As technology develops, much of the industrial work has been replaced or simplified by computational devices (e.g. robots), and automation has become of great importance. Computer programming languages, as human and computer interaction tools, are usually difficult to understand by non-programmers. Thus, the most difficult part is the communication between the developers and software users in the software development process. In order to solve this problem, domain specific languages (DSL) were created. This thesis first describes the classification of modern DSL and the role of syntax and semantics of a language. The Backus-Naur form (BNF) is explained as an important formalism to define the syntax of a computer language. Two programming languages, which are well-known in automation technology, are described as successful examples of early DSL, G-code for controlling of manufacturing tools and a modern language for robot control. Finally the development of a specific DSL for processing finite state machines is presented. The finite state machine is an important pattern to implement sequential behaviour for industrial machinery. This is done with two tools, at first with ANTLR (Another tool for language recognition), then with PYPARSING as parser generators.