According to relevant sources, there are currently at least 1,700 Artificial Intelligence (AI) related startups worldwide. In more than 70 countries around the world, investors have invested more than $14.6 billion in these startups. Investors expect AI to generate revenues of $47 billion by 2020, compared to $8 billion in 2016. So, what kind of application prospects does artificial intelligence have in the aviation industry? What changes will happen in the future due to AI? The so-called AI refers to some computer programs that can be as intelligent as humans, such as logical reasoning, problem solving, and active learning. There are two types of AI performance, one is physical, such as robots , and the other is non-entity, such as Apple Siri and Google Now. Up to now, our understanding of AI can divide the development of AI into four generations. First generation AI: rule-driven reasoning This generation of AI began with the successful integration of people's knowledge into computer programs through the definition of static rules. For example, in the early days, chess programs were run through a set of logical rules defined in advance. In actual operation, they rarely perceive changes in the outside world, and they have no ability to learn and refine knowledge. GPS navigation software is another case of the first generation of AI, which can sense the geographical location, analyze the location information on the map, and give a direction of progress. However, the ability of GPS navigation software to handle path deviations is very limited. It is not difficult to see that the first generation of AI computer programs are still widely used and play their value in various industries. Examples of applications that already exist in the aviation industry are: 1) Display the condition of the aircraft to the pilot in the cockpit through some aircraft instrumentation. For example, the relative values ​​of the attitude of the aircraft and the horizon. 2) Auto-driving function. The flight of the aircraft is automatically controlled according to a predefined flight path without the intervention of the pilot. 3) Automatically adjust the pressure in the aircraft cabin to ensure the safety and comfort of the cabin environment. Second generation AI: learning through big data Historically, dealing with large amounts of data is a headache. But with the development of technology, what is now a headache has become a huge opportunity. The second generation of AI is using these new technologies to learn from big data. Through deep learning architectures such as neural networks, computers gain knowledge through constant experimentation and error handling. The main feature of the second generation AI is learning through data analysis, but it does not have the ability to logical reasoning, nor can it understand the context, and it is difficult to extract knowledge in different fields. This type of learning is still very different from the way people learn. The usual practice is to simulate the problem that needs to be solved through a pre-defined static model, and then continue to use the data to train the model. The second generation of AI is sometimes even very unintelligent. For example, Microsoft's Twitter Bot is an example that will misinterpret the training data and give unreasonable results. Waveguide Adaptor,Coaxial To Waveguide Adapter,Waveguide To Coaxial Adapter,Waveguide To Coaxial Transition Chengdu Zysen Technology Co., Ltd. , https://www.zysenmw.com