Modular Approach to the Planning of the Robot’s Tasks in the Context of Holons and Graph-Based Methods

Abstract

The philosophy of “Industry 4.0” assumes that the contribution of the computer science in the modern industry will intensively grow. The manufacturing tasks are no longer just programs for digitally controlled machines, but are a part of the overall process that includes transport, machine-to-machine communication, and human-machine interfaces that allow the ongoing control of the whole process. Under such circumstances, describing the whole system – or just a part of it – needs to adjust the level of information detail. The key is to find the right level of detail that will guarantee a comprehensible description of the activity performed by the machine, but concomitantly will not contain any information that is redundant at the given level of generality. Robots are special kinds of machines, because apart from the typical handling and manipulation tasks, they also often perform the work that is the part of the machining or assembly process. This involves the higher level of complexity of the code written in the robot’s programming language, but such code is hard to understand for people without the ability of robots’ programming. The paper presents the other way of the description of tasks performed by robots during the manufacturing process that is the modular approach, based on classic, structural programming methods in connection with the holon-based and graph-based methods. Taking into account the dynamic development of modern ways of robots’ programming, the presented method may provide the starting point for further research in the context of task-level programming or programming by demonstration.