A data acquisition system and method based on urban mobile networks

Abstract

The present invention discloses a data collection system and method based on an urban flow network, including: a real-time and prediction subsystem for buses, a temporal distribution subsystem for taxis, and a monitoring subsystem for pedestrians and non-motorized vehicles on roads of an urban flow network system; the data obtained by the real-time and prediction subsystem for buses, the temporal distribution subsystem for taxis, and the monitoring subsystem for pedestrians and non-motorized vehicles on roads are fitted in space and time to establish a temporal spatial distribution map of the city with fine granularity. The data of the three are fitted in space and time to establish a temporal spatial distribution map of the city with fine granularity, thereby constructing an urban flow network system to provide relevant early warnings for public conflict time.

Description

Technical Field

[0001] The present disclosure relates to the field of computer processing technology, and in particular to a data collection system and method based on an urban mobile network. Background Art

[0002] In the field of urban transportation in China, the conflict between traditional taxis and special cars that have emerged based on the Internet is intensifying. Emergency incidents in the taxi industry have become a thorny problem that plagues local government management departments. The emergency warning mechanism for urban transportation and related public events has a lag problem. A certain city's public passenger transport management service center has also become the first case in the country in the Chinese taxi industry to be sued by a special car operator. The key is that it is very likely to induce greater conflicts in the whole society. Simply relying on traditional experience-based management, simple fines and post-emergency plans can no longer effectively and promptly deal with related mass emergencies. Therefore, if we can build an early warning information model that triggers such emergencies, find the relationship between key factors and emergencies, and if we can dig out potential emergencies from the large amount of time series data generated by travel, it will undoubtedly have a high theoretical value and practical significance for conflict prevention. Summary of the invention

[0003] In order to solve the deficiencies of the prior art, the present disclosure provides a data collection system based on an urban mobility network to realize the collection and early warning of urban mobility data.

[0004] In order to achieve the above objectives, this application adopts the following technical solutions:

[0005] A data collection system based on an urban mobile network, comprising: a server, the server being configured to include a bus real-time and prediction subsystem, a taxi temporal distribution subsystem, and a road pedestrian and non-motor vehicle monitoring subsystem, and a network for collecting and transmitting data from the above subsystems constitutes an urban mobile network;

[0006] The bus real-time and prediction subsystem collects and processes the location information and card swiping information of the bus, and uses a neural network for training to obtain bus arrival prediction data;

[0007] The taxi temporal distribution subsystem is used to process the real-time location information and speed of taxis and generate a regional distribution map;

[0008] The non-motor vehicle monitoring subsystem is used to analyze the pedestrians and non-motor vehicles monitored at the urban road intersections to obtain the pedestrian and non-motor vehicle flow diagram at the urban road intersections;

[0009] The urban mobility network system fits the data obtained by the bus real-time and prediction subsystem, the taxi temporal distribution subsystem, and the road pedestrian and non-motor vehicle monitoring subsystem in space and time, and establishes a temporal spatial distribution map of the city in a fine-grained manner.

[0010] According to a further technical solution, the bus real-time and prediction subsystem includes a bus server, and the bus server includes:

[0011] The location information collection module is used to collect and monitor the longitude and latitude location information data of the bus, and the card swiping information collection module is used to collect and monitor the card swiping information data when boarding the bus;

[0012] The neural network prediction module trains the data collected by the location information collection module and the card swiping information collection module through the LSTM neural network, and then performs multi-station, multi-line, and multi-variable arrival predictions for buses running during working hours to obtain accurate arrival times.

[0013] According to a further technical solution, the taxi temporal distribution subsystem includes a taxi server;

[0014] The intelligent devices of the taxi Internet of Vehicles include an adapter box, a meter, a service evaluator, and a reversing camera connected to the on-board terminal. The on-board terminal is connected to the smart ceiling light screen and the dispatch information screen through the adapter box. The on-board terminal transmits the collected data of the smart ceiling light screen, the dispatch information screen, the meter, the service evaluator, and the reversing camera to the taxi server. At the same time, the built-in GPS positioning device and speed measuring device of the on-board terminal send the real-time location information and speed data of the taxi to the taxi server. The taxi server performs relevant calculations through the road traffic temporal analysis module to generate a regional distribution map.

[0015] In a further technical solution, the road traffic temporal analysis module is used for traffic situation analysis based on big data, including forming road network elements, building a real-time traffic network, and collecting and publishing road conditions;

[0016] Realize travel area division: divide the city into several areas according to the density of urban citizens' travel and the distribution of communities, and grid the city areas;

[0017] Realize the analysis of travel patterns in hot spots: calculate the distribution map of vehicle dispatch areas within a period of time;

[0018] The regional distribution maps include: departure area distribution map, destination area distribution map, hot departure area OD map, hot destination area OD map and travel OD trend statistics map.

[0019] In a further technical solution, the non-motor vehicle monitoring subsystem includes a non-motor vehicle server;

[0020] The real-time monitoring modules at the intersections of the city roads send the monitored data to the non-motor vehicle server through the data sending module;

[0021] The target detection and recognition module detects pedestrians and non-motorized vehicles on the road in real-time monitoring through the YOLO neural network, analyzes the targets, and generates pedestrian and non-motorized vehicle flow maps at urban intersections through the urban intersection flow map module, and transmits the detected target information to the non-motorized vehicle server.

[0022] The present application also discloses a data collection method based on an urban mobile network, comprising:

[0023] The bus real-time and prediction subsystem, taxi temporal distribution subsystem, and road pedestrian and non-motorized vehicle monitoring subsystem form an urban mobility network;

[0024] The bus real-time and prediction subsystem is used to collect and process the location information and card swiping information of the bus, and the neural network is used for training to obtain the bus arrival prediction data;

[0025] The taxi temporal distribution subsystem is used to process the real-time location information and speed of taxis and generate a regional distribution map;

[0026] The non-motorized vehicle monitoring subsystem is used to analyze the pedestrians and non-motorized vehicles on the roads monitored at the urban road intersections to obtain the pedestrian and non-motorized vehicle flow diagram at the urban road intersections;

[0027] The data obtained from the bus real-time and prediction subsystem, the taxi temporal distribution subsystem, and the road pedestrian and non-motorized vehicle monitoring subsystem are combined and modeled to construct a dynamic model of the urban flow network. Each subsystem is superimposed with corresponding environmental impact factor indicators, and the data of each subsystem are fitted in space and time. The temporal spatial distribution map of the city is established in a fine-grained manner, so as to realize conflict warning based on the urban flow network.

[0028] A further technical solution is to combine the data obtained from the bus real-time and prediction subsystem, the taxi temporal distribution subsystem, and the road pedestrian and non-motorized vehicle monitoring subsystem for modeling: the data from the bus real-time and prediction subsystem and the data from the taxi temporal distribution subsystem are imported into the map model in parallel. The map model is Amap or Baidu SDK, which reflects the vehicle driving status of the road. The road pedestrian and non-motorized vehicle data from the road pedestrian and non-motorized vehicle monitoring subsystem are imported into the map model.

[0029] A further technical solution is to build a dynamic model of the urban flow network: to use time series to visualize the dynamic model of the spatiotemporal urban flow network, that is, to use the fitting sequence constructed by environmental influencing factors and time series, and to visualize it using kepler.gl technology and open source geospatial toolkit to establish the temporal spatial distribution map of the city in a fine-grained manner, so that buses, taxis, pedestrians at intersections and non-motorized vehicles can be dynamically displayed and moved on the map like particles.

[0030] In a further technical solution, the environmental influencing factor indicators are mainly the time series generated by the temperature, whether there is rain or snow, the wind speed and the time corresponding to the urban mobile network system;

[0031] Each subsystem is superimposed with corresponding environmental impact factor indicators. Specifically, a nonlinear regression fitting is performed using the three influencing factors in the environmental impact factor indicators and the time period to match the data of each day.

[0032] The present disclosure also provides a computer-readable storage medium, in which a plurality of instructions are stored, and the instructions are suitable for being loaded and executed by a processor of a terminal device. The method for public conflict warning based on an urban mobile network is described.

[0033] The present disclosure also provides a terminal device, including a processor and a computer-readable storage medium, the processor is used to implement various instructions; the computer-readable storage medium is used to store multiple instructions, and the instructions are suitable for being loaded and executed by the processor to describe a public conflict warning method based on an urban mobile network.

[0034] Compared with the prior art, the beneficial effects of the present invention are:

[0035] Among the three subsystems disclosed in this disclosure, public transportation reflects the most stable vehicle operation mode in the city and is also the "aorta" of urban transportation; the taxi system is the dynamic tracking of urban vehicles, which can reflect the relevant vehicle dynamics of the entire city, including regional distribution, average speed, hot spots, etc.; road pedestrians and non-motor vehicles reflect the relevant traffic flow at the intersection. The three subsystems are integrated with each other, with complementary functions and complement each other. The data of the three are fitted in space and time, and the temporal spatial distribution map of the city is established with fine granularity, and then the urban flow network system is constructed to provide relevant early warnings for public conflict time. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings in the specification, which constitute a part of the present application, are used to provide further understanding of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute improper limitations on the present application.

[0037] Figure 1 It is a system schematic diagram of an implementation example of the present application;

[0038] Figure 2 is an architecture diagram of a bus real-time and prediction subsystem according to an embodiment of the present application;

[0039] Figure 3 It is an architecture diagram of the taxi temporal distribution subsystem of the embodiment of the present application;

[0040] Figure 4 is an architectural diagram of the road pedestrian and non-motor vehicle subsystem of the embodiment of the present application;

[0041] Figure 5 It is a dynamic distribution diagram of the urban flow network of the implementation example of the present application. DETAILED DESCRIPTION

[0042] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0043] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0044] In a typical implementation of the present application, Figure 1 As shown, a data collection system based on an urban mobile network is provided, including a bus real-time and prediction subsystem 4, a taxi temporal distribution subsystem 5, and a road pedestrian and non-motor vehicle monitoring subsystem 6, all of which are connected to the urban mobile network system.

[0045] The urban mobility network is composed of three systems: the real-time and prediction subsystem of buses, the temporal distribution subsystem of taxis, and the pedestrian and non-motorized vehicle subsystem. They are geographically embedded and complementary. Complex road conditions can be restored to a certain extent and modeled. The three subsystems are embedded and complementary. Buses provide urban passenger flow, taxis can reflect real-time private car traffic information, and pedestrians and non-motorized vehicles on the road can reflect the situation at the intersection. The three subsystems form a network of urban mobility networks, complement each other, and improve the mobility of urban roads.

[0046] The bus real-time and prediction subsystem 4 includes a location information collection module 7, a card swiping information collection module 8, a neural network prediction module 9 and an arrival information collection module. The location information collection module is used to collect and monitor the longitude and latitude location information data of the bus, and the card swiping information collection module is used to collect and monitor the card swiping information data when boarding the bus. The arrival information collection module is used to collect the time information of the vehicle arriving at the station to facilitate the real-time arrival prediction.

[0047] like Figure 2 As shown in , the GPS positioning device 9-1 in the location information collection module collects longitude and latitude information in real time when the bus is running and sends the data information to the bus server through the data sending device 9-2 for data processing, mainly including structural normalization, data missing value interpolation, etc. When the bus arrives at the station, the data sending device 9-2 sends the arrival time, departure time and other information to the bus server. When the bus arrives at the station, passengers can get on the bus by swiping the RFID card 8-1 or scanning the QR code 8-2, and this information is also sent to the bus server through the data sending device 9-2 for data processing.

[0048] The neural network prediction module will train all processed structured historical data through the LSTM neural network 7-1, and then make multi-station, multi-line, and multi-variable arrival predictions for buses running during working hours. In this way, accurate arrival times can be obtained, which firstly alleviates the problem of accurate time values ​​for traffic flow. At the same time, the dispatch of buses and the predicted arrival values ​​play an auxiliary role in the early warning of public conflicts.

[0049] This makes the arrival time of buses accurate to the minute. Once a bus has an abnormal signal or fails to arrive at the station on time, exceeding its corresponding delay threshold time point, it can be considered that a public conflict or emergency has occurred.

[0050] Through the real-time and predictive subsystem of buses, bus routes can also be reversed to ease traffic pressure and reduce the occurrence of public conflicts by easing passenger flow.

[0051] Specifically, the bus location information collection module is a GPS real-time positioning device. The card swiping information collection module is a QR code and RFID card swiping integrated module of the HMXF-A2-CZ model, which has two power input terminals and two digital signal output terminals. It sends data to the bus company data settlement center through a wireless base station for related information processing. The arrival information collection module is a wireless 433 base station, which determines whether the bus has arrived at the station by collecting the distance information of the bus. The information collection module has two power input terminals, four digital signal input terminals, and one digital signal output terminal.

[0052] The taxi temporal distribution subsystem 5 includes a taxi Internet of Vehicles intelligent device 10 and a traffic spatiotemporal situation analysis module 11 based on big data.

[0053] The taxi Internet of Vehicles intelligent device is used to collect information such as the location and operating status of taxi vehicles; the road traffic spatiotemporal situation analysis module mainly conducts relevant analysis on the distribution situation of taxis in the city.

[0054] The intelligent devices of taxi Internet of Vehicles include TGR20 vehicle-mounted terminal and YGL60 adapter box, meter, advertising screen, smart ceiling light screen, dispatch information screen, service evaluator, reversing camera and other devices installed on taxis.

[0055] Figure 3 As shown in , taxis are used to reflect the basic situation of the city's traffic flow. When a taxi receives an order on the road, the TGR20 vehicle terminal 11-1 will transmit the information of the smart ceiling light screen 11-3, the dispatch information screen 11-4, the meter 11-5, the service evaluator 11-6, the reversing camera 11-7 and other devices to the taxi server. The YGL60 adapter box connects the smart ceiling light screen 11-3 and the dispatch information screen 11-4 and aggregates and transfers the data to the TGR20 vehicle terminal. The built-in GPS positioning device and speed measuring device of the vehicle terminal 11-1 will send the real-time location information and speed of the taxi and other related data to the taxi temporal distribution subsystem 5. After all the data are analyzed and processed, the road traffic temporal analysis module 11 performs relevant calculations to generate a regional distribution map 10-1, which will reflect the regional division of taxis, as well as the travel patterns and travel routes of Jinan citizens.

[0056] The taxi server is composed of communication gateway, communication center, database, WEB application server, map server, monitoring and dispatching seats and network equipment (routers, firewalls, switches) and other equipment.

[0057] The taxi data is collected by the intelligent on-board terminal, which is also an integration of hardware equipment, including the intelligent ceiling light screen, TGR20 on-board terminal, GPS positioning equipment, speed measurement equipment, etc.

[0058] Road traffic temporal analysis module: The specific functions implemented are:

[0059] Traffic situation analysis based on big data: including the establishment of road network elements and the construction of real-time traffic network;

[0060] Traffic condition collection and release, etc.

[0061] Travel area division: According to the density of urban citizens' travel and the distribution of communities, the city is divided into 607 areas, and the Jinan area is gridded.

[0062] Analysis of travel patterns in hot areas: Calculate the distribution of vehicle departures in a certain period of time, with the aim of determining which areas have a large number of departures and which areas have a small number of departures.

[0063] Regional distribution map: Contains five distribution maps:

[0064] Distribution map of departure areas: Study the distribution map of the number of departure areas at each time point. The distribution map of departure areas can fully show which area has more departures and which area has fewer departures. An area with more departures means that there is more demand for vehicles in that area, and an area with fewer departures means that there is less demand for vehicles in that area.

[0065] Destination area distribution map: Study the distribution map of the number of arrivals at regional destinations at each time point. The destination area distribution map can fully show which area has more people arriving and which area has fewer people arriving. By comparing the destination area distribution maps of different time periods, we can study the areas where people flow to and from, analyze the flow direction and concentration of people flow, and provide data support for the setting of travel stations.

[0066] Hot departure area OD map (abbreviation of Origin and Destination): Count the areas with the largest number of departures, select the top 20, and then draw an OD map according to the destination. By studying the destination distribution of hot departure areas, we can study where the people who leave from hot departure areas arrive at. According to the distribution at different time points, we can study in detail the travel destination characteristics of people from hot departure areas.

[0067] Hot destination area OD map: Count the areas with the largest number of destinations, select the top 20, and then draw an OD map based on the departure point. By studying the distribution of departure points in hot destination areas, we can study the characteristics of the travel destinations of people from hot departure areas.

[0068] Travel OD trend statistics chart: OD chart that counts the number of taxis from one area to another. By analyzing the distribution of the number of trips from one area to another in different time periods, you can fully understand the OD flow trend of travel.

[0069] Specifically, the spatiotemporal situation analysis of road traffic is to establish a city road network and a real-time traffic network on GIS, and then analyze the data information of taxis through the information collection of taxi Internet of Vehicles intelligent devices, including the regional division of taxis, to calculate the regional distribution of taxi travel; travel pattern analysis is to analyze and study the travel patterns and travel routes of Jinan citizens. In view of the fact that sudden incidents in the taxi industry have become a thorny problem that troubles local government management departments, the urban mobility network system disclosed in this disclosure can monitor and warn related public conflicts through the taxi temporal distribution subsystem, and give early warnings to factors that induce social contradictions and conflicts, and respond quickly.

[0070] The road pedestrian and non-motor vehicle monitoring subsystem 6 includes a real-time monitoring device 12 at the road intersection and a target detection and recognition module 13. The real-time monitoring device at the road intersection is used to obtain information about pedestrians and non-motor vehicles at the intersection; the target detection and recognition module is used to detect real-time targets such as pedestrians and non-motor vehicles at the intersection and calculate the intersection flow.

[0071] like Figure 4 As shown in the figure, the real-time monitoring module 12 of the intersections is distributed at the intersections of the city roads, wherein the high-definition camera 12-1 sends the monitored data to the non-motor vehicle server through the data sending module 12-2, and the target detection and recognition module 13 detects the pedestrians and non-motor vehicles on the road in real-time monitoring through the YOLO neural network 13-1, and analyzes the target. The detected target information is used by the city intersection flow map module 13-2 to generate the pedestrian and non-motor vehicle flow map of the city intersection.

[0072] The hardware of the target detection and recognition module may specifically be a camera, which transmits data to a server of the non-motor vehicle and uses the server to perform target detection and recognition.

[0073] The urban intersection traffic map module is a software model generated by a server based on non-motor vehicles. The urban intersection traffic map refers to the detection of the number of pedestrians and non-motor vehicles at urban intersections after the server performs target detection through camera detection.

[0074] The intersection real-time detection module is used to obtain information about pedestrians and non-motor vehicles at the intersection; the target detection and recognition module is used to detect real-time targets such as pedestrians and non-motor vehicles at the intersection and calculate the intersection flow. The intersection real-time monitoring module is a 5 million POE starlight-level high-definition camera and its related accessories. The pedestrian and non-motor vehicle analysis module uses an end-to-end target detection algorithm based on YOLO3 to detect the number of pedestrian and non-motor vehicle targets, thereby detecting the flow information at the intersection.

[0075] The road traffic spatiotemporal situation analysis module is a data analysis platform built for road mobile vehicles, including road network element analysis, real-time traffic construction, hot spot area map and other related sub-modules. The road intersection real-time monitoring module is a Hikvision 5 million POE starlight-level high-definition camera and its related accessories. The pedestrian and non-motor vehicle analysis module is an end-to-end target detection algorithm based on YOLO3, which detects the number of pedestrians and non-motor vehicle targets.

[0076] When the bus real-time and prediction subsystem 4 detects abnormal information, the taxi spatiotemporal distribution subsystem 5 detects abnormal information, and the road pedestrian and non-motorized vehicle subsystem 6 detects abnormal information, the overall urban mobility network system 1 platform is used to analyze it, and the environmental impact factor indicators 2 are superimposed. Through the public conflict warning execution mechanism 3, according to the prescribed procedures, the relevant departments can take relevant measures on the spot in a timely manner after receiving the public conflict warning signal.

[0077] The environmental influencing factor indicators are mainly the temperature, whether there is rainfall (snowfall), wind speed and the time series generated by the time in the urban flow network system.

[0078] Use these three influencing factors and time periods to perform a nonlinear regression fit and match the data for each day.

[0079] The specific implementation example of the present disclosure adopts a combined modeling of public transportation, taxis, road pedestrians and non-motorized vehicles to construct a dynamic model of the urban flow network, which includes a bus real-time and prediction subsystem, a taxi temporal distribution subsystem and a road pedestrian and non-motorized vehicle monitoring subsystem, and then superimposes corresponding environmental influencing factor indicators to achieve conflict warning based on the urban flow network. Figure 5 It is a dynamic distribution diagram of the urban flow network of the implementation example of the present application.

[0080] Combine public transportation, taxis, road pedestrians and non-motorized vehicles into a model: public transportation and taxi data are imported into the map model (Amap or Baidu SDK) in parallel to reflect the vehicle driving status on the road. Road pedestrian and non-motorized vehicle data are imported into the map model to reflect the pedestrian and non-motorized vehicle data at the intersection.

[0081] About dynamic model: Use time series to visualize the dynamic model of spatiotemporal urban mobility network. That is, the fitting sequence constructed by environmental influencing factors and time series is used. The dynamic model of urban mobility network, just like in the video, buses, taxis, pedestrians at intersections and non-motorized vehicles will be displayed and moved dynamically on the map like particles. The visualization uses kepler.gl technology, an open source geospatial toolkit.

[0082] Among the three subsystems, public transportation reflects the most stable vehicle operation mode in the city and is also the "aorta" of urban transportation; the taxi system is the dynamic tracking of urban vehicles, which can reflect the relevant vehicle dynamics of the entire city, including regional distribution, average speed, hot spots, etc.; road pedestrians and non-motor vehicles reflect the relevant traffic flow at intersections. The three subsystems are integrated with each other, complementary in function, and complement each other. The data of the three are fitted in space and time, and the temporal spatial distribution map of the city is established in fine granularity, and then the urban flow network system is constructed to provide relevant early warnings for public conflict time.

[0083] Public transportation reflects the most stable vehicle operation mode in the city, while the taxi system is a dynamic tracking of urban vehicles, which can reflect the vehicle dynamics in the city. The status of pedestrians on the road can be obtained through the monitoring target pedestrian detection at the intersection. The three complement each other and complement each other. The data of the three are fitted in space and time, and the temporal spatial distribution map of the city is established in fine granularity, and the urban flow network system is constructed to provide relevant early warnings for public conflict time.

[0084] Through the fitting data series of time series and environmental influencing factors, the distribution status of the above three (buses, taxis, pedestrians at intersections and non-motor vehicles) is displayed on the three-dimensional map, that is, the urban flow network system is constructed.

[0085] In the technical solution disclosed in the present invention, when the bus real-time and prediction subsystem detects abnormal information, the taxi spatiotemporal distribution subsystem detects abnormal information, and the road pedestrian and non-motor vehicle monitoring subsystem detects abnormal information, the three subsystems will model and analyze the abnormal information through the overall urban flow network system to warn of public conflicts, so that relevant departments can take measures on the spot in time after receiving the conflict warning signal to prevent the people and public property from being harmed and lost.

[0086] The embodiments of the present disclosure also provide a computer-readable storage medium storing a plurality of instructions, wherein the instructions are suitable for being loaded and executed by a processor of a terminal device, for example, a method for public conflict warning based on an urban mobile network.

[0087] An embodiment of the present disclosure also provides a terminal device, including a processor and a computer-readable storage medium, the processor is used to implement various instructions; the computer-readable storage medium is used to store multiple instructions, and the instructions are suitable for being loaded and executed by the processor to describe a public conflict warning method based on an urban mobile network.

[0088] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A data collection system based on urban mobile network, characterized in that: include: The server is configured to include a bus real-time and prediction subsystem, a taxi temporal distribution subsystem, and a road pedestrian and non-motorized vehicle monitoring subsystem. The network that collects the data of the above subsystems for transmission constitutes an urban flow network, superimposes environmental impact factor indicators, and through the public conflict warning execution mechanism, according to the prescribed process, the relevant departments can take relevant measures on the spot in time after receiving the public conflict warning signal; The bus real-time and prediction subsystem collects and processes the location information and card swiping information of the bus, and uses a neural network for training to obtain bus arrival prediction data; The taxi temporal distribution subsystem is used to process the real-time location information and speed of taxis and generate a regional distribution map; The non-motor vehicle monitoring subsystem is used to analyze the pedestrians and non-motor vehicles monitored at the urban road intersections to obtain the pedestrian and non-motor vehicle flow diagram at the urban road intersections; The urban mobility network system fits the data obtained by the bus real-time and prediction subsystem, the taxi temporal distribution subsystem, and the road pedestrian and non-motor vehicle monitoring subsystem in space and time to establish a temporal spatial distribution map of the city in fine granularity; The bus real-time and prediction subsystem includes a bus server, and the bus server includes: The location information collection module is used to collect and monitor the longitude and latitude location information data of the bus, and the card swiping information collection module is used to collect and monitor the card swiping information data when boarding the bus; The neural network prediction module trains the data collected by the location information collection module and the card swiping information collection module through the LSTM neural network, and then performs multi-station, multi-line, and multi-variable arrival predictions for buses running during working hours to obtain accurate arrival times. At the same time, the scheduling of buses and the predicted arrival values ​​play an auxiliary role in the early warning of public conflicts; The environmental influencing factor indicators are mainly the temperature, whether there is rainfall, snowfall, wind speed and the time series generated by the time in the urban flow network system.

2. A data collection system based on urban mobile network as claimed in claim 1, characterized in that: The taxi temporal distribution subsystem includes a taxi server; The intelligent devices of the taxi Internet of Vehicles include an adapter box, a meter, a service evaluator, and a reversing camera connected to the on-board terminal. The on-board terminal is connected to the smart ceiling light screen and the dispatch information screen through the adapter box. The on-board terminal transmits the collected data of the smart ceiling light screen, the dispatch information screen, the meter, the service evaluator, and the reversing camera to the taxi server. At the same time, the built-in GPS positioning device and speed measuring device of the on-board terminal send the real-time location information and speed data of the taxi to the taxi server. The taxi server performs relevant calculations through the road traffic temporal analysis module to generate a regional distribution map.

3. A data collection system based on urban mobile network as claimed in claim 1, characterized in that: The road traffic temporal analysis module is used for traffic situation analysis based on big data, including forming road network elements, building a real-time traffic network, and collecting and publishing road conditions; Realize travel area division: divide the city into several areas according to the density of urban citizens' travel and the distribution of communities, and grid the city areas; Realize the analysis of travel patterns in hot spots: calculate the distribution map of vehicle dispatch areas within a period of time; The regional distribution maps include: departure area distribution map, destination area distribution map, hot departure area OD map, hot destination area OD map and travel OD trend statistics map.

4. The data collection system based on urban mobile network as claimed in claim 1 is characterized in that: The non-motor vehicle monitoring subsystem includes a non-motor vehicle server; The real-time monitoring modules at the intersections of the city roads send the monitored data to the non-motor vehicle server through the data sending module; The target detection and recognition module detects pedestrians and non-motorized vehicles on the road in real-time monitoring through the YOLO neural network, analyzes the targets, and generates pedestrian and non-motorized vehicle flow maps at urban intersections through the urban intersection flow map module, and transmits the detected target information to the non-motorized vehicle server.

5. A data collection method based on an urban mobile network, characterized in that: include: The bus real-time and prediction subsystem, taxi temporal distribution subsystem, and road pedestrian and non-motorized vehicle monitoring subsystem form an urban mobility network; The bus real-time and prediction subsystem is used to collect and process the location information and card swiping information of the bus, and the neural network is used for training to obtain the bus arrival prediction data; The taxi temporal distribution subsystem is used to process the real-time location information and speed of taxis and generate a regional distribution map; The non-motorized vehicle monitoring subsystem is used to analyze the pedestrians and non-motorized vehicles on the roads monitored at the urban road intersections to obtain the pedestrian and non-motorized vehicle flow diagram at the urban road intersections; The data obtained by the bus real-time and prediction subsystem, taxi temporal distribution subsystem, and road pedestrian and non-motorized vehicle monitoring subsystem are combined and modeled to build a dynamic model of the urban flow network. Each subsystem is superimposed with the corresponding environmental impact factor indicators, and the data of each subsystem is fitted in space and time to establish a temporal spatial distribution map of the city with fine granularity, so as to achieve conflict warning based on the urban flow network; The neural network prediction module trains the data collected by the location information collection module and the card swiping information collection module through the LSTM neural network, and then performs multi-station, multi-line, and multi-variable arrival predictions for buses running during working hours to obtain accurate arrival times. At the same time, the scheduling of buses and the predicted arrival values ​​play an auxiliary role in the early warning of public conflicts.

6. A data collection method based on urban mobile network as claimed in claim 5, characterized in that: The data obtained from the bus real-time and prediction subsystem, the taxi temporal distribution subsystem, and the road pedestrian and non-motorized vehicle monitoring subsystem are combined for modeling: the data from the bus real-time and prediction subsystem and the data from the taxi temporal distribution subsystem are imported into the map model in parallel. The map model is Amap or Baidu SDK, which reflects the vehicle driving status on the road. The road pedestrian and non-motorized vehicle data from the road pedestrian and non-motorized vehicle monitoring subsystem are imported into the map model.

7. A data collection method based on urban mobile network as claimed in claim 5, characterized in that: Constructing a dynamic model of the urban mobility network: To use time series to visualize the dynamic model of the spatiotemporal urban mobility network, that is, to use the fitting sequence constructed by environmental influencing factors and time series, the visualization uses kepler.gl technology and open source geospatial toolkit to establish the temporal spatial distribution map of the city in a fine-grained manner, so that buses, taxis, pedestrians at intersections and non-motorized vehicles can be dynamically displayed and moved on the map like particles.

8. The data collection method based on urban mobile network as claimed in claim 5 is characterized in that: The environmental influencing factor indicators are mainly the temperature, whether there is rain or snow, the wind speed and the time series generated by the time in the urban flow network system; Each subsystem is superimposed with corresponding environmental impact factor indicators. Specifically, a nonlinear regression fitting is performed using the three influencing factors in the environmental impact factor indicators and the time period to match the data of each day.

9. A terminal device, comprising a processor and a computer-readable storage medium, the processor being used to implement various instructions; the computer-readable storage medium being used to store multiple instructions, the instructions being suitable for being loaded by the processor and executing a public conflict warning method based on an urban mobile network as described in any one of claims 5-8.

Images