Finite-state machine control of mechatronic systems

Abstract

This paper describes the design of control concepts and algorithms for complex processes that are characterized more by the occurrence of discrete events than by differential equations representing the laws of physics. Such design is intended for process description in a symbolic, rather than numeric form. The goal is to combine concepts from both computer science and control, in order to develop a meaningful theory for controlling power electronics, process controllers, embedded systems and motion drive systems. Such design improves the firmware quality in a short development time. A designer would have only to specify the actions, events and transitions in terms of simple functions and tables. Generally, such design is highly structured and efficient, programming tasks are readily comprehended and fault diagnostics are easily included into the program structure. An application to the automatic sliding door illustrates the feasibility of this approach. The paper presents the modular finite-state machine, event–condition–action system, motion generation, motion control with load estimation and an example of a digital signal processor system. The limitations and attributes of each technique are discussed, and a state table format is presented with the capability of representing parallel asynchronous sequential processes.