CN111354193B - Highway vehicle abnormal behavior early warning system based on 5G communication - Google Patents

Highway vehicle abnormal behavior early warning system based on 5G communication Download PDF

Info

Publication number
CN111354193B
CN111354193B CN202010119070.4A CN202010119070A CN111354193B CN 111354193 B CN111354193 B CN 111354193B CN 202010119070 A CN202010119070 A CN 202010119070A CN 111354193 B CN111354193 B CN 111354193B
Authority
CN
China
Prior art keywords
vehicle
data
speed
behavior
rsu
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
Application number
CN202010119070.4A
Other languages
Chinese (zh)
Other versions
CN111354193A (en
Inventor
蔡英凤
汪梓豪
吕志军
王海
李祎承
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jiangsu University
Original Assignee
Jiangsu University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Jiangsu University filed Critical Jiangsu University
Priority to CN202010119070.4A priority Critical patent/CN111354193B/en
Publication of CN111354193A publication Critical patent/CN111354193A/en
Application granted granted Critical
Publication of CN111354193B publication Critical patent/CN111354193B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2462—Approximate or statistical queries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/044—Recurrent networks, e.g. Hopfield networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • 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
    • G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
    • G08G1/0116—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
    • 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
    • G08G1/0125—Traffic data processing
    • 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
    • G08G1/0125—Traffic data processing
    • G08G1/0133—Traffic data processing for classifying traffic situation
    • 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
    • G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
    • 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
    • G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
    • G08G1/054—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed photographing overspeeding vehicles
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/025—Services making use of location information using location based information parameters
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/025—Services making use of location information using location based information parameters
    • H04W4/027—Services making use of location information using location based information parameters using movement velocity, acceleration information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/38—Services specially adapted for particular environments, situations or purposes for collecting sensor information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
    • H04W4/44—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for communication between vehicles and infrastructures, e.g. vehicle-to-cloud [V2C] or vehicle-to-home [V2H]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/70—Services for machine-to-machine communication [M2M] or machine type communication [MTC]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W40/00—Communication routing or communication path finding
    • H04W40/24—Connectivity information management, e.g. connectivity discovery or connectivity update
    • H04W40/32—Connectivity information management, e.g. connectivity discovery or connectivity update for defining a routing cluster membership

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Computational Linguistics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Biophysics (AREA)
  • Analytical Chemistry (AREA)
  • Chemical & Material Sciences (AREA)
  • Computing Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Probability & Statistics with Applications (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Fuzzy Systems (AREA)
  • Databases & Information Systems (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention discloses a 5G communication-based highway vehicle abnormal behavior early warning system, which is used for acquiring vehicle data and transmitting the vehicle data to an intelligent RSU when a vehicle enters the detection range of the intelligent RSU and a roadside sensor; the intelligent RSU carries out weighting fusion processing on data with the same physical address and over-small time difference, multi-stage judgment on the vehicle behavior is carried out by utilizing a condition judgment and a cyclic neural network model, finally, probability matrixes of the vehicle belonging to different behaviors in the time period are output, the probability matrixes of the different behaviors are compared with behavior combination data stored in a database, the vehicle behavior is determined according to the similarity, corresponding reminding information is determined according to the behavior, and the corresponding reminding information is sent to the corresponding vehicle; if the similarity is lower than the threshold value, the data are sent to the information processing center, the information processing center carries out secondary processing on the data, and corresponding prompts are directly sent to the vehicle. The invention can obtain more accurate vehicle behavior in the vehicle identification process and simultaneously improve the timeliness of data transmission.

Description

Highway vehicle abnormal behavior early warning system based on 5G communication
Technical Field
The invention belongs to the technical field of intelligent traffic, and particularly relates to a 5G communication-based highway vehicle abnormal behavior early warning system, a vehicle road cooperative information interaction mode, an intelligent road side unit and a vehicle behavior discrimination mode.
Background
With the development of the times, the expressway has become one of important travel ways. The dynamic driving environment on the expressway requires that the driver pay attention to information such as surrounding vehicles all the time, which causes great loss of driver's energy, and the driver cannot pay continuous attention to abnormal behaviors of vehicles on the road, thus traffic accidents such as vehicle collision are easily caused. Therefore, the construction of the highway vehicle abnormal behavior early warning system is necessary, and has important significance for improving the safety of vehicles running on the highway.
The existing highway vehicle monitoring system mainly detects the overspeed of a vehicle, the vehicle speed measurement can be divided into interval speed measurement and fixed point speed measurement, the interval speed measurement utilizes a monitoring camera to obtain the time of the vehicle passing through two speed measurement points, and the average running speed of the vehicle in the interval is calculated; the fixed point speed measurement utilizes a speed measuring radar or a ground induction coil to measure the instantaneous speed of the vehicle passing through a speed measuring point, and the measured speed is compared with the road speed limit to judge whether the vehicle is overspeed or not. In the speed measurement mode, the arrangement distance of the speed measurement sensors is kilometer, only the average speed of the vehicle in the interval and the instantaneous speed of a plurality of speed measurement points can be obtained, only the overspeed behavior in the interval can be detected, and the overspeed behavior of the vehicle monitoring blind area cannot be detected. Moreover, the highway vehicle monitoring system does not have the capability of identifying the behaviors of the road vehicles, and can not carry out early warning reminding on the vehicle drivers and the drivers of the surrounding vehicles for the behaviors of illegal parking of the vehicles occupying emergency lanes, abnormal driving of the vehicles, abnormal inside of the vehicles and the like, so that the safety of the vehicles running on the highway is reduced.
In addition, in the intelligent transportation system, a Road Side Unit (RSU) is used as a communication unit for vehicle-road cooperation, and the road side unit communicates with the vehicle by using DSRC communication technology, but the DSRC communication technology is greatly influenced by the vehicle speed, the vehicle speed for stably transmitting data is kept below 60km/h, and the vehicle speed condition required by an expressway is not met. Meanwhile, in the DSRC communication technology, data transmission is greatly affected under the condition of large road traffic volume, and packet loss may occur. In recent years, the development of the 5G technology is leapfrog, the air interface time delay is about 1ms, the end-to-end time delay is controlled to be at the millisecond level, the communication robustness is good, and the requirements of low time delay and high reliability of the communication of the Internet of vehicles can be met.
Disclosure of Invention
In order to solve the problems, the invention provides a 5G communication-based highway vehicle abnormal behavior early warning system, a road side unit and a road end sensor under the 5G communication condition are arranged, and data and information are transmitted between vehicle roads in a D2D communication mode, so that low-delay and high-efficiency transmission and processing of vehicle-road information are realized; the intelligent roadside unit internal processing module can perform multi-stage judgment based on sensor and vehicle data, can accurately identify abnormal behaviors of vehicles compared with a traditional machine learning algorithm and a deep learning model, has good real-time performance, and can remarkably improve traffic safety of highways.
The invention provides an expressway vehicle abnormal behavior early warning system which is shown in figure 1 and comprises the following parts: the system comprises an intelligent Road Side Unit (RSU), a road side sensor, a vehicle (provided with an on-board T-Box) and an information processing center, wherein the road side sensor and the intelligent RSU perform information interaction processing, the vehicle and the intelligent RSU perform information interaction processing, and the intelligent road side unit RSU and the information processing center complete processing and sending of early warning reminding information.
The intelligent RSU is based on the coverage range of the 5G base station signal, so that an approximate value of the optimal distance of the intelligent RSU arranged at the road end of the given highway and the optimal distance of the millimeter wave radar and the speed measurement camera arranged on the road side sensor are obtained, and the specific implementation can be referred to.
The roadside sensor mainly achieves data acquisition and transmission functions, and when a vehicle enters the detection range of the intelligent RSU and the roadside sensor, a millimeter wave radar and a speed measurement camera arranged at a road end acquire speed and posture data of the vehicle. The roadside sensor then transmits the detected vehicle information to the intelligent RSU, which stores the data within the internal data storage module.
The intelligent RSU and vehicle data transmission communication process comprises the following steps: the intelligent RSU and the vehicle are communicated based on the D2D communication technology, the vehicles are clustered by using a clustering algorithm on the basis of Euclidean distance, the cluster head vehicle transmits acquired vehicle parameter data acquired by a vehicle sensor in each cluster to the nearest intelligent RSU, and the vehicle parameter data are stored in a data storage module in the intelligent RSU. The data stored to the data storage module includes: vehicle MAC address, time, vehicle speed, acceleration, angular velocity, lateral displacement, attitude angle, lane position, geographic coordinates, and the like.
Data processing and prompt information sending: the intelligent RSU leads the stored data into an internal data processing module, and performs weighting fusion processing on the data with the same physical address and small time difference. And after the fused data is input, performing multi-stage judgment on the vehicle behaviors by utilizing the condition judgment and the recurrent neural network model, and finally outputting a probability matrix that the vehicle belongs to different behaviors in the time period. And comparing each behavior probability matrix with the vehicle behavior combination data in the vehicle behavior database, and determining the behavior of the vehicle in the time period after the similarity reaches a certain value. And corresponding the behavior to the behavior in the database, acquiring vehicle reminding information which is to be sent under the behavior, sending the vehicle reminding information to the corresponding vehicle through the wireless communication module, and reminding a driver of carrying out corresponding operation on the vehicle in time. If the information similarity is slightly lower than the set value, the data are sent to an information processing center, the information processing center carries out secondary processing on the data, and corresponding prompts are directly sent to the vehicle.
The invention has the following effective benefits:
(1) the highway roadside unit and the sensor arrangement mode provided by the invention can obtain more accurate vehicle speed information by taking the vehicle speed measurement range as a unit of hundred meters. Meanwhile, data acquired by the millimeter wave radar and the camera and vehicle data uploaded by the vehicle are fused to increase the authenticity and accuracy of the data, and more accurate vehicle behaviors can be obtained in the vehicle identification process.
(2) The intelligent road side unit provided by the invention comprises an instant data processing function, can effectively share the vehicle information processing task in a longer road to the processing module of each road side unit on the road, and divides the judging process into two parts, thereby improving the accuracy and the calculating speed of vehicle behavior identification, reducing the data processing amount and the data processing burden of an information center, and simultaneously improving the timeliness of data transmission.
(3) The prompt information is sent after secondary processing of the intelligent road side unit and the information center, and accuracy of data and vehicle behavior judgment is greatly improved. The D2D communication mode in the 5G communication technology is applied to data interaction between systems, the problem of unstable data transmission of vehicles running at high speed is avoided, meanwhile, the communication delay is reduced, the pressure of a core network and a base station of a communication system is reduced, the frequency spectrum utilization rate and the data throughput are improved, and the data are sent and received more timely.
Drawings
FIG. 1 is a schematic diagram of a vehicle abnormal behavior prompt system
FIG. 2 shows an intelligent RSU internal module
FIG. 3 illustrates a communication connection based on the D2D communication mode
FIG. 4 is a RSU data processing flow chart
FIG. 5 is a flowchart of vehicle abnormal behavior recognition
Detailed Description
The invention will be further explained with reference to the drawings.
The invention provides a 5G communication-based highway vehicle abnormal behavior early warning system, which realizes the functions of vehicle path information interaction and early warning. The intelligent road side unit is internally provided with the data storage and processing module, can perform multi-stage judgment based on sensor and vehicle data, accurately identifies the abnormal behavior of the vehicle, and can send prompt and early warning information aiming at the abnormal behavior in real time by applying 5G communication, thereby greatly improving the traffic safety of the highway.
As shown in fig. 3, the communication processing principle and process of the system are mainly used for communicating with a vehicle entering the communication range of the intelligent RSU through a communication module of the intelligent RSU, acquiring vehicle terminal sensor data, and fusing the vehicle terminal sensor data with vehicle behavior data acquired by the roadside sensor device. The communication of the system is mainly realized by a 5G communication module, a data storage module, a data processing module and a database.
The main setting and operation method of the expressway vehicle abnormal behavior early warning system is as follows:
the first step is as follows: intelligent road side unit and road side sensor arrangement
An intelligent RSU is arranged at each interval distance L of a given highway end, and a millimeter wave radar and a speed measuring camera are arranged on a roadside sensor. For millimeter wave radar, 79Ghz millimeter wave radar using mimo (Multi Input Multi output) technology, which includes a larger transmit-receive antenna array producing a larger transmit aperture and a narrower beam, possesses a very high angular resolution, and can provide a bandwidth of up to 4Ghz and a target separation accuracy of 7.5 cm. The data acquisition requirement of longer distance and higher precision and the requirement of faster data transmission are met, and the matching degree of the MIMO technology and the current 5G communication technology is higher.
The millimeter wave radar is arranged at the position 1-4m away from the lane on the road side, and the interval between the two millimeter wave radars is set to be 200m and is slightly smaller than the maximum detection distance. The speed measuring distance of the speed measuring camera is about 300m, and the camera and the millimeter wave radar are arranged at the same position point, so that data synchronous processing is facilitated.
As shown in fig. 2, the internal module and the device of the intelligent RSU include a 5G wireless communication module, a data storage module, a data processing module, and a database, where the database includes vehicle behavior data and vehicle prompt information data.
The intelligent RSU set distance is established based on the arrangement of the 5G base stations. The longest setting distance L approximate value meets the following conditions:
Figure BDA0002392406020000041
wherein n is the number of lanes, d is the width of lane, and r is the coverage radius of 5G base station signals, generally between 300 and 400 m.
The second step is that: roadside sensor data acquisition and transmission
When the vehicle enters the range of the intelligent RSU and the roadside sensor, the roadside sensor records data of the vehicle, and the intelligent RSU and the roadside sensor communicate to acquire vehicle information detected by the roadside sensor and store the vehicle information in the internal data storage module.
The intelligent RSU communication module directly communicates with the millimeter wave radar and the camera in a D2D communication mode through a PC5 interface, and the millimeter wave radar and the camera which communicate with the intelligent RSU are designated as sensor terminals which can be searched within the RSU set interval distance range. And finally, vehicle video data obtained by shooting through the optical camera and vehicle position, posture and time information obtained by the millimeter wave radar are stored in a storage module of the intelligent RSU, the next step of processing is waited to obtain data such as vehicle license plate information, lane information, vehicle speed, side deflection angle and data obtaining time, the RSU automatically clears data cache every 1 minute to release storage space, and the smoothness and the high efficiency of operation of the processing module are guaranteed.
The third step: data transmission between vehicle and intelligent road side unit
Intelligence RSU communicates with the vehicle, obtains each item of parameter data of vehicle that vehicle self sensor gathered, includes: the vehicle MAC address, time, vehicle speed, acceleration, angular velocity, lateral displacement, attitude angle, lane position, geographical coordinates, etc. are stored in a data storage module (i.e., database).
The intelligent RSU communication module establishes a direct communication link with wireless communication modules in the T-BOX of all vehicles within communication range via the D2D communication mode. The invention adopts a method for grouping and communicating vehicles and intelligent RSUs, wherein the establishment of a communication link needs to carry out resource coordination through a base station, and the grouping and communicating method specifically comprises the following contents:
1. the base station establishes a control communication link with the intelligent RSU and the vehicle in the signal coverage area thereof for communication resource allocation equipment to communicate with each other after establishing connection with the help of the base station. The base station realizes the function of authenticating the equipment access communication and performs connection control and resource allocation on the equipment. The D2D connection on the device layer in the D2D communication mode shares the licensed band with the legacy cellular network.
2. And establishing a terminal communication model at the D2D application server, and firstly numbering each intelligent RSU in the communication range of the base station as a determination point terminal device under the base station. The intelligent RSU establishes a control link with the base station to access the D2D communication authorization frequency band, and is connected with the gateway service and the D2D application server through the core network.
3. And optimizing the node energy consumed by the sensor for transmitting data by vehicles in the base station signal coverage range through an LEACH-REC clustering algorithm. Wherein, clustering processing is carried out on the vehicles by taking the Euclidean distance as a condition.
The Euclidean distance formula is as follows:
Figure BDA0002392406020000051
wherein ViAs the current vehicle speed, VjT is the data updating period and y is the transverse distance between two vehicles.
The clustering coefficient λ can be obtained:
Figure BDA0002392406020000052
wherein ω isiAnd muiAs weighting factors for the velocity difference and Euclidean distance, dD2DThe maximum distance D2D can communicate. λ is used to determine whether the current vehicle can continue to be classified into the cluster: when lambda is too large, the speed difference between the vehicle and the surrounding vehicles is too large or the distance between the vehicles is too large, and the communication requirement cannot be met, and when lambda exceeds a set threshold value, the vehicle i at the moment is moved out of the cluster.
4. And D2D communication is carried out by establishing control link coordination between vehicles in the same cluster and the base station. And marking the vehicle positioned at the forefront position in the cluster as a communication vehicle, and transmitting the vehicle sensor data to the communication vehicle by each vehicle in the communication group, and temporarily storing the data in a storage module of the T-BOX of the communication vehicle. And determining the intelligent RSU with the nearest distance according to the signal strength between the communication vehicle and each intelligent RSU, and establishing a D2D communication link under the control of a base station with the intelligent RSU by using a communication module in the T-BOX. The T-BOX is accessed to a CAN bus through an interface, obtains the vehicle speed, the acceleration, the transverse displacement and the attitude angle detected by sensors such as an acceleration sensor, a gyroscope, a millimeter wave radar, a camera, a vehicle speed sensor and the like in the vehicle, and stores the acquired real-time data in an internal storage medium; the OBU module of the vehicle-mounted unit in the T-BOX can provide information such as gears, oil consumption, fault codes, vehicle MAC addresses and acquisition time of each item of data, and the like, and the information is analyzed in the data processing module, and finally, each item of parameter data of the vehicle is transmitted to the intelligent RSU, and the intelligent RSU receives the data and stores the data in the data storage module.
The fourth step: the intelligent road side unit processes the data and sends prompt information
As shown in fig. 4, the intelligent RSU imports the stored data into an internal data processing module, performs weighted fusion processing on the data with the same MAC address and the time difference value smaller than t, performs multi-stage judgment on the vehicle behavior by using a condition judgment and a recurrent neural network model after inputting the fused data, and finally outputs a probability matrix that the vehicle belongs to different behaviors in the time period. And comparing each behavior probability matrix with the vehicle behavior combination data in the vehicle behavior database, and determining the behavior of the vehicle in the time period when the similarity exceeds k. And calling vehicle reminding information corresponding to the type of behavior in the database, and sending the vehicle reminding information to the corresponding vehicle through the wireless communication module again to remind the vehicle to perform correct operation in time. If the information similarity is less than k, the data are sent to an information processing center, secondary processing is carried out on the data to accurately judge the vehicle behavior, and finally the information processing center directly communicates with the vehicle to send out a corresponding prompt.
The method comprises the following specific implementation steps:
1. inputting the data in the intelligent RSU storage module into the data processing module in a matrix form: firstly, inputting a mark function mark in a data processing module, combining and arranging data with the same mac address according to a time sequence, and outputting a new data matrix M. Performing data fusion processing on M, and recording a matrix after data fusion as M3The fusion process is as follows:
arranging the data according to time sequence to form a data matrix, and making delta t equal to ti+1-tiI belongs to (1, n-1), if delta t is less than m, m is a preset time interval threshold value, all t are carried outi+1The row data is extracted to form a new data matrix M2All of ti+1Line data position and tiThe positions of the row data are in one-to-one correspondence, and the rest positions are taken as 0 vectors. Removing data from the original matrix M to form a new matrix M1Will M1、M2Multiplying the rows of the matrix by a weight coefficient w1i、w2i(w1i+w2iMatrix M 'obtained after 1)'1And M'2Adding to obtain a new matrix M3As a data matrix for further processing.
Figure BDA0002392406020000071
Wherein A is the mac address of the vehicle, a is the acceleration of the vehicle, v is the speed of the vehicle, theta is the pitch angle, psi is the course angle,
Figure BDA0002392406020000072
is roll angle, y is lateral displacement, t isData acquisition time.
Figure BDA0002392406020000073
Figure BDA0002392406020000074
w1、w2As a weight coefficient, M3＝M′1+M′2
2. The intelligent road side unit vehicle behavior and prompt information setting: the impending behavior of the vehicle in the highway accident can be judged and predicted by acquiring the posture and state data of the vehicle and the relative data between the two vehicles, so that the vehicle with abnormal behavior and the vehicle with the impending dangerous behavior are prompted to improve the driving safety of the vehicle on the highway, and the determination of all the vehicles is based on the mac address sent by the vehicle to the intelligent RSU.
As shown in FIG. 5, the present invention provides six conditions for determining abnormal behaviors of vehicles, and firstly, a matrix M corresponding to all vehicles is used3And inputting the transposed data into a data processing module. When the vehicle is judged, a condition judgment method is firstly used for determining whether the vehicle is in behaviors of overspeed driving, low speed driving, dangerous vehicle following, emergency lane occupation and the like, and then data is input into a long-short term memory network (LSTM) to judge whether the vehicle is in behaviors of dangerous steering and dangerous lane changing, wherein the condition judgment and the process are as follows:
the model structure and the using method of the recurrent neural network model for dangerous behavior judgment are as follows:
the model consists of 24 LSTM neurons, an implicit layer using a 3-layer cyclic layer stack, and a fully connected layer to output data. And a cross entropy function is adopted as a loss function when the model fits data, fitting of the model data is optimized by using an Adam optimizer, and the learning rate Ir is set to be 0.005. The output data is enabled to be between 0 and 1 through a softmax function.
The model is pre-trained by using a large amount of data before being put into use, the input data are relevant parameters causing each dangerous behavior, and the output data are probability vectors of each dangerous behavior. The related settings can be changed according to the change of the specific loss function and the precision during the test, and finally, the model weight value with the highest test precision is stored and put into a data processing module of the intelligent RSU for practical use.
Overspeed driving judgment conditions:
(1) and determining the lane position information of the vehicle through an in-vehicle GPS, and acquiring the highest speed limit of the lane.
(2) If the speed v of the vehicle speed v fused in the data sent to the RSU by the road end sensor and the vehicle is at a certain momentiWhen the lane speed limit condition is reached, fitting a curve of the acceleration a by acquiring data points of the acceleration a of the vehicle in a sensor detection interval, calculating an integral value of the acceleration curve within the following 15s, and if the sum of the vehicle speed and the integral value of the acceleration within the 15s is more than or equal to the lane speed limit vmaxThen the vehicle is determined to be overspeed.
Figure BDA0002392406020000081
Overspeed of vehicle
viIs the vehicle speed at time i, a is the vehicle acceleration, vmaxThe highest speed limit for the lane.
Low-speed driving judgment conditions:
(1) and determining the lane position information of the vehicle through an in-vehicle GPS, and acquiring the lowest speed limit of the lane.
(2) If the speed v of the vehicle speed v fused in the data sent to the RSU by the road end sensor and the vehicle is at a certain momentiWhen the speed is lower than the lane speed limit condition, fitting a curve of the acceleration a by acquiring data points of the acceleration a of the vehicle in a sensor detection interval, calculating an integral value of the acceleration curve in the following 15s, and if the sum of the vehicle speed and the integral value of the acceleration in the 15s is less than or equal to the lowest speed limit v of the laneminThen, it is determined that the vehicle is traveling at a low speed.
Figure BDA0002392406020000082
Low speed running of vehicle
viThe speed (km/h) of the vehicle at the moment i, a is the acceleration (m/s) of the vehicle, and v isminThe highest speed limit (km/h) of the lane.
And (4) dangerous car following judgment conditions:
(1) the vehicle speed v is judged once, and the section where the vehicle speed is located is determined to be above 100km/h or below 100 km/h.
(2) If the vehicle speed v isiThe Euclidean distance s between the vehicle and the front vehicle is less than or equal to 100km/h, and the vehicle is judged to be dangerous to follow when the Euclidean distance s between the vehicle and the front vehicle is less than or equal to 100 m.
(3) If the vehicle speed v isiThe Euclidean distance s between the vehicle and the front vehicle is less than or equal to 100km/h, and the vehicle is judged as dangerous following when the Euclidean distance s between the vehicle and the front vehicle is less than or equal to 50 m.
And (3) emergency lane occupation judging conditions:
(1) lane position information of the vehicle and geographic coordinate information (x, y) of the vehicle are determined by the in-vehicle GPS.
(2) If the vehicle communicates with the intelligent RSU after the specified time t: its lane position has not changed; and if the geographic coordinate information of the vehicle is not changed, judging that the vehicle is in an emergency behavior.
Dangerous steering behavior judgment process:
(1) the vehicle is about to turn over when turning, and the vehicle angular velocity omega and the vehicle roll angle
Figure BDA0002392406020000091
The vehicle speed v and other parameters are correlated, and in order to identify the side-turning behavior, the vehicle angular speed omega and the vehicle roll angle of the labeled vehicle normal steering turning and side-turning critical point are input into the LSTM model
Figure BDA0002392406020000092
The vehicle speed v is pre-trained.
(2) And inputting the data obtained in real time into the trained model, and outputting the probability of whether the vehicle behavior is about to turn over.
The dangerous lane changing behavior judgment process comprises the following steps:
(1) the dangerous lane changing behavior relates parameters such as a heading angle psi, a vehicle speed v, a transverse displacement y and an angular speed omega of a vehicle with the rear of an adjacent lane with Euclidean distance s of the vehicle behind the adjacent lane, and in order to identify the dangerous lane changing behavior of the vehicle, the heading angle psi, the vehicle speed v, the transverse displacement y and the angular speed omega of the vehicle with a marked normal lane changing and dangerous lane changing are input in an LSTM model to be pre-trained with the Euclidean distance s of the vehicle behind the adjacent lane.
(2) And inputting the data obtained in real time into the trained model, and outputting the probability of whether the vehicle behavior is dangerous lane change.
3. After secondary judgment, the probability P of behaviors such as overspeed driving, low speed driving, dangerous car following, emergency lane occupation, dangerous steering, dangerous lane changing and the like is outputj. And performing similarity calculation on the obtained probability matrix of each behavior and data in a vehicle behavior database, and calculating the total absolute distance between the data by using the Euclidean distance:
Figure BDA0002392406020000093
wherein xiTo output the probability of a certain abnormal behavior in the behavior data vector, yiN is the probability of a certain behavior in a given behavior data vector in the vehicle behavior database, and n is the number of behaviors contained in the data vector at time t.
When k > ktWhen (k)tA similarity threshold) that the vehicle matches a single abnormal behavior or a combination of behaviors in the database. And finally, outputting all the abnormal behavior data of the vehicle, matching the abnormal behavior data with a vehicle prompt information database, and finally determining prompt information to be sent aiming at the abnormal behaviors of different vehicles. When the similarity k between the behavior data and all data in the intelligent RSU behavior database is too small, the intelligent RSU cannot judge the specific content of the behavior, the intelligent RSU takes the 5G base station as a relay and transmits the data to the information processing center, and the information is secondarily identified by the big database stored in the information processing center.
4. The intelligent RSU data processing module sends the finally output prompt information to the communication module, and the communication module directly sends the vehicle abnormal behavior information to the vehicle T-BOX through the established communication link. And the data after secondary identification by the information processing center is directly sent to the cluster head vehicle, the T-BOX of the cluster head vehicle carries out block processing on the acquired information and sends the information corresponding to the vehicles in the cluster to each vehicle in the cluster. And finally, each vehicle T-BOX sends the acquired prompt information to a display screen of the vehicle-mounted terminal to remind a driver of timely carrying out corresponding operation so as to avoid traffic accidents.
The above-listed series of detailed descriptions are merely specific illustrations of possible embodiments of the present invention, and they are not intended to limit the scope of the present invention, and all equivalent means or modifications that do not depart from the technical spirit of the present invention are intended to be included within the scope of the present invention.

Claims (6)

1. The utility model provides a highway vehicle abnormal behavior early warning system based on 5G communication which characterized in that includes: the system comprises an intelligent Road Side Unit (RSU), a road side sensor, a vehicle and an information processing center; the road side sensor and the intelligent road side unit RSU perform information interaction processing, the vehicle and the intelligent road side unit RSU perform information interaction processing, and the intelligent road side unit RSU and the information processing center complete processing and sending of the early warning reminding information;
the information interaction between the vehicle and the intelligent road side unit RSU is as follows:
the intelligent road side unit RSU and the vehicle are communicated based on a D2D communication technology, the vehicles are clustered by using a clustering algorithm on the basis of Euclidean distance, and the cluster head vehicle transmits acquired vehicle parameter data acquired by a vehicle sensor in each cluster to the nearest intelligent road side unit RSU and stores the data in a data storage module in the intelligent road side unit RSU; the vehicle parameters collected by the vehicle sensors comprise: vehicle MAC address, time, vehicle speed, acceleration, angular velocity, lateral displacement, attitude angle, lane position, and geographic coordinates;
the communication between intelligence road side unit RSU and the vehicle specifically designs into:
(1) the base station establishes a control communication link with the intelligent road side unit RSU and vehicles within the signal coverage range thereof so as to enable allocation equipment of communication resources to communicate with each other after establishing connection with the help of the base station; the base station realizes the function of authenticating the access communication of the equipment and performs connection control and resource allocation on the equipment; the D2D connection on the device layer in the D2D communication mode shares the licensed band with the legacy cellular network;
(2) establishing a terminal communication model at a D2D application server, firstly numbering each intelligent Road Side Unit (RSU) in a communication range of a base station, and taking the RSU as a determined point terminal device under the base station; the method comprises the following steps that an intelligent RSU and a base station establish a control link to access a D2D communication authorization frequency band, and are connected with a gateway service and a D2D application server through a core network;
(3) optimizing node energy consumed by sensor data transmission by vehicles in a base station signal coverage range through an LEACH-REC clustering algorithm; clustering the vehicles by taking the Euclidean distance as a condition;
the Euclidean distance formula is as follows:
Figure FDA0003150605800000011
wherein ViAs the current vehicle speed, VjThe speed of surrounding vehicles is determined, T is a data updating period, and y is the transverse distance between two vehicles;
obtaining a clustering coefficient lambda:
Figure FDA0003150605800000021
wherein ω isiAnd muiRespectively weighting coefficients for the velocity difference and the euclidean distance,
Figure FDA0003150605800000022
representing the average speed of the vehicle, dD2DFor the maximum distance of D2D communication, λ is used to determine whether the current vehicle can continue to be classified into the cluster: when lambda is too large, the speed difference between the vehicle and the surrounding vehicles is too large or the distance between the vehicles is too largeThe communication requirement cannot be met, and when the lambda exceeds a set threshold value, the current vehicle at the current moment is moved out of the cluster;
(4) vehicles in the same cluster establish a control link with a base station to coordinate to carry out D2D communication; marking the vehicle at the forefront position in the cluster as a communication vehicle, and sending the vehicle sensor data to the communication vehicle by each vehicle in the communication group and temporarily storing the data in a storage module of a T-BOX of the communication vehicle; determining the nearest intelligent Road Side Unit (RSU) according to the signal strength between the communication vehicle and each intelligent RSU, and establishing a D2D communication link based on base station control with the intelligent RSU by using a communication module in the T-BOX; the T-BOX is accessed to the CAN bus through an interface, obtains the vehicle speed, the acceleration, the transverse displacement and the attitude angle detected by an acceleration sensor, a gyroscope, a millimeter wave radar, a camera and a vehicle speed sensor in the vehicle, and stores the acquired real-time data in an internal storage medium; the on-board unit (OBU) module in the T-BOX can provide gear, oil consumption, fault codes, vehicle Media Access Control (MAC) addresses and acquisition time information of various data, the acquisition time information is analyzed by the data processing module, finally, various parameter data of the vehicle are transmitted to the intelligent Road Side Unit (RSU) together, and the intelligent Road Side Unit (RSU) receives the data and stores the data into the data storage module.
2. The system of claim 1, wherein the intelligent Road Side Unit (RSU) and the road side sensor are spaced at an interval L, and the L is required to satisfy the following conditions:
Figure FDA0003150605800000023
wherein n is the number of lanes, d is the width of lane, and r is the coverage radius of 5G base station signals, generally between 300 and 400 m.
3. The system of claim 1, wherein the roadside sensor comprises a millimeter wave radar and a speed measurement camera to achieve data acquisition and transmission functions, when a vehicle enters the detection range of the intelligent roadside unit (RSU) and the roadside sensor, the millimeter wave radar and the speed measurement camera arranged at the roadside acquire the speed and attitude data of the vehicle passing through the roadside and transmit the detected vehicle information to the intelligent roadside unit (RSU), and the intelligent roadside unit (RSU) stores the data in the internal data storage module.
4. The system of claim 1, wherein the intelligent RSU and the information processing center complete processing and sending of early warning reminding information:
the intelligent road side unit RSU performs weighted fusion processing on data with the same physical address and over-small time difference, performs multi-stage judgment on vehicle behaviors by using condition judgment and a recurrent neural network model, and finally outputs a probability matrix that the current time period of the vehicle belongs to different behaviors; comparing each behavior probability matrix with vehicle behavior combination data in a vehicle behavior database, determining the behavior of the vehicle in the time period after the similarity reaches a certain value, corresponding the behavior to the behavior in the database, acquiring vehicle reminding information to be sent under the behavior, sending the vehicle reminding information to the corresponding vehicle through a wireless communication module, and reminding a driver of carrying out corresponding operation on the vehicle in time; if the information similarity is slightly lower than a set value, sending the data to an information processing center, carrying out secondary processing on the data by the information processing center, and directly sending corresponding early warning prompts to the vehicle; the method comprises the following specific steps:
the data in the intelligent road side unit RSU storage module is input into the data processing module in a matrix form: combining and arranging data of the MAC addresses of the same vehicle according to a time sequence by using a mark function mark in an input data processing module, and outputting a new data matrix M; performing data fusion processing on M, and recording a matrix after data fusion as M3The fusion process is as follows:
arranging the data according to time sequence to form a data matrix, and making delta t equal to ti+1-tiI.e. is (1, n-1) if Δt is less than m, m is the preset time interval threshold value, all t are addedi+1The row data is extracted to form a new data matrix M2All of ti+1Line data position and tiThe positions of the row data are in one-to-one correspondence, and the rest positions are taken as 0 vectors; removing data from the original matrix M to form a new matrix M1Will M1、M2Multiplying the rows of the matrix by a weight coefficient w1i、w2iMatrix M 'obtained thereafter'1And M'2Adding to obtain a new matrix M3As a data matrix for further processing;
Figure FDA0003150605800000041
wherein A is the MAC address of the vehicle, a is the acceleration of the vehicle, v is the speed of the vehicle, theta is the pitch angle, psi is the course angle,
Figure FDA0003150605800000042
is the roll angle, y is the lateral displacement, t is the data acquisition time;
Figure FDA0003150605800000043
Figure FDA0003150605800000044
M3＝M′1+M′2
the intelligent road side unit vehicle behavior and prompt information setting: the impending behavior of the vehicle in the highway accident can be judged and predicted by acquiring the posture and state data of the vehicle and the relative data between the two vehicles, the vehicle with abnormal behavior and the vehicle with impending dangerous behavior are prompted, and the determination of all the vehicles is based on the vehicle MAC address sent by the vehicle to the intelligent road side unit RSU.
5. The system of claim 4, wherein the abnormal behavior of the vehicle comprises six types, and the conditions for determining the abnormal behavior of the vehicle are as follows:
overspeed driving judgment conditions:
(1) determining the lane position information of the vehicle through an in-vehicle GPS, and acquiring the highest speed limit of the lane;
(2) if the speed v of the vehicle speed v at a certain moment after the fusion of the road end sensor and the data sent by the vehicle to the intelligent road side unit RSUiWhen the lane speed limit condition is reached, fitting a curve of the acceleration a by acquiring data points of the acceleration a of the vehicle in a sensor detection interval, calculating an integral value of the acceleration curve within the following 15s, and if the sum of the vehicle speed and the integral value of the acceleration within the 15s is more than or equal to the lane speed limit vmaxJudging that the vehicle is overspeed;
Figure FDA0003150605800000051
overspeed of vehicle
viIs the vehicle speed at time i, a is the vehicle acceleration, vmaxThe highest speed limit of the lane;
low-speed driving judgment conditions:
(1) determining the lane position information of the vehicle through an in-vehicle GPS, and acquiring the lowest speed limit of the lane;
(2) if the speed v of the vehicle speed v at a certain moment after the fusion of the road end sensor and the data sent by the vehicle to the intelligent road side unit RSUiWhen the speed is lower than the lane speed limit condition, fitting a curve of the acceleration a by acquiring data points of the acceleration a of the vehicle in a sensor detection interval, calculating an integral value of the acceleration curve in the following 15s, and if the sum of the vehicle speed and the integral value of the acceleration in the 15s is less than or equal to the lowest speed limit v of the laneminJudging that the vehicle runs at a low speed;
Figure FDA0003150605800000052
low speed running of vehicle
viIs the vehicle speed at time i, a is the vehicle acceleration, vminThe highest speed limit of the lane;
and (4) dangerous car following judgment conditions:
(1) the vehicle speed v is judged once, and the section where the vehicle speed is located is determined to be above 100km/h or below 100 km/h;
(2) if the vehicle speed v isiThe Euclidean distance s between the vehicle and the front vehicle is less than or equal to 100km/h, and the vehicle is judged to be dangerous following when the Euclidean distance s between the vehicle and the front vehicle is less than or equal to 100 m;
(3) if the vehicle speed v isiThe Euclidean distance s between the vehicle and the front vehicle is less than or equal to 100km/h, and the vehicle is judged to be dangerous following when the Euclidean distance s between the vehicle and the front vehicle is less than or equal to 50 m;
and (3) emergency lane occupation judging conditions:
(1) determining lane position information of the vehicle and geographical coordinate information (x, y) of the vehicle through an in-vehicle GPS;
(2) if the vehicle communicates with the intelligent road side unit RSU after the specified time t: if the lane position is not changed and the geographic coordinate information of the vehicle is not changed, judging that the vehicle is an emergency behavior;
dangerous steering behavior judgment process:
(1) the vehicle is about to turn over when turning, and the vehicle angular velocity omega and the vehicle roll angle
Figure FDA0003150605800000053
The vehicle speed v parameters are correlated, and in order to identify the side-turning behavior, the vehicle angular speed omega and the vehicle roll angle of the labeled vehicle normal steering turning and the vehicle side-turning critical point are input into the LSTM model
Figure FDA0003150605800000054
Pre-training the vehicle speed v;
(2) inputting data obtained in real time into a trained model, and outputting the probability of whether the vehicle behavior is about to turn over;
the dangerous lane changing behavior judgment process comprises the following steps:
(1) the dangerous lane changing behavior associates the course angle psi, the vehicle speed v, the transverse displacement y and the angular speed omega of the vehicle with the Euclidean distance s parameters of the vehicle behind the adjacent lane, and inputs the course angle psi, the vehicle speed v, the transverse displacement y and the angular speed omega of the vehicle and the Euclidean distance s of the vehicle behind the adjacent lane when the labeled vehicle is normally changed and the dangerous lane is changed into an LSTM model for pre-training in order to identify the dangerous lane changing behavior of the vehicle;
(2) and inputting the data obtained in real time into the trained model, and outputting the probability of whether the vehicle behavior is dangerous lane change.
6. The system of claim 4, wherein the data after the secondary processing by the information processing center is directly sent to cluster-head vehicles, the T-BOX of the cluster-head vehicles performs block processing on the acquired information, the information corresponding to the cluster-head vehicles is respectively sent to the cluster-head vehicles, and finally the T-BOX of each vehicle sends the acquired prompt information to a display screen of a vehicle-mounted terminal to remind a driver of timely performing corresponding operations.
CN202010119070.4A 2020-02-26 2020-02-26 Highway vehicle abnormal behavior early warning system based on 5G communication Active CN111354193B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010119070.4A CN111354193B (en) 2020-02-26 2020-02-26 Highway vehicle abnormal behavior early warning system based on 5G communication

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010119070.4A CN111354193B (en) 2020-02-26 2020-02-26 Highway vehicle abnormal behavior early warning system based on 5G communication

Publications (2)

Publication Number Publication Date
CN111354193A CN111354193A (en) 2020-06-30
CN111354193B true CN111354193B (en) 2021-09-10

Family

ID=71197212

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010119070.4A Active CN111354193B (en) 2020-02-26 2020-02-26 Highway vehicle abnormal behavior early warning system based on 5G communication

Country Status (1)

Country Link
CN (1) CN111354193B (en)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112634646B (en) * 2020-12-11 2022-09-16 宜宾市天珑通讯有限公司 Intelligent traffic navigation system of 5G communication network
CN113207101B (en) * 2021-04-13 2021-12-07 山东曙光照信息技术有限公司 Information processing method based on 5G city component sensor and Internet of things cloud platform
CN113297441A (en) * 2021-05-13 2021-08-24 江苏南工科技集团有限公司 Multi-protocol fusion analysis method based on Internet of things technology
CN113470358A (en) * 2021-07-09 2021-10-01 南昌路兴交通工程监理咨询有限公司 Highway mobile monitoring system
CN113715834B (en) * 2021-08-24 2023-05-16 广州五舟软件技术有限公司 Vehicle driving behavior recognition method, device, electronic equipment and storage medium
CN114360293A (en) * 2022-01-12 2022-04-15 武汉中交交通工程有限责任公司 ETC technology-based expressway parking early warning method
CN114501378A (en) * 2022-03-06 2022-05-13 南京理工大学 Special environment emergency communication method and system based on vehicle-road cooperation
CN114898548B (en) * 2022-03-07 2023-03-14 北京汽车研究总院有限公司 Emergency lane occupation detection method, equipment, device and storage medium
CN115022821B (en) * 2022-06-20 2023-06-02 中国联合网络通信集团有限公司 Internet of vehicles safety communication method, RSU, vehicle, device and system
CN115909749B (en) * 2023-01-09 2023-07-14 广州通达汽车电气股份有限公司 Vehicle running road risk early warning method, device, equipment and storage medium

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN205177162U (en) * 2015-11-29 2016-04-20 吉林大学 Highway trackside early warning device
CN106627592A (en) * 2016-12-12 2017-05-10 深圳市元征科技股份有限公司 Early warning method, device and system for vehicle traveling
CN107454649A (en) * 2017-07-25 2017-12-08 西安交通大学 1 hop and 2 hop clusters in VANETs based on density estimation and the adaptive sub-clustering method deposited
CN107798897A (en) * 2016-09-06 2018-03-13 财团法人工业技术研究院 Roadside detection system, roadside device and roadside communication method thereof
CN109302709A (en) * 2018-09-14 2019-02-01 重庆邮电大学 The unloading of car networking task and resource allocation policy towards mobile edge calculations
CN109360426A (en) * 2018-11-23 2019-02-19 湖南车路协同智能科技有限公司 A kind of hypervelocity safe early warning method, device, system and readable storage medium storing program for executing
CN109886304A (en) * 2019-01-22 2019-06-14 江苏大学 A kind of complex road condition following peripheral vehicle behavior recognition methods based on HMM-SVM bilayer improved model
CN110049528A (en) * 2019-04-25 2019-07-23 华侨大学 Mobile trust data collection method based on trust value effectiveness in a kind of Sensor Network
CN110351686A (en) * 2018-04-02 2019-10-18 京东方科技集团股份有限公司 Method, on board unit and the car networking data transmission system implemented in car networking data transmission system

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20210007472A (en) * 2019-07-11 2021-01-20 엘지전자 주식회사 Vehicle terminal and operation method thereof

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN205177162U (en) * 2015-11-29 2016-04-20 吉林大学 Highway trackside early warning device
CN107798897A (en) * 2016-09-06 2018-03-13 财团法人工业技术研究院 Roadside detection system, roadside device and roadside communication method thereof
CN106627592A (en) * 2016-12-12 2017-05-10 深圳市元征科技股份有限公司 Early warning method, device and system for vehicle traveling
CN107454649A (en) * 2017-07-25 2017-12-08 西安交通大学 1 hop and 2 hop clusters in VANETs based on density estimation and the adaptive sub-clustering method deposited
CN110351686A (en) * 2018-04-02 2019-10-18 京东方科技集团股份有限公司 Method, on board unit and the car networking data transmission system implemented in car networking data transmission system
CN109302709A (en) * 2018-09-14 2019-02-01 重庆邮电大学 The unloading of car networking task and resource allocation policy towards mobile edge calculations
CN109360426A (en) * 2018-11-23 2019-02-19 湖南车路协同智能科技有限公司 A kind of hypervelocity safe early warning method, device, system and readable storage medium storing program for executing
CN109886304A (en) * 2019-01-22 2019-06-14 江苏大学 A kind of complex road condition following peripheral vehicle behavior recognition methods based on HMM-SVM bilayer improved model
CN110049528A (en) * 2019-04-25 2019-07-23 华侨大学 Mobile trust data collection method based on trust value effectiveness in a kind of Sensor Network

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
"A Comparative Study of State-of-the-Art Deep Learning Algorithms for Vehicle Detection";H. Wang;《IEEE Intelligent Transportation Systems Magazine》;20191231;第11卷(第2期);第82-95页 *
"城市环境下车联网路由算法研究";袁凯;《中国优秀硕士学位论文全文数据库 工程科技II辑》;20200215(第2期);第C034-227页 *
"基于改进BP网络的车载光通信网络路由设计";梁文祯等;《激光杂志》;20181231;第39卷(第8期);第71-75页 *
"采用64线激光雷达的实时道路障碍物检测与分类算法的研究";娄新雨等;《汽车工程》;20191231;第41卷(第8期);第779-784 *

Also Published As

Publication number Publication date
CN111354193A (en) 2020-06-30

Similar Documents

Publication Publication Date Title
CN111354193B (en) Highway vehicle abnormal behavior early warning system based on 5G communication
CN111325978B (en) Whole-process monitoring and warning system and method for abnormal behaviors of vehicles on expressway
CN107346612B (en) Vehicle anti-collision method and system based on Internet of vehicles
US20190171227A1 (en) Optimization of mission efficiency through platoon opportunity assessment
CN113593273B (en) No-signal control road intersection collision early warning method based on V2I communication
US10762363B2 (en) Road sign recognition for connected vehicles
CN100589143C (en) Method and appaatus for judging the traveling state of a floating vehicle
CN107830865A (en) A kind of vehicle target sorting technique, device, system and computer program product
CN112712697B (en) Lane-level traffic state discrimination method oriented to vehicle-road cooperative application
CN110647164A (en) Assisting flying drones in selecting and approaching vehicles to achieve improved flight distances
CN110329259B (en) Vehicle automatic following system and method based on multi-sensor fusion
US20200286382A1 (en) Data-to-camera (d2c) based filters for improved object detection in images based on vehicle-to-everything communication
CN104575035A (en) Intersection self-adaptation control method based on car networking environment
CN111402613B (en) Method for selecting lane of toll station for automatically driving vehicle
CN103164962A (en) Mountain road sharp turn section real-time vehicle speed early warning method
CN108749814A (en) A kind of intelligent driving vehicle travel control method
CN112907955A (en) Evaluation method of vehicle-road cooperative system based on information fusion
CN104773190A (en) Auxiliary traction control system for train
CN116013101B (en) System and method for suggesting speed of signal-free intersection based on network environment
CN112258850A (en) Edge side multi-sensor data fusion system of vehicle-road cooperative system
CN116434523A (en) Vehicle active safety control method and device based on constraint degree in information perception scene
US10930145B2 (en) Traffic system for predicting and providing traffic signal switching timing
CN114926984A (en) Real-time traffic conflict collection and road safety evaluation method
CN113240901B (en) Intelligent vehicle lane changing behavior risk level determination method and device
CN109827587A (en) A kind of intelligent route planning system of new-energy automobile

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
GR01 Patent grant
GR01 Patent grant