CN117437774A - AI-based multi-mode traffic flow prediction system - Google Patents
AI-based multi-mode traffic flow prediction system Download PDFInfo
- Publication number
- CN117437774A CN117437774A CN202311280054.3A CN202311280054A CN117437774A CN 117437774 A CN117437774 A CN 117437774A CN 202311280054 A CN202311280054 A CN 202311280054A CN 117437774 A CN117437774 A CN 117437774A
- Authority
- CN
- China
- Prior art keywords
- data
- unit
- traffic
- module
- prediction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013480 data collection Methods 0.000 claims abstract description 60
- 238000004458 analytical method Methods 0.000 claims abstract description 28
- 238000000605 extraction Methods 0.000 claims abstract description 22
- 238000007405 data analysis Methods 0.000 claims abstract description 19
- 238000007781 pre-processing Methods 0.000 claims abstract description 17
- 230000004044 response Effects 0.000 claims abstract description 14
- 238000011217 control strategy Methods 0.000 claims abstract description 8
- 238000013500 data storage Methods 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 24
- 238000012549 training Methods 0.000 claims description 21
- 238000012544 monitoring process Methods 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 12
- 238000005457 optimization Methods 0.000 claims description 10
- 230000010354 integration Effects 0.000 claims description 9
- 238000004140 cleaning Methods 0.000 claims description 8
- 238000000556 factor analysis Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000012731 temporal analysis Methods 0.000 claims description 6
- 238000000700 time series analysis Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 4
- 238000010219 correlation analysis Methods 0.000 claims description 3
- 238000013479 data entry Methods 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 230000007613 environmental effect Effects 0.000 claims description 3
- 238000001556 precipitation Methods 0.000 claims description 3
- 230000001932 seasonal effect Effects 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000003912 environmental pollution Methods 0.000 abstract description 2
- 238000007726 management method Methods 0.000 description 7
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- 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
- G08G1/0137—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
- G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Analytical Chemistry (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Chemical & Material Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to the technical field of traffic flow prediction, in particular to an AI-based multi-mode traffic flow prediction system, which comprises a data collection module, a data preprocessing module, a data analysis and feature extraction module, an AI prediction model module, a real-time event response module, a prediction result analysis module and a traffic control module; wherein, the data collection module: the system comprises a data collection module, a data collection module and a data storage module, wherein the data collection module is used for collecting multi-mode traffic data, and the multi-mode traffic data comprises traffic flow, people flow, public traffic flow, traffic signal states and weather conditions; and a data preprocessing module: raw data from the data collection module is received, cleaned, normalized and integrated, and a data quality report is generated and passed to the real-time event response module. According to the invention, the prediction system based on the AI effectively improves urban traffic efficiency, enhances resident traveling experience and is beneficial to reducing environmental pollution by analyzing the multi-mode traffic data in real time and adaptively adjusting the control strategy.
Description
Technical Field
The invention relates to the technical field of traffic flow prediction, in particular to an AI-based multi-mode traffic flow prediction system.
Background
With the acceleration of the urban process, urban traffic flow management becomes an increasingly significant challenge, and the traditional traffic management method mainly depends on experience judgment and a static traffic model, and often cannot respond to sudden traffic events and daily traffic fluctuation in time, so that the problems of traffic jam, travel delay, environmental pollution and the like are increasingly aggravated, and great inconvenience is brought to urban residents.
In order to manage urban traffic more precisely and efficiently, researchers and engineers seek to adopt more advanced technical methods, wherein data-based methods, especially using big data and artificial intelligence techniques, offer new possibilities for traffic prediction and management, and by collecting multi-modal traffic data, such as traffic flow, people flow, bus flow, traffic signal status and weather conditions, more comprehensive information can be provided for traffic management, however, how to efficiently process, analyze and transform these data into practical traffic control strategies remains a key issue to be solved.
In addition, the quality and the integrity of data are critical to the accuracy of traffic prediction, and in practical application, due to sensor faults, data transmission interruption or other reasons, data loss or abnormality may occur, how to effectively identify and process the data problems, so that the accuracy and the robustness of a prediction model are ensured, and the method is a hot spot of current research.
In summary, an AI-based multi-modal traffic flow prediction system is developed, which can collect, process and analyze traffic data in real time, and automatically adjust traffic control strategies according to prediction results, thus being an urgent requirement and research trend of current urban traffic management.
Disclosure of Invention
Based on the above objects, the present invention provides an AI-based multi-modal traffic flow prediction system.
The multi-mode traffic flow prediction system based on the AI comprises a data collection module, a data preprocessing module, a data analysis and feature extraction module, an AI prediction model module, a real-time event response module, a prediction result analysis module and a traffic control module; wherein,
and a data collection module: the system comprises a data collection module, a data collection module and a data storage module, wherein the data collection module is used for collecting multi-mode traffic data, and the multi-mode traffic data comprises traffic flow, people flow, public traffic flow, traffic signal states and weather conditions;
and a data preprocessing module: the method comprises the steps of receiving original data from a data collection module, cleaning, normalizing and integrating the original data, generating a data quality report and transmitting the data quality report to a real-time event response module;
and the data analysis and feature extraction module is used for: receiving the processed data from the data preprocessing module, and performing data analysis to extract key features;
AI prediction model module: key features from the data analysis and feature extraction module are received, and traffic flow prediction is performed based on a pre-trained AI model;
a real-time event response module: receiving a data quality report and key features to identify factors which can influence the prediction accuracy in real time, thereby adjusting parameters of an AI prediction model module;
the prediction result analysis module: receiving a prediction result of the AI prediction model module, and analyzing and optimizing the result;
and a traffic control module: and receiving the prediction result of the prediction result analysis module, and adjusting the traffic control measures according to the prediction result.
Further, the data collection module comprises a traffic flow data collection unit, a people flow data collection unit, a public traffic flow data collection unit, a traffic signal state data collection unit and a weather condition data collection unit; wherein,
traffic flow data collection unit: equipped with the vehicle counter, is used for monitoring the vehicle flow of each crossing and main road in real time, the vehicle flow data that will collect is marked in the form of the time stamp at the same time;
a people flow data collection unit: the method comprises the steps of utilizing ground and overhead human body sensors or CCTV cameras to monitor the human flow of sidewalks and intersections in real time and correlating the data with time and place information;
bus flow data collection unit: collecting the position, speed, direction and passenger capacity of each bus through connection with an API (application program interface) of the bus system or a GPS (global positioning system) tracking system;
traffic signal status data collection unit: acquiring the state of a traffic light in real time by using a sensor or an interface with a traffic signal control system;
weather condition data collection unit: and the weather information of the local air temperature, humidity, wind speed and precipitation is collected in real time through the connection with a weather station or an online weather service interface.
Further, the data preprocessing module comprises a data cleaning unit, a data standardization unit, a data integration unit and a data quality report generation unit; wherein,
and a data cleaning unit: abnormal values, repeated values and irrelevant information in the data can be identified and deleted, and lost or incomplete data entries are processed at the same time, so that the accuracy of subsequent analysis is ensured;
data normalization unit: various data are converted into a unified standard format through a preset algorithm, and when the data of the traffic flow and the people flow are converted into the hourly flow, the consistency of the data in the subsequent analysis is ensured;
a data integration unit: the data processing system is used for automatically identifying and integrating data from different data collection units, combining the data into a structured data set, and particularly integrating data of traffic flow, people flow, public traffic flow, traffic signal states and weather conditions according to time and place;
a data quality report generation unit: for generating a data quality report based on the raw data and the processed data, the quality report listing details of the data cleansing, criteria for data normalization, and structural descriptions of data integration.
Further, the data analysis and feature extraction module specifically includes:
time series analysis unit: performing time series analysis on the received traffic flow, people flow and bus flow data to identify the periodicity, trending and potential seasonal influence of the traffic flow;
correlation analysis unit: analyzing the interrelationship among traffic signal state, traffic flow, people flow and public traffic flow;
an environmental factor analysis unit: analyzing the relation between the weather condition data and traffic flow, and specifically, analyzing the influence of rainy days or snowy days on traffic flow and people flow;
feature extraction unit: the feature extraction unit can also automatically select and optimize features according to analysis results so as to be used by an AI prediction model.
Further, the AI prediction model module comprises a model training unit, a model optimizing unit and a real-time prediction unit; wherein,
model training unit: the method comprises the steps of using historical traffic data as a training set, combining extracted key feature traffic peak time, main traffic bottleneck section and weather impact indexes on traffic, training a model, wherein the training process is based on a deep learning regression model, and the specific algorithm is as follows:
wherein, L (θ) is a loss function, representing the average squared difference of the predicted value and the true value; n is the number of training samples; y is i Is the true value of the i-th sample; x is x i Is the input feature of the ith sample; f is a predictive function associated with the model parameter θ;
model optimizing unit: optimizing model parameters by using a random gradient descent algorithm, wherein for random gradient descent, a specific updating rule is as follows:
wherein θ t For model parameters at time t, η is the learning rate,gradient at time t for the loss function L;
real-time prediction unit: and receiving key features from the data analysis and feature extraction module, and inputting the key features into the pre-trained AI model to obtain a real-time prediction result of traffic flow.
Further, the real-time event response module comprises a data quality monitoring unit, a characteristic deviation identifying unit, an influence factor analyzing unit and a parameter adjustment recommending unit; wherein,
data quality monitoring unit: the data quality report is used for receiving and reading the data quality report from the data preprocessing module, and monitoring is carried out by setting a threshold value that the missing data is not more than 2% and the abnormal data is not more than 1%, so that the data integrity and consistency are ensured;
feature deviation recognition unit: comparing the received key features with the historical average feature values, and triggering a deviation alarm when any key feature deviates by more than 10% from the historical average value;
influence factor analysis unit: combining the results of the data quality monitoring unit and the characteristic deviation identifying unit, and utilizing a preset logic rule, namely judging that the risk is high when the data is missing by more than 2% and the characteristic deviation is more than 10%;
parameter adjustment recommendation unit: when the risk is identified as high, the parameter adjustment recommendation unit automatically adjusts the model weight.
Further, the formula of automatically adjusting the model weight by the parameter adjustment recommendation unit is as follows:
wherein w is new Is a new weight, w old Is the original weight, alpha is a predefined learning rate, set to 0.01,is the gradient of the loss function J with respect to the weight.
Further, the prediction result analysis module specifically includes:
and a result analysis unit: the method comprises the steps of receiving a prediction result from an A1 prediction model module, carrying out structural display on the prediction result, and specifically, carrying out comprehensive analysis on the quality and accuracy of the prediction result by using clear index standards including average error rate, maximum error and error distribution;
prediction error calculation unit: calculating the error between the predicted and actual data, the specific formula is:
wherein y is i,pred Is the predicted value of the ith sample, y i,true Is the true value of the ith sample, N is the number of samples;
optimization suggestion unit: based on the result of the prediction error calculation unit, the optimization suggestion unit will generate a specific optimization suggestion including readjusting parameters of the AI model, increasing the amount of training data, or the feature factor when the error exceeds a preset threshold.
Further, the traffic control module specifically includes:
a result receiving unit: receiving a prediction result, wherein the prediction result comprises future traffic flow, people flow, public traffic flow and traffic signal states of each road section;
control strategy generation unit: calculating a traffic pressure value of each road section based on the received prediction result, wherein the specific formula is expressed as follows:
P i =α·V car,i +β·V ped,i +γ·V bus,i ,
wherein P is i Is the traffic pressure value of the ith road section, V car,i 、V ped,i And V bus,i The traffic flow, the people flow and the public traffic flow predicted value of the ith road section are respectively, and alpha, beta and gamma are weights of various flow types and are preset to be fixed values;
traffic signal adjustment unit: when P of a road section i When the preset upper limit value of 1000 units of pressure value is exceeded, the traffic signal adjusting unit can automatically adjust the traffic signal of the road section.
The invention has the beneficial effects that:
according to the invention, the multi-modal traffic flow prediction system based on the AI is introduced, so that the multi-modal traffic data of the city can be comprehensively and real-timely collected, the system ensures the quality and the integrity of the data by utilizing an advanced data processing and analyzing method, and compared with the traditional traffic management method, the system can more accurately capture subtle changes and potential trends of traffic flow when predicting the traffic flow, thereby greatly improving the accuracy and the instantaneity of the prediction.
According to the invention, the traffic control strategy can be adjusted in real time according to the prediction result, when the traffic pressure increase possibly occurs in a certain road section or area, the system can automatically adjust the traffic signal time length or take other control measures, so that the traffic is smooth, the possibility of traffic jam is reduced, and the self-adaptive control strategy ensures that urban traffic management is more intelligent and efficient.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only of the invention and that other drawings can be obtained from them without inventive effort for a person skilled in the art.
Fig. 1 is a schematic diagram of a traffic flow prediction system according to an embodiment of the present invention.
Detailed Description
The present invention will be further described in detail with reference to specific embodiments in order to make the objects, technical solutions and advantages of the present invention more apparent.
It is to be noted that unless otherwise defined, technical or scientific terms used herein should be taken in a general sense as understood by one of ordinary skill in the art to which the present invention belongs. The terms "first," "second," and the like, as used herein, do not denote any order, quantity, or importance, but rather are used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that elements or items preceding the word are included in the element or item listed after the word and equivalents thereof, but does not exclude other elements or items. The terms "connected" or "connected," and the like, are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", etc. are used merely to indicate relative positional relationships, which may also be changed when the absolute position of the object to be described is changed.
As shown in fig. 1, the AI-based multi-modal traffic flow prediction system includes a data collection module, a data preprocessing module, a data analysis and feature extraction module, an AI prediction model module, a real-time event response module, a prediction result analysis module, and a traffic control module; wherein,
and a data collection module: the system comprises a data collection module, a data collection module and a data storage module, wherein the data collection module is used for collecting multi-mode traffic data, and the multi-mode traffic data comprises traffic flow, people flow, public traffic flow, traffic signal states and weather conditions;
and a data preprocessing module: the method comprises the steps of receiving original data from a data collection module, cleaning, normalizing and integrating the original data, generating a data quality report and transmitting the data quality report to a real-time event response module;
and the data analysis and feature extraction module is used for: receiving the processed data from the data preprocessing module, and performing data analysis to extract key features;
AI prediction model module: key features from the data analysis and feature extraction module are received, and traffic flow prediction is performed based on a pre-trained AI model;
a real-time event response module: receiving a data quality report and key features to identify factors which can influence the prediction accuracy in real time, thereby adjusting parameters of an AI prediction model module;
the prediction result analysis module: receiving a prediction result of the AI prediction model module, and analyzing and optimizing the result;
and a traffic control module: and receiving the prediction result of the prediction result analysis module, and adjusting the traffic control measures according to the prediction result.
The data collection module comprises a traffic flow data collection unit, a people flow data collection unit, a bus flow data collection unit, a traffic signal state data collection unit and a weather condition data collection unit; wherein,
traffic flow data collection unit: equipped with the vehicle counter, is used for monitoring the vehicle flow of each crossing and main road in real time, the vehicle flow data that will collect is marked in the form of the time stamp at the same time;
a people flow data collection unit: the method comprises the steps of utilizing ground and overhead human body sensors or CCTV cameras to monitor the human flow of sidewalks and intersections in real time and correlating the data with time and place information;
bus flow data collection unit: collecting the position, speed, direction and passenger capacity of each bus through connection with an API (application program interface) of the bus system or a GPS (global positioning system) tracking system;
traffic signal status data collection unit: acquiring the state (the duration, the period and the like of the traffic light) of the traffic light in real time by using a sensor or an interface with a traffic signal control system;
weather condition data collection unit: and the weather information of the local air temperature, humidity, wind speed and precipitation is collected in real time through the connection with a weather station or an online weather service interface.
The data preprocessing module comprises a data cleaning unit, a data standardization unit, a data integration unit and a data quality report generation unit; wherein,
and a data cleaning unit: abnormal values, repeated values and irrelevant information in the data can be identified and deleted, and lost or incomplete data entries are processed at the same time, so that the accuracy of subsequent analysis is ensured;
data normalization unit: various data are converted into a unified standard format through a preset algorithm, and when the data of the traffic flow and the people flow are converted into the hourly flow, the consistency of the data in the subsequent analysis is ensured;
a data integration unit: the data processing system is used for automatically identifying and integrating data from different data collection units, combining the data into a structured data set, and particularly integrating data of traffic flow, people flow, public traffic flow, traffic signal states and weather conditions according to time and place;
a data quality report generation unit: for generating a data quality report based on the raw data and the processed data, the quality report listing details of the data cleansing, criteria for data normalization, and structural descriptions of data integration.
The data analysis and feature extraction module specifically comprises:
time series analysis unit: performing time series analysis on the received traffic flow, people flow and bus flow data to identify the periodicity, trending and potential seasonal influence of the traffic flow;
correlation analysis unit: analyzing the interrelationship among traffic signal state, traffic flow, people flow and public traffic flow;
an environmental factor analysis unit: analyzing the relation between the weather condition data and traffic flow, and specifically, analyzing the influence of rainy days or snowy days on traffic flow and people flow;
feature extraction unit: the feature extraction unit can also automatically select and optimize features according to analysis results so as to be used by an AI prediction model.
The AI prediction model module comprises a model training unit, a model optimizing unit and a real-time prediction unit; wherein,
model training unit: the method comprises the steps of using historical traffic data as a training set, combining extracted key feature traffic peak time, main traffic bottleneck section and weather impact indexes on traffic, training a model, wherein the training process is based on a deep learning regression model, and the specific algorithm is as follows:
wherein, L (θ) is a loss function, representing the average squared difference of the predicted value and the true value; n is the number of training samples; y is i Is the true value of the i-th sample; x is x i Is the input of the ith sampleFeatures; f is a predictive function associated with the model parameter θ;
model optimizing unit: optimizing model parameters by using a random gradient descent (SGD) algorithm, wherein for random gradient descent, a specific updating rule is as follows:
wherein θ t For model parameters at time t, η is the learning rate,gradient at time t for the loss function L;
real-time prediction unit: and receiving key features from the data analysis and feature extraction module, and inputting the key features into the pre-trained AI model to obtain a real-time prediction result of traffic flow.
The real-time event response module comprises a data quality monitoring unit, a characteristic deviation identifying unit, an influence factor analyzing unit and a parameter adjusting recommending unit; wherein,
data quality monitoring unit: the data quality report is used for receiving and reading the data quality report from the data preprocessing module, and monitoring is carried out by setting a threshold value that the missing data is not more than 2% and the abnormal data is not more than 1%, so that the data integrity and consistency are ensured;
feature deviation recognition unit: comparing the received key features with the historical average feature values, and triggering a deviation alarm when any key feature deviates by more than 10% from the historical average value;
influence factor analysis unit: combining the results of the data quality monitoring unit and the characteristic deviation identifying unit, and utilizing a preset logic rule, namely judging that the risk is high when the data is missing by more than 2% and the characteristic deviation is more than 10%;
parameter adjustment recommendation unit: when the risk is identified as high, the parameter adjustment recommendation unit automatically adjusts the model weight.
The formula of automatically adjusting the model weight by the parameter adjustment recommendation unit is as follows:
wherein w is new Is a new weight, w old Is the original weight, alpha is a predefined learning rate, set to 0.01,is the gradient of the loss function J with respect to the weight.
The prediction result analysis module specifically comprises:
and a result analysis unit: the method comprises the steps of receiving a prediction result from an A1 prediction model module, carrying out structural display on the prediction result, and specifically, carrying out comprehensive analysis on the quality and accuracy of the prediction result by using clear index standards including average error rate, maximum error and error distribution;
prediction error calculation unit: calculating the error between the predicted and actual data, the specific formula is:
wherein y is i,pred Is the predicted value of the ith sample, y i,true Is the true value of the ith sample, N is the number of samples;
optimization suggestion unit: based on the result of the prediction error calculation unit, the optimization suggestion unit will generate a specific optimization suggestion including readjusting the parameters of the AI model, increasing the amount of training data or the feature factors when the error exceeds a preset threshold.
The traffic control module specifically comprises:
a result receiving unit: receiving a prediction result, wherein the prediction result comprises future traffic flow, people flow, public traffic flow and traffic signal states of each road section;
control strategy generation unit: calculating a traffic pressure value of each road section based on the received prediction result, wherein the specific formula is expressed as follows:
P i =α·V car,i +β·V ped,i +γ·V bus,i ,
wherein P is i Is the traffic pressure value of the ith road section, V car,i 、V ped,i And V bus,i The traffic flow, the people flow and the public traffic flow predicted value of the ith road section are respectively, and alpha, beta and gamma are weights of various flow types and are preset to be fixed values;
traffic signal adjustment unit: when P of a road section i When the preset upper limit value of 1000 units of pressure value is exceeded, the traffic signal adjusting unit can automatically adjust the traffic signal of the road section, for example, the green light time is prolonged or the red light time is shortened, so that the flow can be ensured to pass smoothly.
The present invention is intended to embrace all such alternatives, modifications and variances which fall within the broad scope of the appended claims. Therefore, any omission, modification, equivalent replacement, improvement, etc. of the present invention should be included in the scope of the present invention.
Claims (9)
1. The multi-mode traffic flow prediction system based on the AI is characterized by comprising a data collection module, a data preprocessing module, a data analysis and feature extraction module, an AI prediction model module, a real-time event response module, a prediction result analysis module and a traffic control module; wherein,
and a data collection module: the system comprises a data collection module, a data collection module and a data storage module, wherein the data collection module is used for collecting multi-mode traffic data, and the multi-mode traffic data comprises traffic flow, people flow, public traffic flow, traffic signal states and weather conditions;
and a data preprocessing module: the method comprises the steps of receiving original data from a data collection module, cleaning, normalizing and integrating the original data, generating a data quality report and transmitting the data quality report to a real-time event response module;
and the data analysis and feature extraction module is used for: receiving the processed data from the data preprocessing module, and performing data analysis to extract key features;
AI prediction model module: key features from the data analysis and feature extraction module are received, and traffic flow prediction is performed based on a pre-trained AI model;
a real-time event response module: receiving a data quality report and key features to identify factors which can influence the prediction accuracy in real time, thereby adjusting parameters of an AI prediction model module;
the prediction result analysis module: receiving a prediction result of the AI prediction model module, and analyzing and optimizing the result;
and a traffic control module: and receiving the prediction result of the prediction result analysis module, and adjusting the traffic control measures according to the prediction result.
2. The AI-based multimodal traffic flow prediction system of claim 1 wherein the data collection module comprises a traffic flow data collection unit, a people flow data collection unit, a bus flow data collection unit, a traffic signal status data collection unit, and a weather condition data collection unit; wherein,
traffic flow data collection unit: equipped with the vehicle counter, is used for monitoring the vehicle flow of each crossing and main road in real time, the vehicle flow data that will collect is marked in the form of the time stamp at the same time;
a people flow data collection unit: the method comprises the steps of utilizing ground and overhead human body sensors or CCTV cameras to monitor the human flow of sidewalks and intersections in real time and correlating the data with time and place information;
bus flow data collection unit: collecting the position, speed, direction and passenger capacity of each bus through connection with an API (application program interface) of the bus system or a GPS (global positioning system) tracking system;
traffic signal status data collection unit: acquiring the state of a traffic light in real time by using a sensor or an interface with a traffic signal control system;
weather condition data collection unit: and the weather information of the local air temperature, humidity, wind speed and precipitation is collected in real time through the connection with a weather station or an online weather service interface.
3. The AI-based multimodal traffic flow prediction system of claim 2 wherein the data preprocessing module comprises a data cleansing unit, a data normalization unit, a data integration unit, and a data quality report generation unit; wherein,
and a data cleaning unit: abnormal values, repeated values and irrelevant information in the data can be identified and deleted, and lost or incomplete data entries are processed at the same time, so that the accuracy of subsequent analysis is ensured;
data normalization unit: various data are converted into a unified standard format through a preset algorithm, and when the data of the traffic flow and the people flow are converted into the hourly flow, the consistency of the data in the subsequent analysis is ensured;
a data integration unit: the data processing system is used for automatically identifying and integrating data from different data collection units, combining the data into a structured data set, and particularly integrating data of traffic flow, people flow, public traffic flow, traffic signal states and weather conditions according to time and place;
a data quality report generation unit: for generating a data quality report based on the raw data and the processed data, the quality report listing details of the data cleansing, criteria for data normalization, and structural descriptions of data integration.
4. The AI-based multimodal traffic flow prediction system of claim 3 wherein the data analysis and feature extraction module specifically comprises:
time series analysis unit: performing time series analysis on the received traffic flow, people flow and bus flow data to identify the periodicity, trending and potential seasonal influence of the traffic flow;
correlation analysis unit: analyzing the interrelationship among traffic signal state, traffic flow, people flow and public traffic flow;
an environmental factor analysis unit: analyzing the relation between the weather condition data and traffic flow, and specifically, analyzing the influence of rainy days or snowy days on traffic flow and people flow;
feature extraction unit: the feature extraction unit can also automatically select and optimize features according to analysis results so as to be used by an AI prediction model.
5. The AI-based multimodal traffic flow prediction system of claim 4 wherein the AI prediction model module comprises a model training unit, a model optimization unit, and a real-time prediction unit; wherein,
model training unit: the method comprises the steps of using historical traffic data as a training set, combining extracted key feature traffic peak time, main traffic bottleneck section and weather impact indexes on traffic, training a model, wherein the training process is based on a deep learning regression model, and the specific algorithm is as follows:
wherein, L (θ) is a loss function, representing the average squared difference of the predicted value and the true value; n is the number of training samples; y is i Is the true value of the i-th sample; x is x i Is the input feature of the ith sample; f is a predictive function associated with the model parameter θ;
model optimizing unit: optimizing model parameters by using a random gradient descent algorithm, wherein for random gradient descent, a specific updating rule is as follows:
wherein θ t For model parameters at time t, η is the learning rate,at time as a loss function LA gradient of t;
real-time prediction unit: and receiving key features from the data analysis and feature extraction module, and inputting the key features into the pre-trained AI model to obtain a real-time prediction result of traffic flow.
6. The AI-based multimodal traffic flow prediction system of claim 5 wherein the real-time event response module includes a data quality monitoring unit, a feature bias identification unit, an impact factor analysis unit, and a parameter adjustment recommendation unit; wherein,
data quality monitoring unit: the data quality report is used for receiving and reading the data quality report from the data preprocessing module, and monitoring is carried out by setting a threshold value that the missing data is not more than 2% and the abnormal data is not more than 1%, so that the data integrity and consistency are ensured;
feature deviation recognition unit: comparing the received key features with the historical average feature values, and triggering a deviation alarm when any key feature deviates by more than 10% from the historical average value;
influence factor analysis unit: combining the results of the data quality monitoring unit and the characteristic deviation identifying unit, and utilizing a preset logic rule, namely judging that the risk is high when the data is missing by more than 2% and the characteristic deviation is more than 10%;
parameter adjustment recommendation unit: when the risk is identified as high, the parameter adjustment recommendation unit automatically adjusts the model weight.
7. The AI-based multimodal traffic flow prediction system of claim 6 wherein the formula for automatically adjusting the model weights by the parameter adjustment recommendation unit is:
wherein w is new Is a new weight, w old Is the original weight, alpha is a predefined learning rate, set to 0.01,is the gradient of the loss function J with respect to the weight.
8. The AI-based multimodal traffic flow prediction system of claim 7 wherein the prediction outcome analysis module specifically comprises:
and a result analysis unit: the method comprises the steps of receiving a prediction result from an A1 prediction model module, carrying out structural display on the prediction result, and specifically, carrying out comprehensive analysis on the quality and accuracy of the prediction result by using clear index standards including average error rate, maximum error and error distribution;
prediction error calculation unit: calculating the error between the predicted and actual data, the specific formula is:
wherein y is i,pred Is the predicted value of the ith sample, y i,true Is the true value of the ith sample, N is the number of samples;
optimization suggestion unit: based on the result of the prediction error calculation unit, the optimization suggestion unit will generate a specific optimization suggestion including readjusting parameters of the AI model, increasing the amount of training data, or the feature factor when the error exceeds a preset threshold.
9. The AI-based multimodal traffic flow prediction system of claim 8 wherein the traffic control module specifically comprises:
a result receiving unit: receiving a prediction result, wherein the prediction result comprises future traffic flow, people flow, public traffic flow and traffic signal states of each road section;
control strategy generation unit: calculating a traffic pressure value of each road section based on the received prediction result, wherein the specific formula is expressed as follows:
P i =α·V car,i +β·V ped,i +γ·V bus,i ,
wherein P is i Is the traffic pressure value of the ith road section, V car,i 、V ped,i And V bus,i The traffic flow, the people flow and the public traffic flow predicted value of the ith road section are respectively, and alpha, beta and gamma are weights of various flow types and are preset to be fixed values;
traffic signal adjustment unit: when P of a road section i When the preset upper limit value of 1000 units of pressure value is exceeded, the traffic signal adjusting unit can automatically adjust the traffic signal of the road section.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311280054.3A CN117437774A (en) | 2023-10-07 | 2023-10-07 | AI-based multi-mode traffic flow prediction system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311280054.3A CN117437774A (en) | 2023-10-07 | 2023-10-07 | AI-based multi-mode traffic flow prediction system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN117437774A true CN117437774A (en) | 2024-01-23 |
Family
ID=89550693
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311280054.3A Pending CN117437774A (en) | 2023-10-07 | 2023-10-07 | AI-based multi-mode traffic flow prediction system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117437774A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117744950A (en) * | 2024-01-24 | 2024-03-22 | 深圳宇翊技术股份有限公司 | Travel demand analysis method, device, equipment and storage medium |
-
2023
- 2023-10-07 CN CN202311280054.3A patent/CN117437774A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117744950A (en) * | 2024-01-24 | 2024-03-22 | 深圳宇翊技术股份有限公司 | Travel demand analysis method, device, equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109754597B (en) | Urban road regional congestion regulation and control strategy recommendation system and method | |
CN108083044B (en) | Elevator on-demand maintenance system and method based on big data analysis | |
CN103258427B (en) | Urban expressway traffic real-time monitoring system and method based on information physical network | |
CN109785618B (en) | Short-term traffic flow prediction method based on combinational logic | |
CN104778837A (en) | Multi-time scale forecasting method for road traffic running situation | |
CN117437774A (en) | AI-based multi-mode traffic flow prediction system | |
CN112016735B (en) | Patrol route planning method and system based on traffic violation hotspot prediction and readable storage medium | |
CN112874589A (en) | Automatic train planned operation diagram adjusting system and method | |
CN113159374B (en) | Data-driven urban traffic flow rate mode identification and real-time prediction early warning method | |
CN111640304B (en) | Automatic quantitative extraction method for traffic jam propagation characteristics of continuous flow traffic facility | |
CN114267173B (en) | Multisource data fusion method, device and equipment for space-time characteristics of expressway | |
CN116631186B (en) | Expressway traffic accident risk assessment method and system based on dangerous driving event data | |
CN112767684A (en) | Highway traffic jam detection method based on charging data | |
CN116822769A (en) | Intelligent traffic route optimizing system based on artificial intelligence | |
CN116168356B (en) | Vehicle damage judging method based on computer vision | |
CN113361825A (en) | Early warning method and system for trampling accident | |
CN106910334B (en) | Method and device for predicting road section conditions based on big data | |
CN114549075A (en) | Regional parking evaluation method based on parking and video big data | |
CN107730882B (en) | Road congestion prediction system and method based on artificial intelligence | |
CN114266490B (en) | Efficient and accurate comprehensive transportation network security risk point identification method | |
CN116050581A (en) | Smart city subway driving scheduling optimization method and Internet of things system | |
CN113706860B (en) | Intelligent timing traffic light control method based on raspberry group | |
CN110796301B (en) | Passenger flow prediction method and device based on IC card data | |
CN114219970A (en) | Image processing method and system for traffic management | |
CN112418492A (en) | Passenger flow data acquisition and analysis system based on artificial intelligence |
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 |