WO2021218235A1 - 一种道路交通拥堵预警方法及系统 - Google Patents

一种道路交通拥堵预警方法及系统 Download PDF

Info

Publication number
WO2021218235A1
WO2021218235A1 PCT/CN2021/070686 CN2021070686W WO2021218235A1 WO 2021218235 A1 WO2021218235 A1 WO 2021218235A1 CN 2021070686 W CN2021070686 W CN 2021070686W WO 2021218235 A1 WO2021218235 A1 WO 2021218235A1
Authority
WO
WIPO (PCT)
Prior art keywords
road
fuzzy
data
traffic
source
Prior art date
Application number
PCT/CN2021/070686
Other languages
English (en)
French (fr)
Inventor
张萌萌
黄基
于悦
Original Assignee
山东交通学院
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 山东交通学院 filed Critical 山东交通学院
Priority to US17/609,408 priority Critical patent/US20220222556A1/en
Publication of WO2021218235A1 publication Critical patent/WO2021218235A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/02—Computing arrangements based on specific mathematical models using fuzzy logic
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/40—Business processes related to the transportation industry
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/04—Inference or reasoning models
    • G06N5/048—Fuzzy inferencing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N7/00—Computing arrangements based on specific mathematical models
    • G06N7/02—Computing arrangements based on specific mathematical models using fuzzy logic
    • G06N7/023—Learning or tuning the parameters of a fuzzy system
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • 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
    • 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
    • G08G1/0112—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from the vehicle, e.g. floating car data [FCD]
    • 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
    • G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
    • 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
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • G08G1/0145—Measuring and analyzing of parameters relative to traffic conditions for specific applications for active traffic flow control
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/09—Arrangements for giving variable traffic instructions
    • G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
    • G08G1/0967—Systems involving transmission of highway information, e.g. weather, speed limits
    • G08G1/096708—Systems involving transmission of highway information, e.g. weather, speed limits where the received information might be used to generate an automatic action on the vehicle control
    • G08G1/096716—Systems involving transmission of highway information, e.g. weather, speed limits where the received information might be used to generate an automatic action on the vehicle control where the received information does not generate an automatic action on the vehicle control
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/09—Arrangements for giving variable traffic instructions
    • G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
    • G08G1/0967—Systems involving transmission of highway information, e.g. weather, speed limits
    • G08G1/096733—Systems involving transmission of highway information, e.g. weather, speed limits where a selection of the information might take place
    • G08G1/096741—Systems involving transmission of highway information, e.g. weather, speed limits where a selection of the information might take place where the source of the transmitted information selects which information to transmit to each vehicle
    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/09—Arrangements for giving variable traffic instructions
    • G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
    • G08G1/0967—Systems involving transmission of highway information, e.g. weather, speed limits
    • G08G1/096766—Systems involving transmission of highway information, e.g. weather, speed limits where the system is characterised by the origin of the information transmission
    • G08G1/096775—Systems involving transmission of highway information, e.g. weather, speed limits where the system is characterised by the origin of the information transmission where the origin of the information is a central station

Definitions

  • the present disclosure relates to the field of intelligent transportation technology, and in particular to a road traffic congestion warning method and system.
  • Perception and group experience play an important role in urban road congestion, ignoring the improvement of traffic control rules by the management experience of traffic managers; the construction of urban road traffic early warning platform does not take into account the complexity of the traffic system and the human traffic system It plays an important role in the early warning of congestion for drivers and does not form a complete congestion warning network.
  • the present disclosure proposes a road traffic congestion warning method and system, which collects multi-source traffic parameters of people, vehicles, roads and the environment, realizes multi-source data fusion through fuzzy logic inference and minimum variance weighted average method, and The algorithm of nuclear over-limit learning machine is used to calculate road congestion indicators.
  • congestion indicators are used to judge congestion, and a human-machine hybrid enhanced intelligent multi-source data fusion system that uses the wisdom of road participant groups is constructed.
  • the present disclosure provides a road traffic congestion warning method, including:
  • the feature membership functions are fuzzy weighted and averaged, and the weighted average membership functions of different feature quantities are defuzzified to obtain multiple Source fusion traffic data;
  • the present disclosure provides a road traffic congestion warning system, including:
  • the first fuzzy weight calculation module is used to classify features according to the acquired multi-source traffic data, and construct the corresponding feature membership function, and use the minimum weighted average algorithm for the feature membership function to obtain the first fuzzy weight;
  • the second fuzzy weight calculation module is used to construct an artificial membership function using expert evaluation method for multi-source data, and calculate the second fuzzy weight
  • the fusion module is used to perform fuzzy weighted average on the feature membership function according to the fused fuzzy weight obtained after the fusion of the first fuzzy weight and the second fuzzy weight, and perform the weighted average membership function of the obtained different feature quantities Unblur, obtain multi-source fusion traffic data;
  • the model building module is used to build a road traffic congestion model using the nuclear overrun learning machine group algorithm for multi-source integrated traffic data, and calculate the best road traffic congestion index;
  • the congestion warning module is used to obtain current multi-source traffic data, predict the current congestion index according to the road traffic congestion model, and compare with the best road traffic congestion index to provide early warning of whether the current road is congested.
  • the present disclosure provides an electronic device, including a memory, a processor, and computer instructions stored in the memory and running on the processor. When the computer instructions are executed by the processor, the method described in the first aspect is completed. .
  • the present disclosure provides a computer-readable storage medium for storing computer instructions that, when executed by a processor, complete the method described in the first aspect.
  • This disclosure analyzes the characteristics of different road congestion indexes from four aspects of road traffic, combines the management experience of traffic managers with machine intelligence through the traffic management platform, and builds an intelligent congestion warning platform based on man-machine hybrid enhanced management and control rule base; Introduce the role of humans into the calculation loop of the congestion warning system, and closely couple the ability of humans to deal with fuzzy and uncertain problems with the ability of precision calculations of machines to form an advanced cognitive response mechanism for human-machine collaborative work, two-way communication and control of information , Which combines human perception and cognitive ability with powerful computing and storage capabilities of computers to form a '1+1>2' human-machine hybrid enhanced intelligent form.
  • This disclosure organically merges human and machine intelligence.
  • Road traffic participants obtain road traffic data and road feature information through their own senses and verify each other with real-time data provided by each sensor, and make independent judgments based on the experience provided by verification experts.
  • the ability of the machine to quickly and accurately calculate and the ability of humans to deal with fuzzy problems are brought into play, and the reliability and flexibility of the system will be greatly improved.
  • FIG. 1 is a flowchart of a human-machine hybrid enhanced intelligent multi-source data fusion subsystem provided in Embodiment 1 of the disclosure;
  • FIG. 2 is a diagram of a data collection sub-module of a sensing device provided in Embodiment 1 of the disclosure
  • FIG. 3 is a diagram of a sub-module of traffic participant data collection provided by Embodiment 1 of the present disclosure
  • FIG. 4 is a diagram of a human-machine hybrid enhanced multi-source data acquisition subsystem provided by Embodiment 1 of the disclosure
  • FIG. 6 is a general diagram of an urban road traffic early warning platform based on human-machine hybrid enhanced intelligence provided by Embodiment 1 of the disclosure.
  • this embodiment provides a road traffic congestion warning method, including:
  • S1 Perform feature classification according to the acquired multi-source traffic data, and construct the corresponding feature membership function, and use the minimum weighted average algorithm for the feature membership function to obtain the first fuzzy weight;
  • S2 Use expert evaluation method to construct artificial membership function for multi-source data, and calculate the second fuzzy weight
  • S3 Perform fuzzy weighted average on the feature membership function according to the fused fuzzy weight obtained after the first fuzzy weight and the second fuzzy weight are fused, and defuzzify the weighted average membership function of the obtained different feature quantities, Obtain multi-source fusion traffic data;
  • S5 Obtain current multi-source traffic data, predict the current congestion index based on the road traffic congestion model, and compare with the best road traffic congestion index to warn whether the current road is congested.
  • step S1 feature classification is performed according to the acquired multi-source traffic data to obtain road features, human features, environmental features, and Car characteristics
  • the road characteristics include traffic flow, number of lanes and road grade
  • the human characteristics include driver behavior characteristics and pedestrian behavior characteristics, that is, driver's road familiarity and driving skills, mental state, driving habits, reaction time, pedestrian and driver traffic violation information records, and pedestrian crossing time;
  • the environmental characteristics include road weather and traffic accident information, understandably, and can also include information on large-scale activities around the road, etc.;
  • the vehicle characteristics include the position of the vehicle, speed, headway distance, and vehicle condition.
  • the process of collecting multi-source data consists of various sensing devices and traffic participants.
  • the fixed sensing device laid on the road network collects traffic flow.
  • Road traffic data such as the number of lanes, vehicle speed, road weather, traffic accidents, etc.
  • fixed sensor equipment includes traffic sensor equipment such as infrared, geomagnetism, radar, coil, and video;
  • the mobile sensor equipment installed on the vehicle collects vehicle traffic data such as the position of the vehicle on the road, vehicle acceleration, head distance, driver operation behavior, and driver behavior characteristic data; mobile sensor equipment includes car navigation equipment, license plate recognition Traffic sensing equipment such as equipment.
  • the data collected by traffic participants mainly includes human perception, management experience, and policy activity information; the first is to use pedestrians on urban roads to provide their trajectories, passing time through intersections, sections with frequent accidents, and roads.
  • Basic information on the road such as infrastructure conditions; the second is to rely on the driver of the driving vehicle to provide the driver’s perception information of the driver’s road familiarity, driving skills, mental state, driving habits, and reaction time; the third is to provide traffic congestion through the traffic manager Status determination rules, road traffic management experience, road section real-time traffic information, congestion handling policy, urban traffic control plan and surrounding large-scale activities and other management and control policy experience information.
  • a human-machine hybrid urban road traffic information collection platform is established by using sensor equipment installed on urban roads and vehicles and traffic participants to give full play to the advantages of human-machine hybrid enhanced intelligence.
  • the preprocessed data is multi-source data fusion, which specifically includes:
  • the pre-processed multi-source data is cleaned twice, obviously unreasonable data can be eliminated for problematic data, and experienced traffic managers and data processing experts are invited to perform data analysis, and give full play to the human-machine hybrid and enhanced intelligence. Advantage.
  • the corresponding feature membership function is constructed, including:
  • the minimum variance weighted average algorithm that minimizes the signal variance is used to obtain the first fuzzy weight ⁇ i corresponding to the minimum total mean square error, including:
  • Order x i (a) d i (a) + b i (a), d i (a) is the true value of the signal, B i (a) is a Gaussian characteristic of the noise a time the i-th signal, which corresponds to the variance Is ⁇ i 2 ;
  • W ⁇ 1 , ⁇ 2 ,..., ⁇ i ⁇ is an unknown weight matrix, which satisfies
  • X ⁇ x 1 , x 2 ,..., x i ⁇ is the data collected by different collection methods at time a, and the variance ⁇ i 2 can be written as
  • the step S2 specifically includes: establishing the fuzzy weight value of the artificial intelligence membership degree by inviting the expert group based on the experience of its traffic management experience
  • the first fuzzy weight and the second fuzzy weight are fused to obtain the fused fuzzy weight
  • Use membership function represents fuzzy weight According to the fuzzy weight, different membership functions are fuzzy weighted, and the weighted average membership functions of different feature quantities are obtained.
  • the center of gravity method is adopted, namely Perform defuzzification to obtain multi-source fusion traffic data, that is, traffic flow Q, reaction time T, vehicle speed V, headway distance L, and vehicle acceleration a.
  • constructing a road traffic congestion model specifically includes:
  • the calculation of the optimal road traffic congestion index includes:
  • ⁇ i [ ⁇ i1 , ⁇ i2 ,..., ⁇ in ] T be the input weight of the i-th hidden layer node;
  • ⁇ i [ ⁇ i1 , ⁇ i2 ,..., ⁇ in ] T is the input weight of the i-th hidden layer node;
  • the output weight of i hidden layer nodes, b i is the bias of the i-th hidden layer node, and ⁇ i *x j is the inner product of ⁇ i and x j;
  • the excitation function is infinitely differentiable, that is, the weights of the input layer and the hidden layer bias can be randomly assigned, and the input weight ⁇ i and the hidden layer bias b i are fixed to train the single hidden layer feedforward neural network. ;
  • the least square solution can only exist when min ⁇ H ⁇ -T ⁇ and min ⁇ are satisfied at the same time.
  • H + is the augmented inverse matrix of the hidden layer matrix H. If it is assumed that the output function h(x) of the hidden layer node is unknown, the kernel function is introduced into it to form the core overrun learning machine group algorithm;
  • a periodic function is p is the period of the kernel function, and the form of the period kernel obtained by the RBF kernel is It can be written as follows:
  • K(x i ,y j ) is the kernel function
  • the output formula of KELM can be written in the following form:
  • the biggest advantage of the model is that unknowns such as the number of hidden layer nodes, initial weights, and offsets need not be considered when solving, and the inner product form of the kernel function and the kernel function K( ⁇ , ⁇ ) are directly used.
  • the value of the prediction function can be calculated in a specific form, and the best road traffic congestion index Y'can be obtained conveniently and quickly.
  • the minimum weighted average algorithm and fuzzy judgment reasoning are used to combine human traffic management experience with the machine’s fast and accurate computing capabilities to calculate accurate road traffic fusion data; and use these data to extract road features and use nuclear super
  • the limited learning machine group algorithm predicts the best road traffic congestion index Y′, giving full play to the advantages of human-machine hybrid intelligent enhancement, and quickly and accurately predicting the road congestion index provides strong data support for the human-machine hybrid enhanced intelligent congestion warning platform.
  • step S5 the process of judging the road congestion warning is as follows:
  • the current congestion index is predicted according to the road traffic congestion model
  • the road congestion warning signal can be sent to the cloud of the car navigation by the on-board communication unit and the road test network connection facility at the same time to detect the driving speed of the vehicle on the road section. If the vehicle arrives at the road section, adjust the road network signal timing Plan, carry out congestion warning, and provide a new planned route.
  • the traffic management department can allow drivers to participate in on-site inspections of congested road sections through real-name authentication of the mobile app. Feeling as the criterion to ensure the highest human control, the cloud clears the road congestion signal.
  • the car navigation or mobile phone app can be used to help the driver get the traffic information of the road section ahead, and the early warning system will determine the congestion state according to the congestion status of the road section; at the same time, it can detect the real-time speed of the vehicle and predict it. After judging how long the vehicle will drive into the congested section and give several reasonable routes to avoid the congestion; at the same time, the large LED screen at the intersection in front of the congested section will give information on the congestion situation, traffic flow and other information of the road section, in preparation for entering Vehicles on congested road sections are given timely congestion warning.
  • the traffic management center introduces the experience and wisdom of traffic managers and the professional analysis of experts, establishes a new signal optimization model, and issues the real-time calculation of the optimal signal tuning plan to each An intersection reasonably regulates the signal light phase of each intersection, forms a green wave band as much as possible, and establishes a regionally linked signal timing optimization network; at the same time, each intersection has certain self-optimization and control capabilities.
  • the green wave of the main road is given priority to avoid the occurrence of greater congestion, and a distributed intelligent management and control platform with adaptive capabilities is formed.
  • This embodiment realizes the collaboration of machine intelligence and human intelligence through the Internet brain regulating the traffic signal phases and roadside warning facilities of each intersection, and integrates machine intelligence and swarm intelligence into urban roads to jointly solve the problem of urban congestion in congestion warning.
  • This embodiment provides a road traffic jam early warning system, including:
  • the first fuzzy weight calculation module is used to classify features according to the acquired multi-source traffic data, and construct the corresponding feature membership function, and use the minimum weighted average algorithm for the feature membership function to obtain the first fuzzy weight;
  • the second fuzzy weight calculation module is used to construct an artificial membership function using expert evaluation method for multi-source data, and calculate the second fuzzy weight
  • the fusion module is used to perform fuzzy weighted average on the feature membership function according to the fused fuzzy weight obtained after the fusion of the first fuzzy weight and the second fuzzy weight, and perform the weighted average membership function of the obtained different feature quantities Unblur, obtain multi-source fusion traffic data;
  • the model building module is used to build a road traffic congestion model using the nuclear overrun learning machine group algorithm for multi-source integrated traffic data, and calculate the best road traffic congestion index;
  • the congestion warning module is used to obtain current multi-source traffic data, predict the current congestion index according to the road traffic congestion model, and compare with the best road traffic congestion index to provide early warning of whether the current road is congested.
  • each of the foregoing modules corresponds to steps S1 to S5 in Embodiment 1, and the implementation examples and application scenarios of the foregoing modules and corresponding steps are the same, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above-mentioned modules can be executed in a computer system such as a set of computer-executable instructions as a part of the system.
  • this embodiment provides an early warning platform, including a human-machine hybrid enhanced intelligent multi-source data collection subsystem, a human-machine hybrid enhanced intelligent multi-source data fusion subsystem, and a human-machine hybrid enhanced intelligent congestion early warning subsystem;
  • the human-machine hybrid enhanced intelligent multi-source data acquisition subsystem is composed of various sensing devices and traffic participants;
  • sensing equipment include fixed sensing equipment laid on the road network and mobile sensing equipment installed on vehicles, which collect road traffic data such as traffic flow, number of lanes, vehicle speed, road weather, and traffic accident conditions; collect road sections Vehicle traffic data such as the location of the vehicle on board, vehicle acceleration, headway distance, driver operation behavior and other vehicle traffic data and driver behavior characteristic data.
  • road traffic data such as traffic flow, number of lanes, vehicle speed, road weather, and traffic accident conditions
  • Vehicle traffic data such as the location of the vehicle on board, vehicle acceleration, headway distance, driver operation behavior and other vehicle traffic data and driver behavior characteristic data.
  • the data collected by traffic participants mainly includes human perception, management experience, and policy activity information; the first is to use pedestrians on urban roads to provide road foundations such as their trajectories, transit time through intersections, sections with frequent accidents, and road infrastructure conditions. Information; the second is to rely on the driver of the driving vehicle to provide the driver’s perception information of the driver’s road familiarity and driving skills, mental state, driving habits, reaction time, etc.; the third is to provide traffic congestion status determination rules and road traffic through the traffic manager Management experience, real-time traffic information on road sections, congestion handling policies, urban traffic control plans and surrounding large-scale activities and other management and control policy experience information.
  • the human-machine hybrid enhanced multi-source data acquisition subsystem uses the sensor equipment installed on urban roads and vehicles and traffic participants to establish a human-machine hybrid urban road traffic information acquisition platform.
  • the subsystem converts the traffic data provided by traffic participants into App points through the mobile phone APP. These points can be used to redeem small gifts such as gas cards or high-speed transit coupons to encourage traffic participants to provide us with perceptual information.
  • a local data collection platform of vehicle-vehicle, vehicle-road, vehicle-person, vehicle-infrastructure is formed by the interconnection among the vehicles on the road, traffic participants, and road infrastructure, which greatly improves the collection of different data
  • the environment perception ability of the source obtains more accurate road traffic data.
  • the human-machine hybrid enhanced intelligent multi-source data fusion subsystem performs simple classification processing on the traffic data collected by the data collection subsystem, and then the classified data is transmitted to the data processing department corresponding to the smart city big data center for multi-source Data preprocessing of heterogeneous data.
  • the congestion index calculated by the multi-source data fusion subsystem is used to judge the congestion.
  • the congestion index of the monitored section is greater than the congestion index, the driver or pedestrian is trafficked by adjusting the signal timing plan of each section of the city. Induce and solve the congestion problem of urban roads;
  • the man-machine hybrid enhanced intelligent congestion early warning subsystem is composed of road early warning facilities and vehicle early warning equipment.
  • the multi-source signal data fusion module provides traffic parameters of the road section.
  • the on-board communication unit and the roadside The networked facilities also send road congestion signals to the cloud of car navigation;
  • the traffic management department can allow drivers to participate in on-site inspections of congested roads through mobile app real-name authentication.
  • the feelings of most drivers will be used to ensure that people Has the highest control, the cloud clears the road congestion signal.
  • the car navigation or mobile phone app can be used to help the driver get the traffic information of the road section ahead, and the early warning system will determine the congestion state according to the congestion condition of the road section; at the same time, it detects the real-time speed of the vehicle and predicts how long the vehicle will be. It will drive into the congested section and give several reasonable routes to avoid congestion; at the same time, the large LED screen at the intersection in front of the congested section will give information on the congestion condition and traffic flow of the road section, and provide timely information for vehicles preparing to enter the congested section. Early warning of congestion.
  • the traffic management center introduces the experience and wisdom of traffic managers and the professional analysis of experts, establishes a new signal optimization model, and issues the real-time calculation of the optimal signal tuning plan to each An intersection reasonably regulates the signal light phase of each intersection, forms a green wave band as much as possible, and establishes a regionally linked signal timing optimization network; at the same time, each intersection has certain self-optimization and control capabilities.
  • the green wave of the main road is given priority to avoid the occurrence of greater congestion, and a distributed intelligent management and control platform with adaptive capabilities is formed.
  • the platform realizes the collaboration of machine intelligence and human intelligence through the Internet brain to control the phase of traffic signals and roadside warning facilities at each intersection, and integrates machine intelligence and group intelligence into urban roads to jointly solve the problem of urban congestion in congestion warning ;
  • the platform utilizes human group intelligence and machine intelligence to cooperate with each other, and gives full play to the super-computing ability of the machine to process large amounts of data and the ability of road traffic participants and managers to judge road congestion, forming a human-oriented, man-machine combination Human-machine hybrid enhanced intelligent road warning system.
  • An electronic device includes a memory, a processor, and computer instructions stored on the memory and running on the processor.
  • the computer instructions are executed by the processor, the method described in Embodiment 1 is completed. For the sake of brevity, I will not repeat them here.
  • the processor may be the central processing unit CPU, the processor may also be other general-purpose processors, digital signal processors DSP, application-specific integrated circuits ASIC, ready-made programmable gate array FPGAs or other programmable logic devices. , Discrete gates or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like.
  • the memory may include a read-only memory and a random access memory, and provides instructions and data to the processor, and a part of the memory may also include a non-volatile random access memory.
  • the memory can also store device type information.
  • a computer-readable storage medium is used to store computer instructions, and when the computer instructions are executed by a processor, the method described in Embodiment 1 is completed.
  • Embodiment 1 may be directly embodied as being executed by a hardware processor, or executed by a combination of hardware and software modules in the processor.
  • the software module may be located in a mature storage medium in the field, such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, registers.
  • the storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.

Landscapes

  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Business, Economics & Management (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Economics (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Fuzzy Systems (AREA)
  • Automation & Control Theory (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Atmospheric Sciences (AREA)
  • Primary Health Care (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Biomedical Technology (AREA)
  • Mathematical Analysis (AREA)
  • Computational Mathematics (AREA)
  • Molecular Biology (AREA)
  • Algebra (AREA)
  • Operations Research (AREA)
  • Development Economics (AREA)
  • Quality & Reliability (AREA)
  • Entrepreneurship & Innovation (AREA)

Abstract

本公开公开了一种道路交通拥堵预警方法及系统，包括根据获取的多源交通数据进行特征分类，并构建相应的特征隶属度函数，得到第一模糊权值；对多源数据采用专家评价法构建人工隶属度函数，并计算第二模糊权值；根据第一模糊权值和第二模糊权值融合后得到的融合模糊权值，对特征隶属度函数进行模糊加权平均，对得到的不同特征量的加权平均值隶属度函数进行解模糊，得到多源融合交通数据；采用核超限学习机群组算法构建道路交通拥堵模型，计算最佳道路交通拥堵指数；获取当前多源交通数据，预测当前拥堵指数，通过与最佳道路交通拥堵指数进行比较，对当前道路是否拥堵进行预警。构建了发挥道路参与者群体智慧的人机混合增强智能多源数据融合体系。

Description

一种道路交通拥堵预警方法及系统 技术领域
本公开涉及智能交通技术领域，特别是涉及一种道路交通拥堵预警方法及系统。
背景技术
本部分的陈述仅仅是提供了与本公开相关的背景技术信息，不必然构成在先技术。
近年来，随着道路行驶车辆的增多，使得道路上的行车环境更加复杂，交通拥堵的几率大大增加，车辆在道路上的行车安全遭遇极大威胁，而且行车速度的下降和到达目的地时间也会增加；在道路运行环境中，车辆的行驶状态受多种因素的影响，例如车辆状态、天气环境、行人或驾驶员的行为等；而发明人发现，传统的道路交通预警平台存在部分缺陷，传统的道路拥堵预警平台多以单一的传感设备作为数据采集来源，仅考虑路况或车辆本身的影响，数据的采集端采集的数据不够完善；道路的拥堵程度评价指标过于简单，并未考虑人类感知和群体经验在城市道路拥堵中的重要作用，忽略了交通管理者管理经验对交通管控规则的完善；在城市道路交通预警平台的构建中并未考虑到交通系统的复杂性和人在交通系统中的重要作用，对驾驶员的拥堵预警并未形成完整的拥堵预警网络。
发明内容
为了解决上述问题，本公开提出了一种道路交通拥堵预警方法及系统，采集人、车、路和环境的多源交通参数，通过模糊逻辑推断和最小方差加权平均法实现多源数据融合，并采用核超限学习机算法计算道路拥堵指标，在道路拥堵预警阶段，利用拥堵指标进行拥堵判断，构建了发挥道路参与者群体智慧的人机混合增强智能多源数据融合体系。
为了实现上述目的，本公开采用如下技术方案：
第一方面，本公开提供一种道路交通拥堵预警方法，包括：
根据获取的多源交通数据进行特征分类，并构建相应的特征隶属度函数，对特征隶属度函数采用最小加权平均算法得到第一模糊权值；
对多源数据采用专家评价法构建人工隶属度函数，并计算第二模糊权值；
根据第一模糊权值和第二模糊权值融合后得到的融合模糊权值，对特征隶属度函数进行模糊加权平均，对得到的不同特征量的加权平均值隶属度函数进行解模糊，得到多源融合交通数据；
对多源融交通合数据采用核超限学习机群组算法构建道路交通拥堵模型，计算最佳道路 交通拥堵指数；
获取当前多源交通数据，根据道路交通拥堵模型预测当前拥堵指数，通过与最佳道路交通拥堵指数进行比较，对当前道路是否拥堵进行预警。
第二方面，本公开提供一种道路交通拥堵预警系统，包括：
第一模糊权值计算模块，用于根据获取的多源交通数据进行特征分类，并构建相应的特征隶属度函数，对特征隶属度函数采用最小加权平均算法得到第一模糊权值；
第二模糊权值计算模块，用于对多源数据采用专家评价法构建人工隶属度函数，并计算第二模糊权值；
融合模块，用于根据第一模糊权值和第二模糊权值融合后得到的融合模糊权值，对特征隶属度函数进行模糊加权平均，对得到的不同特征量的加权平均值隶属度函数进行解模糊，得到多源融合交通数据；
模型构建模块，用于对多源融交通合数据采用核超限学习机群组算法构建道路交通拥堵模型，计算最佳道路交通拥堵指数；
拥堵预警模块，用于获取当前多源交通数据，根据道路交通拥堵模型预测当前拥堵指数，通过与最佳道路交通拥堵指数进行比较，对当前道路是否拥堵进行预警。
第三方面，本公开提供一种电子设备，包括存储器和处理器以及存储在存储器上并在处理器上运行的计算机指令，所述计算机指令被处理器运行时，完成第一方面所述的方法。
第四方面，本公开提供一种计算机可读存储介质，用于存储计算机指令，所述计算机指令被处理器执行时，完成第一方面所述的方法。
与现有技术相比，本公开的有益效果为：
本公开从道路交通四方面因素对不同道路拥堵指数进行特征分析，通过交通管理平台将交通管理者的管理经验与机器智能相结合，构建一个基于人机混合增强管控规则库的智能拥堵预警平台；把人的作用引入到拥堵预警系统的计算回路中，将人处理模糊、不确定问题的能力与机器精密计算的能力紧密耦合，形成人机协同工作、信息双向交流与控制的高级认知响应机制，使人的感知、认知能力和计算机强大的运算和存储能力相结合，构成‘1+1>2’的人机混合增强智能形态。
本公开将人和机器智能有机合并，道路交通参与者通过自身的感官获取道路交通数据和道路特征信息并与每个传感器提供的实时数据进行相互验证，根据验证专家提供的经验进行独立判断，充分的发挥了机器快速精确计算的能力和人处理模糊问题的能力，系统的可靠性 和灵活性将大幅提高。
附图说明
构成本公开的一部分的说明书附图用来提供对本公开的进一步理解，本公开的示意性实施例及其说明用于解释本公开，并不构成对本公开的不当限定。
图1为本公开实施例1提供的人机混合增强智能多源数据融合子系统流程图；
图2为本公开实施例1提供的传感设备数据采集子模块图；
图3为本公开实施例1提供的交通参与者数据采集子模块图；
图4为本公开实施例1提供的人机混合增强多源数据采集子系统图；
图5为本公开实施例1提供的人机混合增强智能拥堵预警子系统流程图；
图6为本公开实施例1提供的基于人机混合增强智能的城市道路交通预警平台总图。
具体实施方式：
下面结合附图与实施例对本公开做进一步说明。
应该指出，以下详细说明都是例示性的，旨在对本公开提供进一步的说明。除非另有指明，本文使用的所有技术和科学术语具有与本公开所属技术领域的普通技术人员通常理解的相同含义。
需要注意的是，这里所使用的术语仅是为了描述具体实施方式，而非意图限制根据本公开的示例性实施方式。如在这里所使用的，除非上下文另外明确指出，否则单数形式也意图包括复数形式，此外，还应当理解的是，当在本说明书中使用术语“包含”和/或“包括”时，其指明存在特征、步骤、操作、器件、组件和/或它们的组合。
实施例1
如图1所示，本实施例提供一种道路交通拥堵预警方法，包括：
S1：根据获取的多源交通数据进行特征分类，并构建相应的特征隶属度函数，对特征隶属度函数采用最小加权平均算法得到第一模糊权值；
S2：对多源数据采用专家评价法构建人工隶属度函数，并计算第二模糊权值；
S3：根据第一模糊权值和第二模糊权值融合后得到的融合模糊权值，对特征隶属度函数进行模糊加权平均，对得到的不同特征量的加权平均值隶属度函数进行解模糊，得到多源融合交通数据；
S4：对多源融交通合数据采用核超限学习机群组算法构建道路交通拥堵模型，计算最佳道路交通拥堵指数；
S5：获取当前多源交通数据，根据道路交通拥堵模型预测当前拥堵指数，通过与最佳道路交通拥堵指数进行比较，对当前道路是否拥堵进行预警。
在本实施例中，将人、车、路和环境四方面作为道路交通拥堵的影响因素，在步骤S1中，根据获取的多源交通数据进行特征分类后得到路特征、人特征、环境特征和车特征；
其中，所述路特征包括交通流量、车道数和道路等级；
所述人特征包括驾驶员行为特征和行人行为特征，即驾驶员道路熟悉度和驾驶技术、精神状态、驾驶习惯、反应时间、行人与驾驶员的交通违法信息记录、行人路口通行时间；
所述环境特征包括道路天气和交通事故信息，可以理解的，还可包括道路周边大型活动信息等；
所述车特征包括车辆的位置、车速、车头间距和车况。
在本实施例中，如图2所示，对多源数据的采集过程由各类传感设备和交通参与者组成，车辆在城市道路行驶时，路网上铺设的固定式传感设备采集交通流量、车道数、车速、道路天气、交通事故发生状况等道路交通数据；固定式传感设备包括红外、地磁、雷达、线圈、视频等交通传感设备；
车辆上安装的移动式传感设备采集路段上车辆的位置、车辆加速度、车头间距、驾驶员操作行为等车辆交通数据及驾驶员的行为特征数据；移动式传感设备包括车载导航设备、车牌识别设备等交通传感设备。
如图3所示，交通参与者采集的数据主要有人类感知、管理经验和政策活动信息三大类；一是利用城市道路上的行人提供其运动轨迹、通过路口通行时间、多发事故路段、道路基础设施状况等道路基础信息；二是依靠行驶车辆的驾驶员提供驾驶员道路熟悉度和驾驶技术、精神状态、驾驶习惯、反应时间等驾驶员的感知信息；三是通过交通管理者提供交通拥堵状态判定规则、道路交通管理经验、路段实时交通信息、拥堵处理政策、城市交通管控方案和周边大型活动情况等管控政策经验信息。
在本实施例中，利用城市道路和车辆安装的传感设备和交通参与者建立一个人机混合的城市道路交通信息采集平台，充分发挥人机混合增强智能优势。
在本实施例中，如图4所示，获取多源交通数据并分类后，对各类特征数据进行预处理，对预处理后的数据进行多源数据融合，具体包括：
对于多源异构数据，由于采集的数据种类不同，所采用的数据处理方法也不同，对无法直接量化的天气等道路特征信息和精神状态等人的感知数据利用交通管理者的经验或者专家 的意见为其设置等级标准；
将不易处理的文字图片信息转化为数字，再将其与车速、车头间距等数字化数据合并处理，利用缺失数据剔除、近似值修复补齐、错误数据校正和较大的数据归一化处理等方法进行数据预处理；
对预处理后的多源数据进行二次清洗，对出现问题的数据可将明显不合理的数据剔除，并邀请经验丰富的交通管理者与数据处理专家进行数据辨析，充分发挥人机混合增强智能优势。
所述步骤S1中，特征分类完成后，构建相应的特征隶属度函数，包括：
(1)以人特征、车特征、路特征、环境特征建立四个不同的特征论域A JI、B JI、C JI、D JI，
(2)根据每个特征论域中包含的特征数据，设交通流量、车道数、道路等级数据为A 11,A 12,...,A 1I、A 21,A 22,...,A 2I、A 31,A 32,...,A 3I；驾驶员道路熟悉度和驾驶技术、精神状态、驾驶习惯、反应时间、行人与驾驶员的交通违法信息记录、行人路口通行时间数据为B 11,B 12,...,B 1I、B 21,B 22,...,B 2I、B 31,B 32,...,B 3I、B 41,B 42,...,B 4I、B 51,B 52,...,B 5I、B 61,B 62,...,B 6I；车辆的位置、车速、车头间距、车况数据为C 11,C 12,...,C 1I、C 21,C 22,...,C 2I、C 31,C 32,...,C 3I、C 41,C 42,...,C 4I；道路天气、道路周边大型活动信息和实时交通事故数据为D 11,D 12,...,D 1I、D 21,D 22,...,D 2I、D 31,D 32,...,D 3I；
(3)以各个特征论域为输入变量，利用交通参与者与交通管理者的群体经验和机器智能将定量数据划分出不同等级，并依据不同等级标准设置不同区域范围进行定性分析，建立模糊推理规则表A ij,B ij,C ij,D ij；
(4)运用模糊推理规则表建立对应模糊推理等级论域的模糊子集
Figure PCTCN2021070686-appb-000001
Figure PCTCN2021070686-appb-000002
(5)通过模糊映射得出人、车、路、环境模糊子集对应的特征隶属度函数为：
Figure PCTCN2021070686-appb-000003
对特征隶属度函数利用信号方差最小化的最小方差加权平均算法得到对应总均方差最小时的第一模糊权值ω i，包括：
(1)在a时刻，设人、车、路、环境四个不同的特征论域检测到的信号为x 1(a)、x 2(a)、x 3(a)、x 4(a)；
令x i(a)＝d i(a)+b i(a)，d i(a)是信号的真实值，b i(a)是a时刻第i个信号的高斯特性噪声，其对应方差为σ i 2；
(2)不同数据源得到的信息加权平均结果为：
Figure PCTCN2021070686-appb-000004
W＝{ω 1，ω 2，...,ω i}为未知权值矩阵，满足
Figure PCTCN2021070686-appb-000005
X＝{x 1，x 2，...,x i}为a时刻不同采集方法采集到的数据，方差σ i 2可记为
Figure PCTCN2021070686-appb-000006
Figure PCTCN2021070686-appb-000007
(3)运用柯西不等式得出公式：
Figure PCTCN2021070686-appb-000008
据该公式推断出，当且仅当ω 1σ 1 2＝ω 2σ 2 2＝...＝ω iσ i 2，满足
Figure PCTCN2021070686-appb-000009
取得最小值时，对应总的均方差也取最小极值；
(4)利用多元函数极值的方法，计算出总均方差最小时隶属度的模糊权值：
Figure PCTCN2021070686-appb-000010
所述步骤S2中，具体包括：通过邀请专家组依据其交通管理经验的经验建立人工智能隶属度的模糊权值为
Figure PCTCN2021070686-appb-000011
所述步骤S3中，第一模糊权值和第二模糊权值融合后得到融合模糊权值
Figure PCTCN2021070686-appb-000012
采用隶属度函数
Figure PCTCN2021070686-appb-000013
表示模糊权重
Figure PCTCN2021070686-appb-000014
根据模糊权重，对不同的隶属度函数模糊加权，得到不同特征量的加权平均值隶属度函数
Figure PCTCN2021070686-appb-000015
Figure PCTCN2021070686-appb-000016
所述步骤S3中，采用重心法，即
Figure PCTCN2021070686-appb-000017
进行解模糊，得到多源融合交通数据，即交通流量Q、反应时间T、车速V、车头间距L、车辆加速度a。
所述步骤S4中，构建道路交通拥堵模型具体包括：
(1)选取交通流量Q、反应时间T、车速V、车头间距L、相邻路段交通拥堵指数Y等道路拥堵影响因素为输入样本；
(2)将不同特征的输入样本输入不同的核超限学习子模型中训练，为每个路段生成一个独立子模型，同时进行并行运算处理，形成能够预测整个路网的拥堵指数的道路交通网络模型。
计算最佳道路交通拥堵指数具体包括：
设5个不重复的输入样本为(x i,t i)，则x i＝[x i1,x i2,...,x in] T∈R 5是一个5维的输入；同时设t i＝[t i1,t i2,...,t in] T∈R m是输入x i对应的m维输出，且该模型有
Figure PCTCN2021070686-appb-000018
个隐层节点，激励函数g(x)记为
Figure PCTCN2021070686-appb-000019
设ω i＝[ω i1,ω i2,...,ω in] T是第i个隐层节点的输入权重；β i＝[β i1,β i2,...,β in] T是第i个隐层节点的输出权重，b i是第i个隐层节点的偏置，ω i*x j是ω i和x j的内积；
当神经网络输入与输出完全拟合，即误差为
Figure PCTCN2021070686-appb-000020
时，则存在β i、ω i和b i使得
Figure PCTCN2021070686-appb-000021
得到最佳输出；此时H为隐层节点的输出矩阵记为Hβ＝T；
Figure PCTCN2021070686-appb-000022
同时根据超限学习机理论，激励函数是无限可微，即输入层的权重和隐层偏置可以随机赋值，固定输入权重ω i和隐层偏置b i训练该单隐层前馈神经网络；
当线性系统Hβ＝T中的一个满足最小二乘的
Figure PCTCN2021070686-appb-000023
即
Figure PCTCN2021070686-appb-000024
因为在大多数情况下，隐层节点的数量
Figure PCTCN2021070686-appb-000025
和输入的不重复的训练样本数N不相等，即
Figure PCTCN2021070686-appb-000026
时；
此时可以转向求使损失函数‖Hβ-T‖最小的β，即
Figure PCTCN2021070686-appb-000027
根据极小范数解准则，即同时满足min‖Hβ-T‖和min‖β‖才存在最小二乘解
Figure PCTCN2021070686-appb-000028
Figure PCTCN2021070686-appb-000029
H +是隐层矩阵H的增广逆矩阵，若假设隐层节点的输出函数h(x)是未知的，将核函数引入其中形成核超限学习机群组算法；
超限学习机算法的随机矩阵H TH被核矩阵代替，建立不同的核函数的核超限学习机模型；根据核函数理论对和函数进行分类，核函数K(μ,ν)＝包括RBF核函数、线性核函数和多项式核函数等。
通常设定RBF核为K(μ,ν)＝exp[-(μ-ν 2/γ)]为核函数，但对明显带有周期性的特征输 入构造子模型时加入核函数的周期特性，如某周期函数为
Figure PCTCN2021070686-appb-000030
p为该核函数具有的周期，则由RBF核得到的周期核形式为
Figure PCTCN2021070686-appb-000031
可以写成如下形式：
Figure PCTCN2021070686-appb-000032
Figure PCTCN2021070686-appb-000033
K(x i,y j)是核函数，KELM的输出公式可以写成以下形式：
Figure PCTCN2021070686-appb-000034
C惩罚因子常量，利用公式中的惩罚因子C和核参数γ对学习机的泛化能力进行调优处理，并得出最佳道路交通拥堵指数Y′。
在本实施例中，该模型最大优点在求解时不必考虑隐含层节点数、初始权值和偏移量等未知量，直接利用内核函数的内积形式和核函数K(μ,ν)的特定形式便可计算出预测函数的值，能方便快捷的得到最佳道路交通拥堵指数Y′。
在本实施例中，通过最小加权平均算法和模糊判断推理判断将人的交通管理经验与机器快速精准运算能力相结合，计算出精确的道路交通融合数据；并利用这些数据提取道路特征运用核超限学习机群组算法预测出最佳道路交通拥堵指数Y′，充分发挥人机混合智能增强的优势，快速精准的预测道路拥堵指数为人机混合增强智能拥堵预警平台提供强有力的数据支撑。
所述步骤S5中，如图5所示，对道路拥堵预警判断过程为：
当车辆驶入某一段道路上，获取当前路段的多源交通数据；
通过对道路拥堵的影响因素进行特征提取，根据道路交通拥堵模型预测当前拥堵指数；
通过与最佳道路交通拥堵指数进行比较，当该路段交通流量达到预测拥堵指数上限时，发出道路拥堵预警信号；否则，车辆正常行驶。
在本实施例中，道路拥堵预警信号可由车载通信单元和路测网联设施同时向车载导航的云端发出，对路段行驶车辆的行驶速度进行检测，如果车辆到达该路段，调整路网信号配时方案，进行拥堵预警，并提供新的规划路线。
另外，可以理解的，交通管理部门可通过手机app实名认证的方式让驾驶员参与拥堵路段进行实地检测，当拥堵路段大部分驾驶员的拥堵感官与拥堵指标发生冲突时，以大部分驾驶员的感觉为准确保人的最高控制权，则云端清理该道路拥堵信号。
另外，可以理解的，云端收到拥堵信号后，可利用车载导航或手机app帮助驾驶员得到前方路段的交通信息，并且预警系统会根据路段拥堵的状况判定拥堵状态；同时检测车辆实时车速，预判多长时间之后车辆将行驶入拥堵路段并给出几条合理的规避拥堵的路线；同时拥堵路段前交叉口的LED大屏会给出路段的拥堵情况、车流量等信息，为准备驶入拥堵路段的车辆给予及时的拥堵预警。
交通管理中心通过对城市内每一条道路的拥堵情况的实时状况分析，引入交通管理者的经验智慧和专家的专业分析，建立新的信号优化模型将实时计算最优信号调优方案下发到每一个路口合理的调控各个路口的信号灯相位，尽可能的形成绿波带，建立区域联动的信号配时优化网络；同时各个路口具有一定的自我优化和调控的能力，当主干路与次干路上的拥堵车流发生交汇时，优先考虑主干路的绿波通行避免更大的拥堵状况的发生，形成具有自适应能力的分布式智能管控平台。本实施例通过互联网大脑调控各路口的交通信号相位和路侧预警设施实现了机器智能与人类智慧的协作，将机器智能与群体智能融入城市道路在拥堵预警中共同解决城市拥堵问题。
实施例2
本实施例提供一种道路交通拥堵预警系统，包括：
第一模糊权值计算模块，用于根据获取的多源交通数据进行特征分类，并构建相应的特征隶属度函数，对特征隶属度函数采用最小加权平均算法得到第一模糊权值；
第二模糊权值计算模块，用于对多源数据采用专家评价法构建人工隶属度函数，并计算第二模糊权值；
融合模块，用于根据第一模糊权值和第二模糊权值融合后得到的融合模糊权值，对特征隶属度函数进行模糊加权平均，对得到的不同特征量的加权平均值隶属度函数进行解模糊，得到多源融合交通数据；
模型构建模块，用于对多源融交通合数据采用核超限学习机群组算法构建道路交通拥堵模型，计算最佳道路交通拥堵指数；
拥堵预警模块，用于获取当前多源交通数据，根据道路交通拥堵模型预测当前拥堵指数，通过与最佳道路交通拥堵指数进行比较，对当前道路是否拥堵进行预警。
此处需要说明的是，上述各个模块对应于实施例1中的步骤S1至S5，上述模块与对应的步骤所实现的示例和应用场景相同，但不限于上述实施例1所公开的内容。需要说明的是，上述模块作为系统的一部分可以在诸如一组计算机可执行指令的计算机系统中执行。
实施例3
如图6所示，本实施例提供一种预警平台，包括人机混合增强智能多源数据采集子系统、人机混合增强智能多源数据融合子系统和人机混合增强智能拥堵预警子系统；
所述人机混合增强智能多源数据采集子系统由各类传感设备和交通参与者组成；
各类传感设备包括路网上铺设的固定式传感设备、车辆上安装的移动式传感设备，分别采集交通流量、车道数、车速、道路天气、交通事故发生状况等道路交通数据；采集路段上车辆的位置、车辆加速度、车头间距、驾驶员操作行为等车辆交通数据及驾驶员的行为特征数据。
交通参与者采集的数据主要有人类感知、管理经验和政策活动信息三大类；一是利用城市道路上的行人提供其运动轨迹、通过路口通行时间、多发事故路段、道路基础设施状况等道路基础信息；二是依靠行驶车辆的驾驶员提供驾驶员道路熟悉度和驾驶技术、精神状态、驾驶习惯、反应时间等驾驶员的感知信息；三是通过交通管理者提供交通拥堵状态判定规则、道路交通管理经验、路段实时交通信息、拥堵处理政策、城市交通管控方案和周边大型活动情况等管控政策经验信息。
人机混合增强多源数据采集子系统利用城市道路和车辆安装的传感设备和交通参与者建立一个人机混合的城市道路交通信息采集平台。该子系统通过手机APP将交通参与者提供的交通数据转化为App积分，这些积分可用来兑换加油卡或高速通行优惠券等小礼品，以此激励交通参与者为我们提供感知信息。同时利用道路上行驶的车辆、交通参与者、道路基础设施之间互联互通形成一个车-车、车-路、车-人、车-基础设施的局域数据采集平台，极大地提高不同数据采集来源的环境感知能力得到更为准确的道路交通数据。
所述人机混合增强智能多源数据融合子系统将数据采集子系统采集的交通数据进行简单的分类处理，之后分类后的数据分别传至智慧城市大数据中心对应的数据处理部门，进行多源异构数据的数据预处理。
可以理解的，人机混合增强智能多源数据融合子系统所实现的对多源交通数据构建隶属度函数、模糊权值的融合、道路交通拥堵模型的构建等过程与实施例1中所述的方法对应，在此不再赘述。
在道路拥堵预警阶段，利用多源数据融合子系统计算出的拥堵指标进行拥堵判断，当监测路段道路交通指标大于拥堵指标时，通过调整城市各路段的信号配时方案对驾驶员或行人进行交通诱导解决城市道路的拥堵问题；
所述人机混合增强智能拥堵预警子系统由道路预警设施、车辆预警设备组成。在当车辆驶入某一段道路上，多源信号数据融合模块提供路段交通参数，通过对道路拥堵的影响因素进行特征提取，当该路段交通流量达到预测拥堵指数上限时，车载通信单元和路侧网联设施同时向车载导航的云端发出道路拥堵信号；
同时交通管理部门可通过手机app实名认证的方式让驾驶员参与拥堵路段进行实地检测，当拥堵路段大部分驾驶员的拥堵感官与拥堵指标发生冲突时，以大部分驾驶员的感觉为准确保人的最高控制权，则云端清理该道路拥堵信号。
云端收到拥堵信号后，可利用车载导航或手机app帮助驾驶员得到前方路段的交通信息，并且预警系统会根据路段拥堵的状况判定拥堵状态；同时检测车辆实时车速，预判多长时间之后车辆将行驶入拥堵路段并给出几条合理的规避拥堵的路线；同时拥堵路段前交叉口的LED大屏会给出路段的拥堵情况、车流量等信息，为准备驶入拥堵路段的车辆给予及时的拥堵预警。
交通管理中心通过对城市内每一条道路的拥堵情况的实时状况分析，引入交通管理者的经验智慧和专家的专业分析，建立新的信号优化模型将实时计算最优信号调优方案下发到每一个路口合理的调控各个路口的信号灯相位，尽可能的形成绿波带，建立区域联动的信号配时优化网络；同时各个路口具有一定的自我优化和调控的能力，当主干路与次干路上的拥堵车流发生交汇时，优先考虑主干路的绿波通行避免更大的拥堵状况的发生，形成具有自适应能力的分布式智能管控平台。
在本实施中，该平台通过互联网大脑调控各路口的交通信号相位和路侧预警设施实现了机器智能与人类智慧的协作，将机器智能与群体智能融入城市道路在拥堵预警中共同解决城市拥堵问题；
该平台利用人类的群体智慧与机器智能相互协作，充分发挥了机器处理大量数据精准的超运算能力和道路交通参与者及管理者对道路拥堵的判断能力，形成了以人为主，人机结合的人机混合增强智能道路预警系统。
在更多实施例中，还提供：
一种电子设备，包括存储器和处理器以及存储在存储器上并在处理器上运行的计算机指令，所述计算机指令被处理器运行时，完成实施例1中所述的方法。为了简洁，在此不再赘述。
应理解，本实施例中，处理器可以是中央处理单元CPU，处理器还可以是其他通用处理 器、数字信号处理器DSP、专用集成电路ASIC，现成可编程门阵列FPGA或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。
存储器可以包括只读存储器和随机存取存储器，并向处理器提供指令和数据、存储器的一部分还可以包括非易失性随机存储器。例如，存储器还可以存储设备类型的信息。
一种计算机可读存储介质，用于存储计算机指令，所述计算机指令被处理器执行时，完成实施例1中所述的方法。
实施例1中的方法可以直接体现为硬件处理器执行完成，或者用处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器、闪存、只读存储器、可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。该存储介质位于存储器，处理器读取存储器中的信息，结合其硬件完成上述方法的步骤。为避免重复，这里不再详细描述。
本领域普通技术人员可以意识到，结合本实施例描述的各示例的单元即算法步骤，能够以电子硬件或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
以上仅为本公开的优选实施例而已，并不用于限制本公开，对于本领域的技术人员来说，本公开可以有各种更改和变化。凡在本公开的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本公开的保护范围之内。
上述虽然结合附图对本公开的具体实施方式进行了描述，但并非对本公开保护范围的限制，所属领域技术人员应该明白，在本公开的技术方案的基础上，本领域技术人员不需要付出创造性劳动即可做出的各种修改或变形仍在本公开的保护范围以内。

Claims (10)

  1. 一种道路交通拥堵预警方法，其特征在于，包括：
    根据获取的多源交通数据进行特征分类，并构建相应的特征隶属度函数，对特征隶属度函数采用最小加权平均算法得到第一模糊权值；
    对多源数据采用专家评价法构建人工隶属度函数，并计算第二模糊权值；
    根据第一模糊权值和第二模糊权值融合后得到的融合模糊权值，对特征隶属度函数进行模糊加权平均，对得到的不同特征量的加权平均值隶属度函数进行解模糊，得到多源融合交通数据；
    对多源融交通合数据采用核超限学习机群组算法构建道路交通拥堵模型，计算最佳道路交通拥堵指数；
    获取当前多源交通数据，根据道路交通拥堵模型预测当前拥堵指数，通过与最佳道路交通拥堵指数进行比较，对当前道路是否拥堵进行预警。
  2. 如权利要求1所述的一种道路交通拥堵预警方法，其特征在于，
    所述多源交通数据进行特征分类后得到路特征、人特征、环境特征和车特征；所述路特征包括交通流量、车道数和道路等级；所述人特征包括驾驶员行为特征和行人行为特征；所述环境特征包括道路天气和交通事故信息；所述车特征包括车辆的位置、车速、车头间距和车况。
  3. 如权利要求1所述的一种道路交通拥堵预警方法，其特征在于，
    根据特征分类结果分别构建特征论域，并对特征论域中特征数据划分等级，构建相应的模糊推理规则表，根据模糊推理规则表建立对应模糊推理等级论域的模糊子集，通过模糊映射得到特征隶属度函数。
  4. 如权利要求3所述的一种道路交通拥堵预警方法，其特征在于，
    对不同特征论域的数据进行加权平均，采用柯西不等式得到总均方差的极小值，利用多元函数极值计算总均方差极小值时的第一模糊权值。
  5. 如权利要求1所述的一种道路交通拥堵预警方法，其特征在于，
    所述解模糊采用重心法，得到的多源交通融合数据包括交通流量、反应时间、车速、车头间距、车辆加速度。
  6. 如权利要求1所述的一种道路交通拥堵预警方法，其特征在于，
    获取多源交通融合数据作为输入样本对核超限学习子模型进行训练，得到不同特征量的子模型；
    对不同特征量的子模型进行并行运算，构建道路交通拥堵模型。
  7. 如权利要求1所述的一种道路交通拥堵预警方法，其特征在于，
    根据核超限学习机群组算法的内核函数的内积形式和核函数计算最佳道路交通拥堵指数。
  8. 一种道路交通拥堵预警系统，其特征在于，包括：
    第一模糊权值计算模块，用于根据获取的多源交通数据进行特征分类，并构建相应的特征隶属度函数，对特征隶属度函数采用最小加权平均算法得到第一模糊权值；
    第二模糊权值计算模块，用于对多源数据采用专家评价法构建人工隶属度函数，并计算第二模糊权值；
    融合模块，用于根据第一模糊权值和第二模糊权值融合后得到的融合模糊权值，对特征隶属度函数进行模糊加权平均，对得到的不同特征量的加权平均值隶属度函数进行解模糊，得到多源融合交通数据；
    模型构建模块，用于对多源融交通合数据采用核超限学习机群组算法构建道路交通拥堵模型，计算最佳道路交通拥堵指数；
    拥堵预警模块，用于获取当前多源交通数据，根据道路交通拥堵模型预测当前拥堵指数，通过与最佳道路交通拥堵指数进行比较，对当前道路是否拥堵进行预警。
  9. 一种电子设备，其特征在于，包括存储器和处理器以及存储在存储器上并在处理器上运行的计算机指令，所述计算机指令被处理器运行时，完成权利要求1-7任一项所述的方法。
  10. 一种计算机可读存储介质，其特征在于，用于存储计算机指令，所述计算机指令被处理器执行时，完成权利要求1-7任一项所述的方法。
PCT/CN2021/070686 2020-04-30 2021-01-07 一种道路交通拥堵预警方法及系统 WO2021218235A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/609,408 US20220222556A1 (en) 2020-04-30 2021-01-07 Road traffic jam early warning method and system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010364099.9A CN111583639B (zh) 2020-04-30 2020-04-30 一种道路交通拥堵预警方法及系统
CN202010364099.9 2020-04-30

Publications (1)

Publication Number Publication Date
WO2021218235A1 true WO2021218235A1 (zh) 2021-11-04

Family

ID=72111895

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/070686 WO2021218235A1 (zh) 2020-04-30 2021-01-07 一种道路交通拥堵预警方法及系统

Country Status (3)

Country Link
US (1) US20220222556A1 (zh)
CN (1) CN111583639B (zh)
WO (1) WO2021218235A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115410375A (zh) * 2022-11-02 2022-11-29 华路易云科技有限公司 一种基于雷卡融合交通数据的融合交通指标集生成方法
CN117765738A (zh) * 2024-02-21 2024-03-26 云南省公路科学技术研究院 道路交通车流量分流控制优化方法、系统、终端及介质

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111583639B (zh) * 2020-04-30 2021-05-14 山东交通学院 一种道路交通拥堵预警方法及系统
CN112735124B (zh) 2020-12-16 2022-05-20 阿波罗智联(北京)科技有限公司 交通数据的分析方法、装置、设备、车辆及存储介质
CN113380033B (zh) * 2021-06-09 2022-08-16 山东交通学院 基于人机混合增强智能的城市交通安全预警方法及系统
CN113450569A (zh) * 2021-06-30 2021-09-28 阿波罗智联(北京)科技有限公司 确定路口状态的方法、装置、电子设备和存储介质
CN115641721B (zh) * 2022-12-06 2023-04-07 南京感动科技有限公司 多来源交通流指标融合及基于融合指标的路况计算方法
CN116434543B (zh) * 2023-03-27 2024-04-05 南京理工大学 一种基于事故分布规律和事故成因的交通隐患点预警方法
CN116050672B (zh) * 2023-03-31 2023-06-20 山东银河建筑科技有限公司 基于人工智能的城市管理方法及系统
CN116384635B (zh) * 2023-06-05 2023-09-01 中国标准化研究院 基于大数据的绿色生态城市信息化管理系统
CN116579905A (zh) * 2023-07-11 2023-08-11 新唐信通(北京)科技有限公司 一种基于生态通行指数改善交通的方法
CN116935654B (zh) * 2023-09-15 2023-12-01 北京安联通科技有限公司 一种基于数据分配值的智慧城市数据分析方法及系统
CN117334042B (zh) * 2023-09-28 2024-05-24 东莞市东莞通股份有限公司 一种基于人工智能的智慧交通管理系统及方法
CN117058888B (zh) * 2023-10-13 2023-12-22 华信纵横科技有限公司 一种交通大数据处理方法及其系统
CN117131581B (zh) * 2023-10-26 2024-02-13 乘木科技(珠海)有限公司 一种数字孪生城市道路的构建系统及方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160321919A1 (en) * 2015-04-30 2016-11-03 Here Global B.V. Method And System For Intelligent Traffic Jam Detection
CN106373390A (zh) * 2015-07-23 2017-02-01 中国国防科技信息中心 基于自适应神经模糊推理系统的道路交通状态评价方法
CN104778837B (zh) * 2015-04-14 2017-12-05 吉林大学 一种道路交通运行态势多时间尺度预测方法
CN108010321A (zh) * 2017-12-25 2018-05-08 北京理工大学 一种交通流预测方法
CN110085026A (zh) * 2019-03-28 2019-08-02 中国公路工程咨询集团有限公司 一种基于聚类分析和马尔科夫模型的交通状态预测方法
CN111583639A (zh) * 2020-04-30 2020-08-25 山东交通学院 一种道路交通拥堵预警方法及系统

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101216998B (zh) * 2008-01-11 2011-04-06 浙江工业大学 基于模糊粗糙集的证据理论城市交通流信息融合方法
JP5555778B2 (ja) * 2010-12-15 2014-07-23 本田技研工業株式会社 渋滞予測方法
CN102637357B (zh) * 2012-03-27 2013-11-06 山东大学 一种区域交通状态评价方法
CN104050809B (zh) * 2014-06-03 2016-07-20 北京航天福道高技术股份有限公司 交通状况自动分析预测方法
CN107248284B (zh) * 2017-08-09 2019-11-05 北方工业大学 基于多源信息融合的实时交通评价方法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104778837B (zh) * 2015-04-14 2017-12-05 吉林大学 一种道路交通运行态势多时间尺度预测方法
US20160321919A1 (en) * 2015-04-30 2016-11-03 Here Global B.V. Method And System For Intelligent Traffic Jam Detection
CN106373390A (zh) * 2015-07-23 2017-02-01 中国国防科技信息中心 基于自适应神经模糊推理系统的道路交通状态评价方法
CN108010321A (zh) * 2017-12-25 2018-05-08 北京理工大学 一种交通流预测方法
CN110085026A (zh) * 2019-03-28 2019-08-02 中国公路工程咨询集团有限公司 一种基于聚类分析和马尔科夫模型的交通状态预测方法
CN111583639A (zh) * 2020-04-30 2020-08-25 山东交通学院 一种道路交通拥堵预警方法及系统

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115410375A (zh) * 2022-11-02 2022-11-29 华路易云科技有限公司 一种基于雷卡融合交通数据的融合交通指标集生成方法
CN117765738A (zh) * 2024-02-21 2024-03-26 云南省公路科学技术研究院 道路交通车流量分流控制优化方法、系统、终端及介质
CN117765738B (zh) * 2024-02-21 2024-04-26 云南省公路科学技术研究院 道路交通车流量分流控制优化方法、系统、终端及介质

Also Published As

Publication number Publication date
CN111583639A (zh) 2020-08-25
US20220222556A1 (en) 2022-07-14
CN111583639B (zh) 2021-05-14

Similar Documents

Publication Publication Date Title
WO2021218235A1 (zh) 一种道路交通拥堵预警方法及系统
US11657708B2 (en) Large-scale real-time traffic flow prediction method based on fuzzy logic and deep LSTM
Gu et al. Utilizing UAV video data for in-depth analysis of drivers’ crash risk at interchange merging areas
Tsai et al. Optimizing hyperparameters of deep learning in predicting bus passengers based on simulated annealing
US20230124864A1 (en) Graph Representation Querying of Machine Learning Models for Traffic or Safety Rules
Goh et al. Bus accident analysis of routes with/without bus priority
Nasrollahi et al. Identification and prioritization of connected vehicle technologies for sustainable development in Iran
Quek et al. A novel self-organizing fuzzy rule-based system for modelling traffic flow behaviour
Kim et al. A comparative analysis of incident service time on urban freeways
Gamel et al. A fog-based Traffic Light Management Strategy (TLMS) based on fuzzy inference engine
CN115662113A (zh) 一种信号交叉口人车博弈冲突风险评估与预警方法
Farrag et al. STIMF: a smart traffic incident management framework
Hu et al. A roadside decision-making methodology based on deep reinforcement learning to simultaneously improve the safety and efficiency of merging zone
CN114267191A (zh) 缓解交通拥堵驾驶员控制系统、方法、介质、设备及应用
Alomari et al. Traffic speed prediction techniques in urban environments
Huang et al. A data-driven operational integrated driving behavioral model on highways
CN116645805A (zh) 一种车队路径匹配和可信度评估方法、系统、设备及终端
Gupta et al. Lstm based real-time smart parking system
CN114926823B (zh) 基于wgcn的车辆驾驶行为预测方法
Wang et al. Efficiency and Reliability Analysis of Self‐Adaptive Two‐Stage Fuzzy Control System in Complex Traffic Environment
CN112729329A (zh) 安全行驶路线推荐方法、装置以及设备
Shamlitskiy et al. Transport stream optimization based on neural network learning algorithms
Xie et al. A collaborative method on reversible lane clearance and signal coordination control in associated intersection
Liu Refined judgment of urban traffic state based on machine learning and edge computing
Lv et al. A Lane‐Changing Decision‐Making Model of Bus Entering considering Bus Priority Based on GRU Neural Network

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21796995

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21796995

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 15/03/2023)