Jiadu Xintai Technology Co., Ltd. Liu Hongyin
1. Development status and characteristics of AIoT
1 ) Definition of AIoT
AIoT is one of the hot words that have become popular in recent years, specifically referring to the Internet of Things under the blessing of artificial intelligence. Through the combination of artificial intelligence (AI) and the Internet of Things (IoT), the traditional Internet of Things has stronger capabilities, higher "IQ", and a higher level of application.
With the advent of 5G, infrastructure construction has been able to support the interconnection of all things at the city level, and it can also achieve extensive communication and real-time connections between devices through various protocols with less manual intervention. AIoT is to add artificial intelligence to the network and equipment connected to everything. Through various AI algorithms, machine learning and other means, equipment and machines can make various decisions and make various equipment more intelligent and sensitive. Through the blessing of AI, IoT devices can change from pure programming/algorithm responses to dynamic decision-making, so as to connect devices, cover more scenarios, and create more value.
2 ) Development status of AIoT
Since 2016, the AIoT industry has begun to develop vigorously. So far, global AI technology and IoT technology have begun to take shape, and China's AIoT is also developing rapidly. It can be seen from several aspects: First, the technical level has accumulated enough. China’s global share of papers in the field of artificial intelligence has increased from 4.26% in 1997 to 27.68% in 2017, far ahead of other countries. The second is that the industry is highly popular and employs a large number of people. As of June 2018, a total of 4,925 artificial intelligence companies have been monitored worldwide, including 2,028 artificial intelligence companies in the United States, ranking first in the world. The total number of artificial intelligence enterprises in China is 1,011, ranking second in the world, followed by the United Kingdom, Canada, and India; third, the government attaches great importance to it. The State Council issued the "New Generation Artificial Intelligence Development Plan" in 2017, proposing a three-step strategy——"The first step is to synchronize the overall technology and application of artificial intelligence with the world's advanced level by 2020. The artificial intelligence industry has become a new important economic growth point, and the application of artificial intelligence technology has become a new way to improve people's livelihood. In 2025, the basic theory of artificial intelligence will achieve a major breakthrough, some technologies and applications will reach the world's leading level, artificial intelligence will become the main driving force for China's industrial upgrading and economic transformation, and the construction of an intelligent society will make positive progress; the third step is to reach the world's leading level in terms of artificial intelligence theory, technology and application in general by 2030, and become the world's major artificial intelligence innovation center, and the smart economy and intelligent society will achieve remarkable results, laying an important foundation for becoming one of the top innovative countries and an economic power."
3 ) Technical characteristics of AIoT
From a technical point of view, AIoT is not a simple system or platform, but a huge technical system, which mainly includes four levels: intelligent sensing equipment, AIoT basic capability platform (AIoT operating system), AIoT common applications, and AIoT industry-specific applications, and is finally delivered through integrated services.
If the AIoT system is compared to a person, the intelligent sensing device is the "five senses" of AIoT. It can complete the collection of various environmental data such as video images, temperature and humidity, sound, pressure, and speed, and process them through technical measures such as access, aggregation, and governance, and aggregate them to the AIoT basic capability platform (AIoT operating system). The AIoT basic capability platform (AIoT operating system) layer is equivalent to the "trunk and brain" of the AIoT system. In addition to being able to connect and control various intelligent sensing devices, it also provides capabilities such as data intelligence, intelligent analysis of twin cities, data processing, and 3D reality. The PaaS form exists. AIoT common applications and AIoT industry-specific applications focus on industries and scenarios, and solve specific industry pain points and problems through data capabilities.
2. Application problems and bottlenecks of AIoT in smart communities
1) IoT standards have not yet been unified. The core of the Internet of Things and artificial intelligence is data. The Internet of Things has a high demand for data access due to its many sensing devices and different characteristics of the carrying network. So far, although there are a series of industry de facto standards, there is still no unified data access and data governance standards, which need to be further improved in the next stage of development.
2) The cost of AI capabilities is relatively high. At present, artificial intelligence mainly relies on supervised learning, through a large number of training models to "teach" artificial intelligence to answer binary or even multi-classification problems such as face recognition and prediction. On the one hand, this requires a large amount of manually calibrated sample data, and on the other hand, it also requires a large amount of computing resources in the training process, and its labor and resource costs are relatively high.
3) The application scenarios are yet to be explored. In the smart community scenario, artificial intelligence and Internet of Things technologies are currently used more, and the face-face access control and other scenarios are still the majority. All kinds of data generated in the operation of the community have not been fully integrated and deeply applied. There is still a lot of room for development in this area.
3. In the era of big data, the in-depth application of AIoT in smart communities
Since the reform and opening up 40 years ago, the rapid growth of economic construction has effectively promoted the vigorous development of housing construction. In the 21st century, the construction of Chinese communities has entered a new period of development, that is, it has entered an era that focuses on "quality". It can be summed up in one sentence: "In the 1970s, we solved the problem of whether or not, and in the 1980s, we solved the problem of size;
With the development of the country's economy, the process of urbanization, and the influx of a large number of migrants, the urban population has the characteristics of a large base, high mobility, scattered living locations, and high difficulty in management and control. Various public security, criminal cases, and other social problems derived from this are also on the rise year by year.
With the rapid economic development, a large number of migrants flow into the city, which not only provides a large amount of labor force for urban development, but also forms a huge demand market for cheap housing. Because of the low rent, there are a large number of migrant workers living together, the population is highly mobile, and the composition of the residents is complex, and various demands will naturally increase. This series of new community management issues has always been a difficult point in the work of grassroots departments in various places, and the original working methods are difficult to meet the new community management work requirements.
From the perspective of demand, the community mainly faces the following pain points: first, it is necessary to expand information reporting channels to fully obtain the opinions and demands of community residents; second, it is necessary to connect various community governance entities, so that information can be exchanged and the status is clear; third, it is necessary to be able to detect the status of various risks and hidden dangers in the community in real time, and to detect problems in a timely manner; fourth, it is necessary to make full use of intelligent equipment to monitor and warn of abnormalities;
Therefore, based on the concept of community block data and making full use of the capabilities of AIoT, the new smart community can deeply explore the following applications:
1 ) Integrated sensing area
The "wisdom" of the community comes from the full perception of community conditions and the comprehensive analysis and application of a large amount of data. Therefore, it is necessary to build an IoT sensory network that is dominated by video images and multiple sensory resources are associated and superimposed, as well as a corresponding sensory big data resource service system. Realize the integration and aggregation of information such as videos, faces, vehicles, and IoT devices, and achieve the association and fusion of information such as people, things, cars, houses, and events. Provide a resource-sharing, open-capable, safe and controllable diversified sensory big data platform for various application scenarios, and support the community to carry out various data intelligence applications.
2 ) Insight into the overall situation with one brain
Integrate various data such as community people, things, things, etc., deconstruct, intersect, and fuse various "strip data" to form a "community data cerebellum" that includes various data files of community people, places, things, things, organizations, etc., clarify the basic situation of the community, and support management and service development.
"Community Data Cerebellum" is based on the full lifecycle management of data, with centralized and unified management as the core, and comprehensively regulates data collection, integration, sharing, openness, application, security, etc., through data aggregation, data fusion, and intelligent allocation, to maximize the value of data, release data dividends, provide timely and effective support for community scientific management and comprehensive, coordinated, and sustainable development, and support the realization of community operations with "clear background, clear situation, wide linkage, smart management, accurate service, knowledge of risks, and quick response."
Data aggregation highlights diversification and multi-channels. While collecting data resources such as grid management, public security, rental housing, public security management, and Internet technology companies, it actively collects data on people, houses, cars, venues, and events within the jurisdiction to form a comprehensive data depiction of the community. At the same time, the rule of law needs to be guaranteed to ensure that the storage and use of data is in accordance with laws and regulations, safe and reasonable.
Data fusion needs to study and determine the update frequency, update method, and update method according to the characteristics of each type of basic data. Through data collision and comparison, the freshness of data is guaranteed.
Intelligent allocation needs to establish a three-level linkage mechanism of districts, streets, and communities to ensure that data can be correlated and information can be connected, so that the thousands of threads above can be twisted into one rope, and the role of data in decision support and analysis can be brought into play.
3 ) One-picture perspective panorama
Build a smart community AR three-dimensional real-scene fusion command and dispatch platform, oriented to key business scenarios such as community management, service, emergency response, prevention and control, etc., to realize comprehensive monitoring and command and dispatch of community operation status. The platform uses three-dimensional scenes as the carrier to monitor and dispatch all key data of the core system of community operation. It supports daily operation monitoring and management from multiple dimensions such as community planning, enterprise operation, asset management, traffic conditions, hidden dangers, environmental monitoring, urban management, grid events, public security prevention and control, emergency response, precision services, etc., as well as emergency command and dispatch management of emergencies. Empowerment.
4 ) One frame responds to every call
Based on the idea of "grid-based, Internet of Things as a means", sort out the existing community responsibility list and response list, improve the construction of basic data collection and management tools, strengthen the ability of intelligent monitoring and early warning, deeply mine and utilize data such as management events, response tasks, community police situations, hidden risks, daily patrol collection, emergency command, etc., plan and design a community management evaluation system, improve grid event reporting and handling capabilities from multiple dimensions such as management, emergency response, and prevention and control, improve key work management and supervision mechanisms, and achieve standardized management procedures.
At the same time, a virtual grid is established through technical support means such as the Internet, integrated communication, and big data, to achieve online and offline integration, to enrich the channels for communication and coordination between the masses and grid members, government departments, and consultation and notification channels, to mobilize the enthusiasm of the masses to participate in community governance and construction, and to enhance the sense of belonging of the masses.
The ultimate goal is to build a grassroots party building pattern and a social governance pattern of co-construction, co-governance and sharing, promote the modernization of grassroots governance, and realize the work effect of "what the people call, the party and the government respond, the streets and towns call, the department responds, the enterprise calls, the service responds, the party calls, and the whole people respond."
5 ) One Code Co-governance and Sharing
At the front end, the two-dimensional code is used as the entrance, and the community and residents are connected through public accounts, small programs and other carriers, making full use of the real-name identity system built by various city codes during the epidemic, converting the health code into a management code/service code, connecting online and offline, and integrating the health code to achieve one-code access and one-net service.
At the back end, efforts will be made to establish a general portal for community precision services to accurately grasp and timely discover the difficulties and needs of the masses; sort out the community service list, determine service work procedures, integrate community service resources, improve service infrastructure, broaden service supply channels, and solve the needs of the masses in a humanized way. Realize the transformation of community governance from management to service, and provide more convenient and efficient community government affairs and people's livelihood services. Finally, a basic framework of smart community services with overall coordination, efficient operation, precise service, and scientific management will be formed.
5. In the era of big data, the application prospect of AIoT in the smart community
In the era of big data, the application prospect of AIoT in the smart community is undoubtedly broad. The source of big data in smart communities comes from AIoT, but it cannot be understood purely from the perspective of data volume—community is the smallest unit and the forefront of social governance. The absolute value of community big data data is not necessarily large, but it covers various management and service departments such as public security, civil affairs, health, land, urban management, meteorology, Guangzhou, and comprehensive management, as well as all aspects of information generated in community production and life. "Block data" deconstruction, intersection, fusion, to be precise, it should be called community block data.
In the process of community block data collection, it includes the filling of data space, the reconstruction of spatial data, the organization of the collection process, and the collection during the organization process, as well as the collection of new data and the derived data after the original data combination. Through the application of block data, higher and more value of data can be mined.
In the field of social governance, under the guidance of block data thinking, the existing behavior of the government will undergo a huge change. For community authorities (subdistrict offices), block data can break barriers, connect isolated islands, and promote the effective development of work through information sharing and work linkage. For the masses, with the normalization of data disclosure, the public's participation in social governance will also be more effective. At the same time, the application of block data can make government decision-making more scientific and forward-looking, and it will become possible to build a service-oriented and intelligent community.