Abstract
The Internet of Things (IoT) and Artificial Intelligence (AI) have been driving forces in propelling the technical innovation of intelligent manufacturing, promoting economic growth, and improving the quality of people’s lives. In an intelligent factory, introducing edge computing is conducive to expanding the computing resources, the network bandwidth, and the storage capacity of the cloud platform to the IoT edge, as well as realizing the resource scheduling and data uplink and downlink processing during the manufacturing and production processes. Moreover, the emotion recognition and interaction of the Affective Interaction Intelligence Robot (iRobot), with the IoT cloud platform as the infrastructure and AI technology as the core competitiveness, can better solve the psychological problems of the user. Accordingly, this has become a hot research topic in the field of intelligent manufacturing. In this paper, we describe an intelligent robot factory (iRobot-Factory), adopt a highly interconnected and deeply integrated intelligent production line, and introduce the overall structure, composition, characteristics, and advantages of such a factory in details from the two aspects of cognitive manufacturing and edge computing. Then, we describe the implementation of the volume production of iRobot using iRobot-Factory and look at the system performance experimental results and analysis of the iRobot-Factory and a traditional factory. The experimental results show that our scheme significantly improved both the chip assembly and the production efficiency, while the number of system instructions also decreased significantly. In addition, we discuss some open issues relating to cloud-end fusion, load balancing, and personalized robots to make reference to promoting the emotion recognition and interaction experience of users.
•This paper sets forth a new iRobot Factory system architecture from six points of view: intelligent terminal, system management, edge-computing node, cloud / cognitive engine, intelligent device unit, and the iRobot-Factory production line layer.•This paper discusses efficient information interaction, multimodal data fusion, and automatic production and puts forward two solutions for modeling and active operation and maintenance based on cognitive manufacturing, communication, and interaction based on edge computing.•This paper carries out the system-performance experiments between the iRobot-Factory and the traditional factory. The experimental results showed that iRobot-Factory significantly improved both chip assembly and production efficiency. The number of system instructions also decreased significantly.