Urban field data sensing and decision support system and method

Abstract

The application discloses a kind of urban field data sensing and decision support system and method, it is related to urban intelligent sensing, Internet of Things, edge computing and data analysis technical field.The system includes AIoT sensing node, edge computing module, low-power wide-area communication module, AI central control platform, data analysis module, report generation module and decision support module.AIoT sensing node is used to collect the environmental state of urban field, people flow activity, space use and equipment state data;Edge computing module is used for local screening, feature extraction, preliminary identification of anomaly, data aggregation and compression, to generate numerical feature data;Low-power wide-area communication module uploads numerical feature data to AI central control platform;AI central control platform accesses, washes, fuses, stores and visualizes to multiple point data, and generates environmental anomaly, hot risk, people flow change, space use and equipment state etc.Index;Report generation module generates structured data analysis report according to index and preset scene template;Decision support module outputs alarm information, evaluation result and decision support information.The application can realize low-power consumption, privacy-friendly, long-term deployment of urban field data acquisition and analysis, and convert scattered field monitoring data into reusable data assets and governance decision support.

Description

Technical Field

[0001] This invention relates to the fields of urban intelligent sensing, Internet of Things, edge computing and data analysis technology, and in particular to an urban on-site data sensing and decision support system and method, which can be used in scenarios such as urban governance, park management, public space optimization, low-carbon governance, campus management, community governance, park management and urban public space status assessment. Background Technology

[0002] With the increasing demand for urban governance, low-carbon development, smart parks, and refined management of public spaces, government departments, park operators, university research teams, planning and design institutions, and government think tanks are increasingly in need of obtaining continuous data from the urban site, such as environmental conditions, pedestrian activity, space usage, changes in thermal risk, noise levels, air quality, and equipment operating status.

[0003] Current methods for collecting urban field data typically suffer from the following problems. First, different entities often deploy equipment ad-hoc for their respective projects. For example, government departments deploy monitoring systems for decision-making, university teams collect data for research projects, think tanks organize data sampling for research reports, and park operators deploy sensing equipment for operation and maintenance. It is difficult to standardize equipment systems, data definitions, and analysis methods across different projects, easily leading to redundant construction and data silos.

[0004] Second, many urban environmental and public space monitoring projects remain at the stage of project-based, outsourced, or one-off sampling. After the project ends, the equipment may be idle, data collection may be interrupted, historical data may be difficult to accumulate continuously, relevant experience may be difficult to reuse, and long-term continuous urban field data assets may not be formed.

[0005] Third, while traditional camera solutions can acquire relatively rich on-site information, they suffer from high energy consumption, complex wiring, and high privacy compliance risks, limiting their adoption in scenarios such as campuses, parks, communities, streets, and public spaces. Traditional sensor nodes, on the other hand, suffer from insufficient battery life, frequent maintenance, limited data dimensions, and insufficient long-distance transmission capabilities, making it difficult to support long-term, multi-location, and low-maintenance-cost outdoor deployments.

[0006] Fourth, existing data platforms mostly remain at the level of data display or single monitoring, making it difficult to unify the access, continuous analysis, and automatic alarm of multi-source data such as environment, people flow, space usage, and equipment status, and further generate structured data analysis reports and assessment results for governance decision-making.

[0007] Therefore, there is a need for an urban field data perception and decision support system and method that can be deployed long-term, operate with low power consumption, is privacy-friendly, can be uniformly analyzed, can generate reports, and can support governance decisions. Summary of the Invention

[0008] The purpose of this invention is to provide an urban field data perception and decision support system and method to solve the problems of redundant construction, data dispersion, high maintenance costs, high privacy risks, difficulty in continuously accumulating data assets, and difficulty in transforming data into governance decision support in existing urban field data collection methods.

[0009] To achieve the above objectives, the present invention provides an urban field data perception and decision support system, comprising: an AIoT sensing node, an edge computing module, a low-power wide-area communication module, an AI central control platform, a data analysis module, a report generation module, and a decision support module.

[0010] The AIoT sensing nodes are used to collect environmental status data, pedestrian activity data, space usage data, and equipment status data at preset urban spatial locations, and associate the collected data with location identifiers and collection time. The urban spatial locations may include one or more locations in campuses, industrial parks, parks, communities, streets, squares, around stations, urban public spaces, or low-carbon governance demonstration areas.

[0011] The environmental status data may include one or more of the following: temperature, humidity, particulate matter, carbon dioxide concentration, noise, light intensity, wind speed, wind direction, air pressure, or other environmental parameters. The pedestrian activity data may include one or more of the following: pedestrian quantity, activity intensity, dwell time, flow rate changes, movement direction, traffic activity status, or space occupancy status. The equipment status data may include one or more of the following: node voltage, battery level, communication status, sensor operating status, equipment offline status, or abnormal status.

[0012] The edge computing module is used to locally filter, extract features, perform preliminary anomaly identification, aggregate and compress data collected by AIoT sensing nodes, and generate numerical feature data. Through local processing by the edge computing module, invalid data backhaul can be reduced, lowering communication bandwidth, cloud computing pressure, and system energy consumption.

[0013] The edge computing module can process the collected data based on preset rules, threshold judgment methods, statistical analysis methods, machine learning models, or combinations thereof. The processing results may include environmental features, pedestrian flow features, space usage features, abnormal state features, equipment status features, or other numerical features used for platform analysis.

[0014] The low-power wide-area communication module is used to upload the numerical feature data to the AI ​​central control platform. The low-power wide-area communication module may include a LoRaWAN communication module, or it may employ NB-IoT, 4G, 5G, Wi-Fi, Ethernet, or other communication methods depending on the deployment environment. Preferably, LoRaWAN communication is used in open outdoor, low-power, and long-distance communication scenarios.

[0015] The AI ​​central control platform is used to access, clean, fuse, store, and visualize numerical feature data from multiple locations, forming a continuous urban site status dataset. This urban site status dataset can be organized and managed according to location, time, scenario, data type, device number, or project task.

[0016] The data analysis module is used to generate one or more of the following indicators based on the urban site status dataset: environmental anomaly indicators, thermal risk indicators, pedestrian flow change indicators, space usage indicators, and equipment status indicators. The data analysis module can perform comparative analysis, trend analysis, anomaly identification, spatial distribution analysis, and comprehensive evaluation of data from different locations, time periods, and scenarios.

[0017] In one embodiment, the data analysis module can calculate the differences between locations, the temporal trend, the degree of abnormal deviation, and the spatial distribution status based on the temperature, humidity, particulate matter, carbon dioxide concentration, noise, and pedestrian activity characteristics at different locations, and generate thermal risk indicators, environmental anomaly indicators, pedestrian flow change indicators, space usage indicators, and equipment status indicators accordingly. These indicators can be categorized, stored, and retrieved according to location, time period, scenario type, or project task.

[0018] The report generation module is used to generate a structured data analysis report based on the indicators and preset scenario templates. The preset scenario templates may include one or more of the following: park assessment template, thermal risk analysis template, public space use assessment template, low-carbon governance template, equipment operation and maintenance template, and policy assessment template.

[0019] The decision support module is used to output alarm information, assessment results, and decision support information based on the structured data analysis report, targeting urban governance, park management, public space optimization, or low-carbon governance. The decision support information may include risk warnings, space optimization suggestions, equipment maintenance suggestions, low-carbon management suggestions, public space operation suggestions, or policy evaluation reference information.

[0020] This invention also provides a method for urban site data perception and decision support, comprising the following steps: collecting environmental status data, pedestrian activity data, space usage-related data, and equipment status data of the urban site through AIoT sensing nodes, and associating the data with location identifiers and collection time; performing local filtering, feature extraction, preliminary anomaly identification, data aggregation, and compression on the collected data at the edge to generate numerical feature data; uploading the numerical feature data to the AI ​​central control platform via low-power wide-area communication; the AI ​​central control platform accessing, cleaning, fusing, storing, and visualizing the numerical feature data from multiple locations to form a continuous urban site status dataset; generating one or more of environmental anomaly indicators, thermal risk indicators, pedestrian flow change indicators, space usage indicators, and equipment status indicators based on the urban site status dataset; generating a structured data analysis report based on the indicators and a preset scenario template; and outputting alarm information, evaluation results, and decision support information for urban governance, park management, public space optimization, or low-carbon governance based on the structured data analysis report.

[0021] Compared with existing technologies, this invention has the following beneficial effects. First, by combining urban field data collection, edge processing, and feature uploading through AIoT sensing nodes and edge computing modules, it can reduce communication bandwidth, cloud computing pressure, and system energy consumption. Second, by uploading numerical feature data instead of raw images, raw audio, or raw sensor waveform data, it can reduce privacy compliance risks and is suitable for public space scenarios such as campuses, parks, communities, and streets. Third, by using an AI central control platform to uniformly access, clean, fuse, store, and visualize continuous data from multiple locations, it can reduce data silos and form a sustainable urban field status dataset. Fourth, through data analysis, report generation, and decision support modules, field data is transformed into indicators, reports, and decision support information, improving the digital support capabilities for urban governance, park management, public space optimization, and low-carbon governance. Fifth, this invention can upgrade from temporary, project-based data collection methods to a sustainable, reusable, and data-assembly-generating urban field data infrastructure. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the overall architecture of the urban field data perception and decision support system provided in an embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram of the AIoT sensing node structure provided in an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram of the urban field data perception and decision support method provided in an embodiment of the present invention. Detailed Implementation

[0025] The present invention will be further described below with reference to the accompanying drawings and embodiments. It should be understood that the following embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0026] like Figure 1 As shown, this embodiment provides an urban field data perception and decision support system, including multiple AIoT sensing nodes, an edge computing module, a low-power wide-area communication module, an AI central control platform, a data analysis module, a report generation module, and a decision support module. Multiple AIoT sensing nodes are distributed and deployed in campuses, industrial parks, parks, communities, streets, or other urban public spaces to continuously collect urban field status data.

[0027] In one embodiment, AIoT sensing nodes are deployed in a low-carbon park. The nodes collect data on temperature, humidity, particulate matter, carbon dioxide concentration, noise, and pedestrian activity, and record the corresponding location identifier and collection time. An edge computing module locally filters the collected data, performs preliminary anomaly identification, and aggregates the data to generate numerical feature data. This numerical feature data is uploaded to the AI ​​central control platform via a LoRaWAN communication module. The AI ​​central control platform cleans, merges, stores, and visualizes the data from multiple locations, generating indicators such as park environmental status, thermal risk, pedestrian flow changes, and equipment status. A report generation module calls a park assessment template or a low-carbon governance template to generate a park assessment report or a low-carbon governance report. A decision support module outputs suggestions for park operation and maintenance, space optimization, or low-carbon management based on the report.

[0028] like Figure 2 As shown, an AIoT sensing node may include an environmental sensing unit, a people movement sensing unit, a power supply unit, a communication unit, an edge computing unit, and an outdoor installation structure. The environmental sensing unit may include temperature and humidity sensors, particulate matter sensors, carbon dioxide sensors, noise sensors, or other environmental sensors. The people movement sensing unit may employ millimeter-wave sensing units, infrared sensing units, radar sensing units, or other non-visual sensing units. The power supply unit may include solar power components and energy storage components. The outdoor installation structure may include a waterproof housing, a breathable structure, mounting brackets, and a sensor fixing structure.

[0029] In one embodiment, the edge computing module is deployed inside the AIoT sensing node or in a nearby edge device. Based on preset rules or machine learning models, the edge computing module performs invalid data filtering, feature extraction, initial anomaly detection, and data aggregation on the collected data. For example, for temperature, humidity, particulate matter, and carbon dioxide concentration data, the edge computing module can generate numerical features such as mean, maximum, minimum, rate of change, and anomaly markers; for pedestrian activity data, the edge computing module can generate numerical features such as traffic flow, density, dwell time, activity intensity, or traffic conditions.

[0030] In one embodiment, the system does not upload the original images, original audio, or original sensor waveform data, but instead uploads numerical feature data processed by the edge computing module. This approach reduces privacy compliance risks and decreases communication bandwidth and energy consumption.

[0031] like Figure 3 As shown, this embodiment provides a method for urban site data perception and decision support. First, AIoT sensing nodes are deployed at preset urban spatial locations to collect environmental status data, pedestrian activity data, space usage-related data, and equipment status data, and the data is associated with location identifiers and collection times. Second, at the edge, the collected data undergoes local filtering, feature extraction, preliminary anomaly identification, data aggregation, and compression to generate numerical feature data. Third, the numerical feature data is uploaded to the AI ​​central control platform via low-power wide-area communication. Then, the AI ​​central control platform accesses, cleans, fuses, stores, and visualizes the numerical feature data from multiple locations, forming a continuous urban site status dataset. Subsequently, based on the urban site status dataset, one or more of the following indicators are generated: environmental anomaly indicators, thermal risk indicators, pedestrian flow change indicators, space usage indicators, and equipment status indicators. Finally, a structured data analysis report is generated based on the indicators and preset scenario templates, and alarm information, evaluation results, and decision support information are output based on the structured data analysis report.

[0032] In one embodiment, the system is applied to public spaces on university campuses. AIoT sensing nodes are deployed around teaching buildings, walkways, plazas, dormitories, or sports fields to collect data on microclimate, pedestrian activity, noise, and air quality. The AI ​​central control platform compares and analyzes data from different locations and time periods to identify heat risk points, peak space usage, abnormal noise, and abnormal equipment conditions. The report generation module generates analysis reports based on campus public space assessment templates, providing data support for green campus construction, campus space management, and public space optimization.

[0033] In another embodiment, the system is applied to urban streets or community public spaces. AIoT sensing nodes are deployed on streets, intersections, community entrances, park trails, or around stations to collect data on environmental conditions, pedestrian activity, and space usage. The AI ​​central control platform generates street space usage indicators, pedestrian flow change indicators, and environmental anomaly indicators based on continuous data, and forms public space optimization reports or community governance reports through a report generation module.

[0034] In another embodiment, the system is applied to research projects of government think tanks or planning and design institutions. Unlike one-time manual sampling, this system can continuously collect data from multiple locations, forming a continuous dataset of urban site conditions. The report generation module, based on policy evaluation templates or planning evaluation templates, writes indicator results, time trends, location comparisons, and anomaly identification results into a structured data analysis report, providing data support for policy evaluation, planning argumentation, and governance recommendations.

[0035] The modules in the above embodiments can be implemented through hardware, software, or a combination of both. The AI ​​central control platform, data analysis module, report generation module, and decision support module can be deployed on cloud servers, local servers, edge servers, or combinations thereof. The low-power wide-area communication module can adopt LoRaWAN, NB-IoT, 4G, 5G, Wi-Fi, Ethernet, or other communication methods depending on the actual deployment environment. The sensor type and number of AIoT sensing nodes can also be adjusted according to specific scenario requirements.

[0036] Those skilled in the art will understand that, without departing from the concept of the present invention, the specific sensor types, communication methods, data analysis models, report templates, and deployment scenarios in the above embodiments can be replaced, combined, or adjusted, and such replacements, combinations, or adjustments should all fall within the protection scope of the present invention.

Claims

1. A city-wide on-site data perception and decision support system, characterized in that, It includes: AIoT sensing nodes, edge computing modules, low-power wide-area communication modules, AI central control platforms, data analysis modules, report generation modules, and decision support modules; the AIoT sensing nodes are used to collect environmental status data, pedestrian activity data, space usage-related data, and equipment status data at preset urban spatial locations, and associate the data with location identifiers and collection times; The edge computing module is used to locally filter, extract features, preliminarily identify anomalies, aggregate and compress data collected by the AIoT sensing nodes to generate numerical feature data; the low-power wide-area communication module is used to upload the numerical feature data to the AI ​​central control platform; the AI ​​central control platform is used to access, clean, fuse, store and visualize the numerical feature data from multiple locations to form a continuous urban site status dataset; the data analysis module is used to generate one or more of the following based on the urban site status dataset: environmental anomaly indicators, thermal risk indicators, pedestrian flow change indicators, space usage indicators and equipment status indicators; the report generation module is used to generate a structured data analysis report based on the indicators and preset scenario templates; the decision support module is used to output alarm information, evaluation results and decision support information for urban governance, park management, public space optimization or low-carbon governance based on the structured data analysis report.

2. The urban field data perception and decision support system according to claim 1, characterized in that, The AIoT sensing node includes an environmental sensing unit, a pedestrian activity sensing unit, a power supply unit, a communication unit, and an outdoor installation structure. The environmental sensing unit is used to collect one or more data from temperature, humidity, particulate matter, carbon dioxide concentration, and noise. The pedestrian activity sensing unit is used to collect one or more data from pedestrian quantity, activity intensity, dwell time, traffic flow changes, or traffic activity status.

3. The urban field data perception and decision support system according to claim 2, characterized in that, The power supply unit includes a solar power supply component and an energy storage component. The outdoor installation structure includes a waterproof shell, a breathable structure, a mounting bracket, and a sensor fixing structure, enabling the AIoT sensing node to be deployed long-term in campuses, parks, communities, streets, or urban public spaces.

4. The urban field data perception and decision support system according to claim 1, characterized in that, The edge computing module processes the collected data based on preset rules and / or machine learning models, and performs invalid data filtering, feature extraction, anomaly initial judgment, data aggregation and low-frequency reporting at the edge to reduce communication bandwidth, cloud computing pressure and system energy consumption.

5. The urban field data perception and decision support system according to claim 1, characterized in that, The low-power wide-area communication module includes a LoRaWAN communication module, which transmits the numerical feature data to the AI ​​central control platform through a gateway, network server, message middleware, or database.

6. The urban field data perception and decision support system according to claim 1, characterized in that, The numerical feature data does not include the original images, original audio, or original sensor waveform data. The AI ​​central control platform analyzes the numerical feature data to achieve privacy-friendly monitoring of the urban site conditions.

7. The urban field data perception and decision support system according to claim 1, characterized in that, The report generation module includes a scenario template library and an automatic generation unit. The scenario template library includes one or more of the following: park assessment template, thermal risk analysis template, public space use assessment template, low-carbon governance template, equipment operation and maintenance template, and policy assessment template. The automatic generation unit is used to generate corresponding data analysis reports based on the analysis results of the data analysis module.

8. A method for urban field data perception and decision support, characterized in that, The process includes the following steps: collecting environmental status data, pedestrian activity data, space usage data, and equipment status data in the urban environment through AIoT sensing nodes, and associating the data with location identifiers and collection time; At the edge, the collected data undergoes local filtering, feature extraction, preliminary anomaly identification, data aggregation, and compression to generate numerical feature data. This numerical feature data is then uploaded to the AI ​​central control platform via low-power wide-area communication. The AI ​​central control platform accesses, cleans, merges, stores, and visualizes the numerical feature data from multiple locations, forming a continuous urban site status dataset. Based on this urban site status dataset, one or more of the following indicators are generated: environmental anomaly indicators, thermal risk indicators, pedestrian flow change indicators, space usage indicators, and equipment status indicators. A structured data analysis report is then generated based on these indicators and a preset scenario template. Based on the structured data analysis report, alarm information, assessment results, and decision support information are output for urban governance, park management, public space optimization, or low-carbon governance.

9. The urban field data perception and decision support method according to claim 8, characterized in that, The environmental status data includes one or more of the following: temperature, humidity, particulate matter, carbon dioxide concentration, and noise. The human activity data includes one or more of the following: number of people, activity intensity, dwell time, flow rate changes, or traffic activity status.

10. The urban field data perception and decision support method according to claim 8, characterized in that, When processing the collected data, the edge device does not upload the original image, original audio, or original sensor waveform data, but instead uploads the numerical feature data that has been filtered, aggregated, compressed, or feature extracted.

11. The urban field data perception and decision support method according to claim 8, characterized in that, The generation of environmental anomaly indicators, thermal risk indicators, pedestrian flow change indicators, space usage indicators, and equipment status indicators includes: based on continuous numerical feature data from multiple locations, comparative analysis of environmental status, pedestrian activity, space usage status, and equipment operation status at different locations, at different times, or in different scenarios to obtain corresponding indicator results.

12. The urban field data perception and decision support method according to claim 8, characterized in that, The generation of structured data analysis reports includes: calling a preset scenario template, writing one or more indicators from environmental anomaly indicators, thermal risk indicators, pedestrian flow change indicators, space usage indicators, and equipment status indicators into the corresponding report structure, and generating a park assessment report, a public space optimization report, a low-carbon governance report, an equipment operation and maintenance report, or a policy assessment report.

Images