Adaptive Supervisor Synthesis for Automated Manufacturing Systems With Multiple Unreliable Resources
针对自动化制造系统中多个资源可能随时失效的问题，提出一种自适应监督控制方法，通过Petri网转换和虹吸分析，设计能自动调整工作模式的控制器，确保系统在资源故障时仍能连续运行。
Deadlock and blocking issues have received more and more attention in automated manufacturing systems (AMSs). This article studies an adaptive supervisory control problem for AMSs with multiple unreliable resources which may fail at any time. By using general Petri nets (PNs), each processing stage in the AMSs under consideration allows to utilize multiple types and quantities of resources. Based on a transformation technique, a general net is first converted to an ordinary net. In an ordinary net, a set of formulas is used to check the undermarked siphons, each of which is designed as a control place to prohibit it from being undermarked. By analyzing the relationship between the unreliable resources and the detected siphon, an adaptive controller is designed which can automatically adjust working modes such that the undermarked siphon caused by resource failures is marked. Finally, an adaptive supervisory controller is iteratively synthesized for AMSs with unreliable resources to ensure the controlled system can execute continuous operations even if unreliable resources malfunction. A representative example is given to illustrate the effectiveness of our approach.