Design of an Intelligent Service Robot based on Deep Learning

Abstract

With the rapid development of science and technology and social needs, various service robots have been widely used in cleaning, medical treatment, tour guide, security and other fields of daily life, and they are constantly changing people's lifestyles and improving people's living standards and quality .Based on the research on the application technology and development of service robot in recent years, depending on the deep learning method in artificial intelligence, this paper proposes and designs an intelligent service robot, which is used to provide users with intelligent navigation, voice interaction and beverage bottle recycling and other functional services in shopping malls, airports, convention and exhibition centers. The robot takes STM32 microprocessor as the core of the control system, carries NVIDIA TX2 artificial intelligence board, and is equipped with autonomous navigation and positioning module and various sensors. Through self-developed application control program, the robot can realize autonomous navigation, walking, voice dialogue and other functions; at the same time, it adopts the target detection algorithm single shot multibox detector based on deep learning (SSD) realizes the recognition of different beverage bottles and recycles them to the robot, and gives users corresponding bonus points. Experiments show that the intelligent service robot can complete the tasks of autonomous navigation, voice interaction and beverage bottle recycling.