Knowledge-based program generation approach for robotic manufacturing systems

• Published in: Robotics and computer-integrated manufacturing Vol. 73; p. 102242
• Main Authors: Zheng, Chen, Xing, Jiajian, Wang, Zhanxi, Qin, Xiansheng, Eynard, Benoît, Li, Jing, Bai, Jing, Zhang, Yicha
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.02.2022; Elsevier BV; Elsevier
• Subjects: Engineering Sciences; Greenhouse effect; Industrial robots; Knowledge-based engineering; Manufacturing; Manufacturing programming; Mechanical engineering; Mechanics; Product models; Program generation; Reasoning; Robotic manufacturing system; Robotics; Semantics; Standardization; System effectiveness; Workpieces; Knowledge-based engineering; Program generation; Manufacturing programming; Robotic manufacturing system
• ISSN: 0736-5845; 1879-2537
• DOI: 10.1016/j.rcim.2021.102242

•Explicit semantic descriptions of the concepts in the domains of product model of workpieces, robotic manufacturing process, robotic manufacturing systems, and robotic manufacturing program are defined, which can be easily understood by users and processed by computers.•A formal description of the intra- and inter-domain relationships among the concepts of the afore-mentioned domains are provided, based on which the reasoning mechanism is established to infer the sequence of basic instruction units of the required robotic manufacturing program.•Based on the semantic descriptions and reasoning mechanism, the proposed automatic program generation approach provides an effective support for the standardization of the rules and knowledge relating to the manufacturing program that have been proven by previous successful manufacturing cases, which can greatly improve the manufacturing stability and production quality. In recent decades, robotic manufacturing systems have been considered as effective solutions for providing more productive manufacturing processes, but with less cost and risk. However, the programming for robotic manufacturing systems is a time-consuming task, and hinders the implementation of robotic manufacturing systems in today's industry. This paper proposes a knowledge-based program-generation approach for robotic manufacturing systems. The proposed approach provides effective support for the standardization of the rules and knowledge related to manufacturing programs that have proven successful in previous manufacturing cases; this can not only increase the programming efficiency, but can also improve the manufacturing stability and production quality. First, an ontological knowledge model is developed to provide an explicit semantic description of the relevant concepts for the robotic manufacturing system, basic instruction units for the program, and product models of the workpieces. Second, a rule-based reasoning mechanism is established to infer the implicit relationships between the basic instruction units of the manufacturing program. Finally, based on the semantic descriptions and reasoning mechanism of the proposed knowledge model, the basic instruction units of the manufacturing program are instantiated based on data extracted from the product models and integrated according to the relationships inferred by the reasoning mechanism, thereby generating the robotic manufacturing program.