Driven by the high-tech of intelligence and automation, and based on "people's yearning for a better life", the smart home industry has become a trendy field of the development of the times. However, when we do social research, we find that for some users who are inconvenient in life, unable to take care of themselves (especially the elderly or the disabled) or more pursuing fast life, the rational use of furniture is not easy. However, as a carrier or form of life, furniture is not only to meet the basic needs of most people, but also to meet all people's yearning and pursuit for a better life. This also requires that it should be developed, iterated and even evolved with the progress of the times. For example, in real life, the disabled cannot use furniture normally, the improper use of electrical appliances in ordinary people's lives leads to large energy consumption of resources, and the voltage overload of electrical appliances detracts from the life of electrical appliances. Smart home applications provide users with a platform and a new way of personalized use of home, with objective research and application value. At present, the most widely used in the market is the maximum matching word segmentation algorithm, which performs semantic matching through the user's speech behavior recognition, but there are still many problems, such as dialect recognition and deaf-mute people. It is under this background that our group has carried out the creation and innovation in this direction. Compared with the smart home products in today's society, this project starts from the idea of neural network gesture recognition, using the ML algorithm and graph model theory currently popular in academia and industry, using radar sensing technology, radar signal matrix processing and other technologies to realize the control of smart home, using convolutional neural network based on distance perception and speed perception to carry out model design, involving artificial intelligence, deep learning, sensors and other aspects, strive to make smart furniture really convenient, through simple actions can realize the control of furniture, so as to better serve people, so that the so-called high-tech is no longer mysterious, but in our lives. This technology is not limited to smart homes. In the future, we will conduct in-depth research and innovation on the control of drones and the automatic driving control of smart cars, which has a very wide range of application prospects and value.