IoT Engineering: An Introduction to IoT + AI

When you go to work in the morning, you input your destination in the real-time traffic app and it will recommend the best driving route. When you encounter a traffic jam, the car music will play music that can soothe your anxious mood, and even use automatic driving to send you to the office. Can handle business. When it’s time to get off work, the automatic driving will automatically set the optimal route, and you can still handle your own affairs on the road. At the same time, the air conditioner, water heater, and rice cooker at home are all ready before you go home. These are all using artificial intelligence technology.

1. Introduction to Artificial Intelligence

Artificial intelligence first appeared in 1956. After years of development, it is now a discipline that studies and develops theories, methods, technologies and application systems for simulating, extending and expanding human intelligence. Artificial intelligence is a branch of computer science. By analyzing the essence of intelligence, it produces technologies that can respond in a way similar to human intelligence. The advantage is that it can quickly process massive data, replace a lot of repetitive labor of humans, and improve the level of intelligence. Research in the field includes language recognition, image recognition, natural language processing and expert systems. Since the birth of artificial intelligence, the theory and technology have continued to mature, and the application fields have also continued to expand. Artificial intelligence is the simulation of the information process of human consciousness and thinking. Artificial intelligence is not human intelligence, but it can think like a human. Artificial intelligence does not have an advantage in the study of natural laws unknown to humans, but it can assist humans in processing massive amounts of data and replace humans in undertaking a lot of repetitive labor.

2. Artificial intelligence related technologies

The subdivided fields of artificial intelligence technology applications include deep learning, computer vision, virtual personal assistants, natural language processing, real-time speech translation, context-aware computing, gesture control, automatic recognition of visual content, recommendation engines, etc.

1. Deep Learning

Deep learning is an important application branch in the field of artificial intelligence. For example, through learning and updating algorithms again and again, AlphaGo first defeated the European Go champion Fan Hui with 5:0 in 2015, and defeated the world Go with 4:1 in 2016. Champion Lee Sedol. AlphaG. The victory made people realize that the power of artificial intelligence can no longer be ignored. In the past ten years or so, the three major elements of algorithms, data and computing have boosted the rapid rise of artificial intelligence. The massive data stored in the Internet for many years has great application value: training machines.

The technical principle of deep learning:

(1) Build a network and randomly initialize the weights of all connections.

(2) Output a large amount of generated data to this network.

(3) The network automatically processes these actions and learns.

(4) If the corresponding action can be found, the weight will be increased, if not found, the weight will be decreased.

(5) The weight of the system is readjusted after the above process.

(6) Can exceed human performance after learning thousands of times.

2. Computer Vision

Computer vision is a discipline dedicated to the study of biological vision technology that allows computers to learn. It is a branch of artificial intelligence that allows computers to recognize objects, scenes and activities from images. Computer vision can be subdivided into different application fields, including medical imaging analysis to improve disease prediction, pre-diagnosis and treatment; face recognition is used in some confidential places, security and surveillance identification, payment and other fields.

The technical principle of computer vision: Computer vision technology uses image processing and other technologies to decompose the image into small tasks that are easy to manage, such as detecting the edge and texture of objects from the image. Classification techniques can be used to determine whether a recognized feature represents a class of objects known to the system for recognition.

3. Speech recognition

Speech recognition technology is to convert speech into text, and recognize and process it. The main applications of speech recognition include medical dictation, speech writing, computer system voice control, telephone customer service, etc.

Speech recognition technology principle:

Use the moving window function to process and frame the sound.

After framing, the sound becomes a digital waveform, and the waveform is extracted as an acoustic sign to become a state.

After feature extraction, the sound becomes a square matrix of N columns. Finally, phonemes are combined into words.

4. Virtual Personal Assistants

Virtual personal assistant products include Siri, Windows 10’s Cortana, etc.

Principles of virtual personal assistant technology (taking Siri as an example):

(1) After the user speaks to Siri, the voice is encoded and converted into a compressed digital file, which contains all the information of the user’s voice.

(2) The voice signal is sent to the user’s Internet service provider (ISP) through the mobile terminal and the mobile network, and the ISP has a cloud computing server.

(3) The modules in the server use technical means to identify the user’s voice content. The working principle of virtual assistant software is “local speech recognition + cloud computing service”.

5. Natural Language Processing

Natural language processing (NLP) is an important direction in the field of computer science and artificial intelligence, which integrates linguistics, computer science, mathematics and other disciplines to integrate language. The purpose of natural language processing is to realize language communication between man and machine.

Principles of language processing technology:

(1) Text encoding lexical analysis.

(2) Syntactic analysis.

(3) Semantic analysis.

(4) Text generation.

(5) Speech recognition.

6. Engine recommendation

Everyone often encounters this phenomenon when surfing the Internet: the website pushes you some relevant website content and advertisement content based on the pages you have browsed before and the keywords you searched. This can sometimes help us save a lot of energy. This is a manifestation of engine recommendation technology.

The purpose of Google’s free search engine is not to make money directly through search engine advertising rankings, but actually to collect a large amount of natural search data to enrich its large database and prepare for the subsequent artificial intelligence database.

Technical principle of engine recommendation: engine recommendation is an information network that actively discovers the current or potential needs of users and actively pushes information to users based on user behavior and attributes (data generated by users browsing websites), through algorithm analysis and processing.

With the development of artificial intelligence technology, more and more fields will use artificial intelligence technology. All aspects of human life such as medical care, education, finance, etc.

3. The combination of artificial intelligence and the Internet of Things

With the rapid development of the underlying technology of artificial intelligence, intelligent machines have now achieved a leap from “understanding the physical world” to “personalized scene implementation”.

Artificial intelligence is a discipline that can help us use mathematical computer tools to solve problems. It can free natural science researchers from a lot of repetitive labor and engage in more innovative work. More specifically, AI can help people understand how their own intelligence is formed.

The heat map of China’s AI industry in the “China-US AI Venture Capital Report” released by Tencent Research Institute shows that medical care is currently the hottest industry for AI applications, followed by the automotive industry that stands out with the development of driverless/assisted driving and other related technologies. The third echelon includes the real economy such as education, manufacturing, transportation, and e-commerce.

The combination of artificial intelligence and the Internet of Things will gradually penetrate into all walks of life and cause revolutionary changes. AI can replace humans in various technical work and part of the mental work in technology and tedious engineering, thus resulting in the social structure that has now been formed. drastic changes. On the other hand, job opportunities in new fields will be created. Therefore, artificial intelligence will readjust the social structure.

Artificial intelligence helps smart cities enter version 2.0. Big data and cloud computing are two assistants in building a smart city. A large amount of data is generated every moment in the fields of urban transportation, municipal administration, energy, and water supply. Artificial intelligence can extract effective information from the large amount of data generated by urban operation and development, and provide new ideas for the development, management, and control of smart cities .

Artificial intelligence provides motivation and methods for the transformation and upgrading of the manufacturing industry. Manufacturing is moving from automation to intelligence. Traditional robots are just automated mechanical devices that cannot intelligently adapt to the environment. Based on artificial intelligence, it is easy to realize intelligent manufacturing systems such as intelligent robots, intelligent factories, and intelligent supply chains. Artificial intelligence makes the manufacturing industry more efficient and brings about disruptive changes in production and organizational models.

The Internet of Things generates massive amounts of data, and these data will be a big gold mine in the era of artificial intelligence. Through the application of artificial intelligence, the Internet of Things will make our society more intelligent.