Modeling, planning, and scheduling of shop-floor assembly process with dynamic cyber-physical interactions: a case study for CPS-based smart industrial robot production

In recent years, the applications of industrial robots are expanding rapidly due to Industry 4.0 oriented evolutions, ranging from automobile industry to almost all manufacturing domains. As demands with rapid product iterations become increasingly fluctuant and customized, the assembly process of industrial robots faces new challenges including dynamic reorganization and reconfiguration, ubiquitous sensing, and communication with time constraints, etc. This paper studies the industrial robot assembly process modeling, planning, and scheduling based on real-time data acquisition and fusion under the framework of advanced shop-floor communication and computing technologies such as wireless sensor, actuator network, and edge computing. Taking the assembly of industrial robots as the specific object, the multi-agent model of industrial robot assemble process is established. Then, the encapsulation, communication, and interaction of agents with real-time data acquisition and fusion are studied. Based on multi-agent reinforcement learning approach, an intelligent planning and scheduling algorithm for industrial robot assembly is proposed, and a simulation case is presented to demonstrate the proposed model and algorithm.