. At present, the factors such as high cost, low precision, incomplete travel chain and the like exist in the recognition of the fatigue driving vehicles on the expressway.
(1) Shujuan and the like propose a method for detecting the driving state of a driver based on multi-physiological signal fusion, which mainly comprises the steps of collecting physiological data such as pulse wave signals, galvanic skin response signals and the like of the driver through equipment, subjectively evaluating and analyzing the state change rule of the driver, and setting three fatigue state grades of awake, mild fatigue and severe fatigue states; tian Haixia the fatigue driving early warning system based on face recognition technology comprises the steps of equipment assembly, software platform construction, detection early warning, fatigue early warning and the like; zhan Changshu and the like propose a driver fatigue state monitor based on a pulse sensor, wherein the device comprises a pulse sensor, a raspberry group processor, a control center, a display screen, a key module, an alarm module and other modules; li Ting and the like propose a wearable fatigue driving monitor and a fatigue driving quantification method, which mainly collect blood oxygen related parameter data through the wearable fatigue driving monitor, establish a relation between blood oxygen parameters and behavior performance functions and realize quantification and classification of fatigue driving. The method for acquiring the fatigue driving related parameters by installing or wearing the instrument has the advantages of high application cost and difficult wide investment.
(2) Huang Li and the like propose to respectively obtain first, second and third fatigue values of the driver based on the change tracks of the face, neck and elbow regions in the detected image, and preset fatigue values are used for judging the fatigue state of the driver; Dongrui et al propose a driver fatigue detection method based on a multi-feature fusion state recognition network, and mainly judge whether a driver is tired or not by recognizing the gestures of eyes, mouth and head; the method for judging the fatigue driving by detecting the facial change characteristics of the driver is easily affected by factors such as image definition, weather and the like.
(3) Wang Miaoyu and the like propose to acquire the passing time of a vehicle passing through a road section based on snap shots at the expressway entrance, and judge whether the vehicle has fatigue driving behavior by judging whether the continuous running time of the vehicle is greater than a set threshold value; wang Sha and the like extract video streams and bayonet images based on a bayonet system of a public security traffic management department to obtain the passing time of a target vehicle type in a section, and judge the illegal behavior of fatigue driving by analyzing whether the target vehicle has a method of stopping for more than 4 hours or stopping for less than 20 minutes in continuous driving time; the method for judging whether the fatigue driving behavior exists only by the fact that the driving time exceeds the threshold value does not consider the traffic accident of the vehicle or the stay driving behavior of entering the service area, and the method has larger error. She Jinsong and the like propose to judge whether a target vehicle enters a service area or not by judging the time of entering, exiting and portal time of the vehicle on the expressway based on the complete traffic data on the expressway, and judge that the vehicle has fatigue driving behavior if the target vehicle does not enter the service area. The method does not consider the situation of accident stop and municipal road passing time.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention firstly provides a highway fatigue driving early warning management method based on multi-source data fusion analysis.
The invention further provides a highway fatigue driving management system based on multi-source data fusion analysis.
In order to achieve the above purpose, the technical scheme of the invention is as follows:
a highway fatigue driving early warning management method based on multisource data fusion analysis is characterized in that a traffic flow database and a geographic information database are used as bases, potential fatigue driving vehicle identification and vehicle stay discriminant analysis are relied on through data association, highway fatigue driving vehicles are comprehensively analyzed, and a scheme for realizing fatigue driving vehicle management is provided for highway operation enterprises and traffic management departments; the method comprises the following steps:
(1) Acquiring traffic flow data;
(2) Fusing traffic flow data and geographic information data;
(3) Identifying potential fatigue driving vehicles;
(4) The potential fatigue driving vehicle stays at a high speed for discriminant analysis;
(5) And judging the fatigue driving vehicle of the expressway, and realizing the management of the fatigue driving vehicle.
Preferably, the step (1) is to acquire traffic flow data, wherein the traffic flow data mainly comprises expressway flow data and road high-definition bayonet license plate identification data;
the expressway pipelining mainly comprises toll station data and ETC portal data;
the toll station data mainly comprises first-type vehicle information and first-type transaction information, wherein the first-type vehicle information comprises license plate numbers, vehicle types, in-out information, passing time and total weight of vehicles and goods, and the first-type transaction information comprises payment types, transaction amounts and charging mileage;
the ETC portal data mainly comprises second-class vehicle information and second-class transaction information, wherein the second-class vehicle information comprises license plate numbers, vehicle types, portal road sections, high speeds and passing time, and the second-class transaction information comprises traffic media and charging mileage;
the toll station data and ETC portal data can establish an association relation through ID codes and license plate numbers;
the road high-definition bayonet license plate identification data mainly comprises an bayonet name, a direction number, a lane number, an elapsed time and a license plate number;
the traffic flow data information of the traffic flow database can establish an association relationship through license plate numbers.
Preferably, the step (2) obtains expressway traffic data, road high-definition license plate identification data and expressway geographic information, wherein the expressway geographic information data mainly comprises a road network model, a service area and a toll station;
the traffic flow data information of the traffic flow database can establish a spatial association relationship with the traffic geographic information model through the toll station position and the gate position;
the road network structure model consists of geographic information files of expressways and urban roads, and comprises a closed expressway and a connected urban road network.
Preferably, the step (3) identifies that there are mainly two scenarios for the potentially fatigue-driven vehicle:
first scenario: only considering a high-speed internal driving scene, and analyzing the potential fatigue driving vehicle through high-speed portal data;
the second scenario: and (3) considering the running scenes of the urban road and the expressway, and carrying out fusion analysis on the expressway portal information, the toll gate information and the road gate information to obtain the potential fatigue driving vehicle.
Preferably, the specific implementation process of the two scene potential fatigue driving vehicle identification methods comprises the following steps:
step11: vehicle travel chain construction
(101) Constructing a single vehicle expressway travel information base; establishing a portal sequence { M ] for vehicles to pass over a highway s1 ,M s2 ,M s3 ......,M sn Time series of vehicle passing through portal { T } and s1 ,T s2 ,T s3 ......T sn (wherein n is the number of frames in the investigation region, M) sj To study the jth portal passed by the vehicle during a period of time, T sj For the time that the vehicle passes the jth portal in the research period, j epsilon (1, n);
(102) Constructing a single vehicle municipal road trip information base; extracting vehicle license plates P of vehicles entering and exiting toll stations, and establishing road high-definition bayonet sequence { X ] of vehicle path k1 ,X k2 ,X k3 ......,X kq Sum { T } through high definition Bayonet identification time series k1 ,T k2 ,T k3 ......,T kq Obtaining a travel track and travel time of the toll road passing in and out of the toll station vehicle; wherein q is the total number of bayonets passed by the vehicle in the study period, X ki To study the ith bayonet, T, passed by the vehicle during the period ki To investigate the time that the vehicle passed the ith bay in the period, i e (1, q).
Step12: vehicle identification for potential fatigue driving
The first scene can directly obtain the continuous running time of the vehicle through a vehicle expressway travel chain, and if the continuous running time exceeds 4 hours, the vehicle is judged to have possible fatigue driving behaviors;
the second scene is combined with toll gate and road gate data to obtain a travel chain of a vehicle from the municipal road to the toll gate, the travel chain of the vehicle municipal road and the expressway are fused, continuous running time of the vehicle on the municipal road and the expressway is obtained, and if the continuous running time exceeds 4 hours, the vehicle is judged to have fatigue driving behaviors possibly.
Preferably, the specific steps of the high-speed stay discriminant analysis of the potential fatigue driving vehicle in the step (4) are as follows:
step21: vehicle behavior feature discrimination based on massive historical data
(201) And constructing a single vehicle travel information base between adjacent portals of the expressway. Establishing a portal sequence { M ] for vehicles to pass over a highway s1 ,M s2 ,M s3 ......,M sn Time series of vehicle passing through portal { T } and s1 ,T s2 ,T s3 ......T sn (wherein n is the number of frames in the investigation region, M) sj To study the jth portal passed by the vehicle during a period of time, T sj For the time that the vehicle passes the jth portal in the research period, j epsilon (1, n); calculating the travel time T of a vehicle passing between adjacent door frames ab ，T ab The travel time sequence { T of the adjacent door frames for the passing of the vehicle can be obtained for the travel time difference of the vehicle passing through the door frame a and the door frame b 1_ab ,T 2_ab ,T p_ab ......,T N_ab Wherein N is the total number of all vehicles passing between the door frame a and the door frame b in the study period, T p_ab There is p e (1, n) for the travel time difference of the vehicle p through the a-th portal and the b-th portal.
(202) And judging the behavior characteristics of the vehicle. The box diagram principle in statistics is adopted as the basis for judging the vehicle behavior, and the box diagram is generally considered to be greater than the upper limit Q in the data set 3 +1.5×(Q 3 -Q 1 ) Or less than the lower limit of Q 1 -1.5×(Q 3 -Q 1 ) Outliers, where Q 1 、Q 3 The first quantile and the third quantile of the data are respectively;
the travel time sequence of each adjacent portal frame pair of the expressway is ordered from small to largeCalculating the upper limit T of the travel time of adjacent portals in service areas kk_max And lower limit T kk_min Travel time threshold range T for the ith adjacent gantry pair i_kk The calculation is as follows:
wherein: dis (Dis) i -road network distance of the ith portal pair;
v min -the lowest running speed between two portal frames of the expressway is selected according to the historical data;
v max -the highest running speed between two portal frames of the expressway, and selecting running conditions according to historical data;
according to the principle of the box diagram, the range of the travel time threshold value is determined by combining historical data among the portal frames of the expressway, so that the values of the middle, upper quartile, lower quartile, upper limit and lower limit in the box diagram are determined, the travel time threshold value experience library for forming each adjacent portal frame pair can be stored in a database as basic data, updated regularly for long-term use, and different types of vehicle behavior state analysis of any period of the adjacent portal frames of the expressway section can be supported. The relation between the travel time of the adjacent portal frame and the behavior characteristics of the vehicle is as follows:
(6) the travel time being less than the lower limit value is attributed to overspeed behavior;
(7) the travel time is greater than the lower limit and less than the lower quartile due to free flow behavior;
(8) the travel time is greater than the lower quartile and less than the upper quartile due to normal driving behavior;
(9) the travel time is greater than the upper quartile and less than the upper limit due to congestion creep behavior;
a travel time greater than the upper limit is due to the dwell behavior.
Step22: vehicle behavior feature discrimination for potential fatigue driving
And building and identifying potential fatigue driving vehicle information through a vehicle travel chain, selecting license plate numbers of the potential fatigue driving vehicles, obtaining travel time of adjacent doors on the upstream and downstream sides of a highway service area, and comparing and analyzing the travel time with a threshold experience database of the same type of vehicle travel time of the adjacent door frame pair, thereby judging the behavior characteristics of the potential fatigue driving vehicles.
Preferably, the step (5) is to judge the fatigue driving vehicle of the expressway, so as to realize the management of the fatigue driving vehicle;
vehicles with continuous driving times exceeding 4 hours are defined as potential fatigue driving vehicles, and long stay after entering a highway is generally a traffic accident or enter a service area. Whether the vehicle stays on the expressway due to traffic accidents or not can be obtained through law enforcement records, rescue records and the like; judging whether the vehicle enters the service area or not based on the travel time recorded by adjacent door frames at the upstream and downstream of the service area; and if the potential fatigue driving vehicle does not have a traffic accident and does not enter the service area, judging the potential fatigue driving vehicle as the fatigue driving vehicle.
A highway fatigue driving early warning management method and system based on multisource data fusion analysis is characterized in that the system is based on a traffic flow database and a geographic information database, and relies on potential fatigue driving vehicle identification and vehicle stay discriminant analysis through data association to comprehensively analyze the highway fatigue driving vehicles, so that a scheme for realizing accurate management of the fatigue driving vehicles is provided for highway operation enterprises and traffic management departments; the method comprises the following modules:
the traffic flow data acquisition module: the traffic flow information transmission module is used for acquiring traffic flow data;
highway geographic information module: establishing a highway geographic information model;
potential fatigue driving vehicle identification module: counting continuous driving time identification by constructing a vehicle travel chain;
the potential fatigue driving vehicle stays at a high speed to distinguish the analysis module;
and the fatigue driving vehicle judging and managing module is used for: and judging the fatigue driving vehicle, and realizing the accurate management of the fatigue driving vehicle.
The computer equipment comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the expressway fatigue driving early warning management method based on multi-source data fusion analysis when executing the computer program.
A computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements the highway fatigue driving early warning management method based on multi-source data fusion analysis.
Compared with the prior art, the invention has the beneficial effects that: the invention aims to provide a highway fatigue driving early warning management method and system based on multi-source data fusion analysis, which support the recognition of a potential fatigue driving vehicle, the stay discriminant analysis of the potential fatigue driving vehicle on a highway and the judgment of the fatigue driving vehicle, provide a scheme for a highway operation enterprise and a traffic management department for realizing the accurate management of the fatigue driving vehicle, and mainly comprise modules of data model construction, the recognition of the potential fatigue driving vehicle, the high-speed stay discriminant of the potential fatigue driving vehicle, the judgment and management of the fatigue driving vehicle and the like.
Drawings
Fig. 1 is a general framework diagram of a highway fatigue driving early warning management method based on multi-source data fusion analysis.
Fig. 2 is a flow chart for identifying potential fatigue driving vehicles constructed based on a traffic data travel chain.
Fig. 3 is a schematic view of adjacent portals (no service areas) of a highway.
Fig. 4 is a schematic view of adjacent portals (with service areas) of a highway.
Fig. 5 is a schematic view of adjacent portals (with service areas and interchange) of a highway.
Fig. 6 is a schematic diagram for distinguishing the behavior characteristics of a vehicle on a highway section where a certain service area in guangzhou is located.
Detailed Description
The invention is further described below with reference to the drawings and detailed description.
As shown in FIG. 1, the invention provides a highway fatigue driving early warning management method based on multi-source data fusion analysis, which is based on a traffic flow database and a geographic information database, and firstly relies on potential fatigue driving vehicle identification and vehicle stay discriminant analysis through data association to comprehensively analyze the highway fatigue driving vehicle, thereby providing a scheme for realizing fatigue driving vehicle management for highway operation enterprises and traffic management departments. The evaluation method comprises four modules of data model construction, potential fatigue driving vehicle identification, potential fatigue driving vehicle high-speed stay judgment, fatigue driving vehicle judgment and management, and the like, and specifically comprises the following five steps: the method comprises the steps of (1) constructing a traffic flow database; (2) fusing traffic flow data with geographic information data; (3) identifying potential fatigue driving vehicles; (4) The potential fatigue driving vehicle stays at a high speed for discriminant analysis; (5) And judging the fatigue driving vehicle of the expressway, and realizing the management of the fatigue driving vehicle.
1. Data model construction
1. And constructing a traffic flow database. Mainly comprises expressway flow data and road high-definition bayonet license plate identification data. (1) The expressway pipelining mainly comprises toll station data and ETC portal data; the toll station data mainly comprises vehicle information (license plate number, vehicle type, in-out information, passing time and total weight of vehicles and goods), transaction information (payment type, transaction amount and charging mileage); ETC portal data mainly comprises vehicle information (license plate number, vehicle type, portal road section, high speed, passing time and the like) and transaction information (passing medium, charging mileage and the like); the toll station data and ETC portal data can establish association relation through ID codes, license plate numbers and other information. (2) The road high definition bayonet license plate identification data mainly comprises an bayonet name, a direction number, a lane number, elapsed time, a license plate number and the like. The traffic flow data information of the traffic flow database can establish an association relationship through license plate numbers.
2. And constructing a geographic information model. The system mainly comprises a road network model, a service area and a toll station. (1) The road network structure model is composed of geographic information files of expressways and urban roads, and comprises two parts of closed expressways (including toll roads and toll roads) and connected urban road networks. (2) The service area and the toll station are the starting point and the finishing point of the expressway and are connected with municipal roads, the service area is a place where passengers and drivers stay for rest, and facilities such as a parking lot, a public toilet, a gas station, a vehicle repair station, a catering department and a canteen are provided, so that the toll station is an important place where vehicles of the expressway rest. (3) And the high-speed portal and the road gate position establish a spatial association relationship with the traffic geographic information model through the portal position and the gate position.
2. Vehicle identification for potential fatigue driving
There are mainly two scenarios for the identification of potential fatigue-driven vehicles:
first scenario: only considering a high-speed internal driving scene, and analyzing the potential fatigue driving vehicle through high-speed portal data;
the second scenario: considering urban road and expressway driving scenes, carrying out fusion analysis on expressway portal information, toll station information and road gate information to obtain potential fatigue driving vehicle
The implementation process of the two scene potential fatigue driving vehicle identification methods is shown in fig. 2, and the specific steps are as follows:
step1 vehicle travel chain construction
(1) Constructing a single vehicle expressway travel information base; establishing a portal sequence { M ] for vehicles to pass over a highway s1 ,M s2 ,M s3 ......,M sn Time series of vehicle passing through portal { T } and s1 ,T s2 ,T s3 ......T sn (wherein n is the number of frames in the investigation region, M) sj To study the jth portal passed by the vehicle during a period of time, T sj For the time that the vehicle passes the jth portal in the research period, j epsilon (1, n);
(2) Constructing a single vehicle municipal road trip information base; extracting vehicle license plates P of vehicles entering and exiting toll stations, and establishing road high-definition bayonet sequence { X ] of vehicle path k1 ,X k2 ,X k3 ......,X kq Sum { T } through high definition Bayonet identification time series k1 ,T k2 ,T k3 ......,T kq And obtaining the travel track and travel time of the vehicles passing in and out of the toll station and the municipal roads. Wherein q is the total number of bayonets passed by the vehicle in the study period, X ki To study the ith bayonet, T, passed by the vehicle during the period ki To investigate the time that the vehicle passed the ith bay in the period, i e (1, q).
Step2 potential fatigue driving vehicle identification
The first scene can directly obtain the continuous running time of the vehicle through a vehicle expressway travel chain, and if the continuous running time exceeds 4 hours, the vehicle is judged to have possible fatigue driving behaviors;
the second scene is combined with toll gate and road gate data to obtain a travel chain of a vehicle from the municipal road to the toll gate, the travel chain of the vehicle municipal road and the expressway are fused, continuous running time of the vehicle on the municipal road and the expressway is obtained, and if the continuous running time exceeds 4 hours, the vehicle is judged to have fatigue driving behaviors possibly.
3. Determination of high-speed stopping of potential fatigue driving vehicle
After the vehicles enter the expressway, the vehicles can generally stay in the service area, the travel time of different vehicle types between adjacent door frame pairs at the upstream and downstream of the service area is analyzed based on door frame flow data statistics, the travel time sets of adjacent door frame sections in different time periods throughout the day are obtained, different travel time thresholds of various vehicles in different time periods are calibrated through massive historical data, the travel time of the vehicles in potential fatigue driving is analyzed, and then whether the vehicles are in a stay state is judged. The specific steps of the high-speed stopping and distinguishing of the potential fatigue driving vehicle are as follows:
step1 vehicle behavior feature discrimination based on massive historical data
(1) And constructing a single vehicle travel information base between adjacent portals of the expressway. Establishing a portal sequence { M ] for vehicles to pass over a highway s1 ,M s2 ,M s3 ......,M sn Time series of vehicle passing through portal { T } and s1 ,T s2 ,T s3 ......T sn (wherein n is the number of frames in the investigation region, M) sj To study the jth portal passed by the vehicle during a period of time, T sj For the time that the vehicle passes the jth portal in the research period, j epsilon (1, n); calculating the travel of a vehicle through between adjacent door framesM T ab ，T ab The travel time sequence { T of the adjacent door frames for the passing of the vehicle can be obtained for the travel time difference of the vehicle passing through the door frame a and the door frame b 1_ab ,T 2_ab ,T p_ab ......,T N_ab Wherein N is the total number of all vehicles passing between the door frame a and the door frame b in the study period, T p_ab There is p e (1, n) for the travel time difference of the vehicle p through the a-th portal and the b-th portal.
(2) And judging the behavior characteristics of the vehicle. The box diagram principle in statistics is adopted as the basis for judging the vehicle behavior, and the box diagram is generally considered to be greater than the upper limit Q in the data set 3 +1.5×(Q 3 -Q 1 ) Or less than the lower limit of Q 1 -1.5×(Q 3 -Q 1 ) Outliers, where Q 1 、Q 3 The first and third quantiles of the data, respectively. The analysis orders the travel time sequence of each adjacent portal pair of the expressway from small to large, and calculates the upper limit T of the travel time of the adjacent portal in the service area kk_max And lower limit T kk_min Travel time threshold range T for the ith adjacent gantry pair i_kk The calculation is as follows:
wherein: dis (Dis) i -road network distance of the ith portal pair;
v min -the lowest running speed between two portal frames of the expressway is selected according to the historical data;
v max -the highest running speed between two portal frames of the expressway, and selecting running conditions according to historical data;
according to the principle of the box diagram, the range of the travel time threshold value is determined by combining historical data among the portal frames of the expressway, so that the values of the middle, upper quartile, lower quartile, upper limit and lower limit in the box diagram are determined, the travel time threshold value experience library for forming each adjacent portal frame pair can be stored in a database as basic data, updated regularly for long-term use, and different types of vehicle behavior state analysis of any period of the adjacent portal frames of the expressway section can be supported. The relation between the travel time of the adjacent portal frame and the behavior characteristics of the vehicle is as follows:
the travel time being less than the lower limit value is attributed to overspeed behavior;
the travel time is greater than the lower limit and less than the lower quartile due to free flow behavior;
the travel time is greater than the lower quartile and less than the upper quartile due to normal driving behavior;
the travel time is greater than the upper quartile and less than the upper limit due to congestion creep behavior;
travel times greater than the upper limit are due to dwell behavior.
In this embodiment, taking a pair of adjacent portals of a road section where a certain service area in Guangzhou is located as an example, selecting an adjacent portal of a certain working day of 10:00-11:00 to process and analyze data, so as to obtain a travel time sequence distribution box graph (shown in fig. 6) of the direction that the adjacent portals of a high-speed road section where the service area is located enter and leave the Guangzhou. As can be seen from fig. 6, the overall travel time bin pattern distribution in and out of the guangzhou at the same time interval has stronger similarity, wherein the median of travel time in and out of the guangzhou direction is 13.7 minutes and 12.6 minutes respectively; the upper limit for entering and exiting Guangzhou is 23.9 minutes and 23.6 minutes, respectively. Taking the exit Guangzhou direction as an example, if the travel time of a certain vehicle passing through the exit Guangzhou direction portal frame pair is 50 minutes, judging that the vehicle stays in the service area according to a relationship model of travel time and vehicle behavior characteristics, namely that the travel time is greater than an upper limit value (23.6 minutes).
Step2 potential fatigue driving vehicle behavior feature discrimination
And building and identifying potential fatigue driving vehicle information through a vehicle travel chain, selecting license plate numbers of the potential fatigue driving vehicles, obtaining travel time of adjacent doors on the upstream and downstream sides of a highway service area, and comparing and analyzing the travel time with a threshold experience database of the same type of vehicle travel time of the adjacent door frame pair, thereby judging the behavior characteristics of the potential fatigue driving vehicles.
4. Fatigue driving vehicle determination and management
Vehicles with continuous driving times exceeding 4 hours are defined as potential fatigue driving vehicles, and long stay after entering a highway is generally a traffic accident or enter a service area. Whether the vehicle stays on the expressway due to traffic accidents or not can be obtained through law enforcement records, rescue records and the like; judging whether the vehicle enters the service area or not based on the travel time recorded by adjacent door frames at the upstream and downstream of the service area; and if the potential fatigue driving vehicle does not have a traffic accident and does not enter the service area, judging the potential fatigue driving vehicle as the fatigue driving vehicle.
The embodiments of the present invention described above do not limit the scope of the present invention. Any modifications, equivalent substitutions and improvements made within the spirit principles of the present invention should be included in the scope of the claims of the present invention.
Claims (10)
1. A highway fatigue driving early warning management method based on multisource data fusion analysis is characterized in that a traffic flow database and a geographic information database are used as bases, potential fatigue driving vehicle identification and vehicle stay discriminant analysis are relied on through data association, highway fatigue driving vehicles are comprehensively analyzed, and a scheme for realizing fatigue driving vehicle management is provided for highway operation enterprises and traffic management departments; the method comprises the following steps:
(1) Acquiring traffic flow data;
(2) Fusing traffic flow data and geographic information data;
(3) Identifying potential fatigue driving vehicles;
(4) The potential fatigue driving vehicle stays at a high speed for discriminant analysis;
(5) And judging the fatigue driving vehicle of the expressway, and realizing the management of the fatigue driving vehicle.
2. The method of claim 1, wherein the step (1) is to collect and obtain traffic data, wherein the traffic data mainly comprises highway flow data and road high definition bayonet license plate identification data;
the expressway pipelining mainly comprises toll station data and ETC portal data;
the toll station data mainly comprises first-type vehicle information and first-type transaction information, wherein the first-type vehicle information comprises license plate numbers, vehicle types, in-out information, passing time and total weight of vehicles and goods, and the first-type transaction information comprises payment types, transaction amounts and charging mileage;
the ETC portal data mainly comprises second-class vehicle information and second-class transaction information, wherein the second-class vehicle information comprises license plate numbers, vehicle types, portal road sections, high speeds and passing time, and the second-class transaction information comprises traffic media and charging mileage;
the toll station data and ETC portal data can establish an association relation through ID codes and license plate numbers;
the road high-definition bayonet license plate identification data mainly comprises an bayonet name, a direction number, a lane number, an elapsed time and a license plate number;
the traffic flow data information of the traffic flow database can establish an association relationship through license plate numbers.
3. The method according to claim 2, wherein the step (2) obtains expressway traffic data, road high definition license plate recognition data, and expressway geographic information, wherein the expressway geographic information data mainly includes a road network model, a service area, and a toll station;
the traffic flow data information of the traffic flow database can establish a spatial association relationship with the traffic geographic information model through the toll station position and the gate position;
the road network structure model consists of geographic information files of expressways and urban roads, and comprises a closed expressway and a connected urban road network.
4. A method according to claim 3, wherein said step (3) identifies to the potentially tired driving vehicle that there are mainly two scenarios:
first scenario: only considering a high-speed internal driving scene, and analyzing the potential fatigue driving vehicle through high-speed portal data;
the second scenario: and (3) considering the running scenes of the urban road and the expressway, and carrying out fusion analysis on the expressway portal information, the toll gate information and the road gate information to obtain the potential fatigue driving vehicle.
5. The method of claim 4, wherein the specific implementation process of the two scene potential fatigue driving vehicle identification methods is:
step11: vehicle travel chain construction
(101) Constructing a single vehicle expressway travel information base; establishing a portal sequence { M ] for vehicles to pass over a highway s1 ,M s2 ,M s3 ......,M sn Time series of vehicle passing through portal { T } and s1 ,T s2 ,T s3 ......T sn (wherein n is the number of frames in the investigation region, M) sj To study the jth portal passed by the vehicle during a period of time, T sj For the time that the vehicle passes the jth portal in the research period, j epsilon (1, n);
(102) Constructing a single vehicle municipal road trip information base; extracting vehicle license plates P of vehicles entering and exiting toll stations, and establishing road high-definition bayonet sequence { X ] of vehicle path k1 ,X k2 ,X k3 ......,X kq Sum { T } through high definition Bayonet identification time series k1 ,T k2 ,T k3 ......,T kq And get in and outJourney track and journey time of toll station vehicles passing through municipal roads; wherein q is the total number of bayonets passed by the vehicle in the study period, X ki To study the ith bayonet, T, passed by the vehicle during the period ki I e (1, q) for the time the vehicle passes the i-th bay in the study period;
step12: vehicle identification for potential fatigue driving
The first scene can directly obtain the continuous running time of the vehicle through a vehicle expressway travel chain, and if the continuous running time exceeds 4 hours, the vehicle is judged to have possible fatigue driving behaviors;
the second scene is combined with toll gate and road gate data to obtain a travel chain of a vehicle from the municipal road to the toll gate, the travel chain of the vehicle municipal road and the expressway are fused, continuous running time of the vehicle on the municipal road and the expressway is obtained, and if the continuous running time exceeds 4 hours, the vehicle is judged to have fatigue driving behaviors possibly.
6. The method of claim 5, wherein the specific steps of the step (4) of the stay discriminant analysis of the potential fatigue driving vehicle at high speed are as follows:
step21: vehicle behavior feature discrimination based on massive historical data
(201) Constructing a single vehicle travel information base between adjacent portal frames of the expressway; establishing a portal sequence { M ] for vehicles to pass over a highway s1 ,M s2 ,M s3 ......,M sn Time series of vehicle passing through portal { T } and s1 ,T s2 ,T s3 ......T sn (wherein n is the number of frames in the investigation region, M) sj To study the jth portal passed by the vehicle during a period of time, T sj For the time that the vehicle passes the jth portal in the research period, j epsilon (1, n); calculating the travel time T of a vehicle passing between adjacent door frames ab ，T ab The travel time sequence { T of the adjacent door frames for the passing of the vehicle can be obtained for the travel time difference of the vehicle passing through the door frame a and the door frame b 1_ab ,T 2_ab ,T p_ab ......,T N_ab (wherein N isThe study period is defined by the total number of all vehicles between the door frame a and the door frame b, T p_ab For the travel time difference of the vehicle p passing through the a-th portal and the b-th portal, p epsilon (1, N);
(202) Judging the behavior characteristics of the vehicle; the box diagram principle in statistics is adopted as the basis for judging the vehicle behavior, and the box diagram is generally considered to be greater than the upper limit Q in the data set 3 +1.5×(Q 3 -Q 1 ) Or less than the lower limit of Q 1 -1.5×(Q 3 -Q 1 ) Outliers, where Q 1 、Q 3 The first quantile and the third quantile of the data are respectively;
for the travel time sequence of each adjacent portal frame pair of the expressway, sorting from small to large, and calculating the upper limit T of the travel time of the adjacent portal frames of the service area kk_max And lower limit T kk_min Travel time threshold range T for the ith adjacent gantry pair i_kk The calculation is as follows:
wherein: dis (Dis) i -road network distance of the ith portal pair;
v min -the lowest running speed between two portal frames of the expressway is selected according to the historical data;
v max -the highest running speed between two portal frames of the expressway, and selecting running conditions according to historical data;
according to the principle of a box diagram, a travel time threshold range is determined by combining historical data among the portal frames of the expressway, so that values of a median, an upper quartile, a lower quartile, an upper limit and a lower limit in the box diagram are determined, a travel time threshold experience library for forming each adjacent portal frame pair can be stored in a database as basic data, updated periodically for long-term use, and different types of vehicle behavior state analysis of any period of the adjacent portal frames of the expressway section can be supported; the relation between the travel time of the adjacent portal frame and the behavior characteristics of the vehicle is as follows:
(1) the travel time being less than the lower limit value is attributed to overspeed behavior;
(2) the travel time is greater than the lower limit and less than the lower quartile due to free flow behavior;
(3) the travel time is greater than the lower quartile and less than the upper quartile due to normal driving behavior;
(4) the travel time is greater than the upper quartile and less than the upper limit due to congestion creep behavior;
(5) travel times greater than the upper limit are due to dwell behavior;
step22: vehicle behavior feature discrimination for potential fatigue driving
And building and identifying potential fatigue driving vehicle information through a vehicle travel chain, selecting license plate numbers of the potential fatigue driving vehicles, obtaining travel time of adjacent doors on the upstream and downstream sides of a highway service area, and comparing and analyzing the travel time with a threshold experience database of the same type of vehicle travel time of the adjacent door frame pair, thereby judging the behavior characteristics of the potential fatigue driving vehicles.
7. The method according to claim 6, wherein the step (5) is to determine a fatigue-driven vehicle on the expressway, and to implement fatigue-driven vehicle management;
defining a vehicle with continuous running time exceeding 4 hours as a potential fatigue driving vehicle, wherein the vehicle stays for a long time after entering a highway and generally has traffic accidents or enters a service area; whether the vehicle stays on the expressway due to traffic accidents or not can be obtained through law enforcement records, rescue records and the like; judging whether the vehicle enters the service area or not based on the travel time recorded by adjacent door frames at the upstream and downstream of the service area; and if the potential fatigue driving vehicle does not have a traffic accident and does not enter the service area, judging the potential fatigue driving vehicle as the fatigue driving vehicle.
8. A highway fatigue driving early warning management system based on multi-source data fusion analysis is characterized in that the system is based on a traffic flow database and a geographic information database, relies on potential fatigue driving vehicle identification and vehicle stay discriminant analysis through data association, comprehensively analyzes the highway fatigue driving vehicles, and provides a scheme for realizing accurate management of the fatigue driving vehicles for highway operation enterprises and traffic management departments; the method comprises the following modules:
the traffic flow data acquisition module: the traffic flow information transmission module is used for acquiring traffic flow data;
highway geographic information module: establishing a highway geographic information model;
potential fatigue driving vehicle identification module: counting continuous driving time identification by constructing a vehicle travel chain;
the potential fatigue driving vehicle stays at a high speed to distinguish the analysis module;
and the fatigue driving vehicle judging and managing module is used for: and judging the fatigue driving vehicle, and realizing the accurate management of the fatigue driving vehicle.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the highway fatigue driving early warning management method based on multi-source data fusion analysis according to any one of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when executed by a processor, implements the highway fatigue driving early warning management method based on multi-source data fusion analysis according to any one of claims 1 to 7.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2022114246656 | 2022-11-14 | ||
| CN202211424665 | 2022-11-14 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117037464A true CN117037464A (en) | 2023-11-10 |
Family
ID=88625068
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310534050.7A Pending CN117037464A (en) | 2022-11-14 | 2023-05-11 | Expressway fatigue driving early warning management method based on multisource data fusion analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117037464A (en) |
-
2023
- 2023-05-11 CN CN202310534050.7A patent/CN117037464A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109448369B (en) | Real-time operation risk calculation method for expressway | |
| Montella et al. | Effects on speed and safety of point-to-point speed enforcement systems: Evaluation on the urban motorway A56 Tangenziale di Napoli | |
| CN110164132B (en) | Method and system for detecting road traffic abnormity | |
| CN109191861B (en) | System and method for detecting abnormal behavior of fee evasion vehicle on expressway based on video detector | |
| Li et al. | Identifying important variables for predicting travel time of freeway with non-recurrent congestion with neural networks | |
| CN110264734A (en) | A kind of vehicle on highway auditing system and its working method | |
| CN107067730B (en) | Network appointment vehicle-man-vehicle inconsistency monitoring method based on bayonet equipment | |
| CN106846801A (en) | A kind of region based on track of vehicle is hovered anomaly detection method | |
| US20220383738A1 (en) | Method for short-term traffic risk prediction of road sections using roadside observation data | |
| Wang et al. | A field investigation of red-light-running in Shanghai, China | |
| CN109979197B (en) | Method and system for constructing highway traffic time map based on fusion data | |
| CN109448384B (en) | A kind of highway danger traffic behavior recognition methods | |
| CN107784832A (en) | A kind of method and apparatus for being used to identify the accident black-spot in traffic route | |
| CN114970943B (en) | Expressway accident type and severity prediction method | |
| CN111899517B (en) | Expressway fatigue driving illegal behavior determination method | |
| CN112767684A (en) | Highway traffic jam detection method based on charging data | |
| Wang et al. | Crash prediction for freeway work zones in real time: A comparison between Convolutional Neural Network and Binary Logistic Regression model | |
| CN111754786A (en) | System for identifying traffic vehicle passing events on highway | |
| CN116631186B (en) | Expressway traffic accident risk assessment method and system based on dangerous driving event data | |
| Chen et al. | Lane-based Distance-Velocity model for evaluating pedestrian-vehicle interaction at non-signalized locations | |
| CN114550445A (en) | Urban area traffic safety state evaluation method and device | |
| CN110807929A (en) | Traffic police auxiliary command information analysis and research and judgment system | |
| CN116935659B (en) | High-speed service area bayonet vehicle auditing system and method thereof | |
| CN111798579A (en) | ETC portal frame-based highway congestion prediction method and system | |
| CN117037464A (en) | Expressway fatigue driving early warning management method based on multisource data fusion analysis |
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 |