Process Oriented Ontology Based Robotic Task Planning And Modelling

This PhD tackles the significant obstacles hindering Small and Medium-sized Enterprises (SMEs) from effectively integrating robotic solutions into their agile manufacturing processes. The core challenges identified include the lack of robotic expertise, comprehensive knowledge, and high programming costs. To mitigate these issues, the primary contribution of this PhD is the creation of RTMN (Robot Task Modelling Notation), an intuitive modelling language designed for planning and controlling robotic processes. This language is further enhanced by an ontology, ORPP (Ontology for Robotic Task Planning), which provides structured knowledge. This combined approach empowers individuals without specialized robotics knowledge to flexibly program and reprogram robots, ultimately facilitating mass customization through readily accessible relevant information. Building upon the foundation of RTMN, subsequent research led to RTMN 2.0, which expands the initial framework to specifically address the complexities of human-robot collaboration, incorporating safety considerations and enabling traceability between business processes and low-level robotic control. The integration of RTMN 2.0 with the ORPP ontology culminates in a comprehensive task planning system featuring a user-friendly graphical interface based on RTMN principles, while ORPP operates in the background as a knowledge base, enabling intelligent querying and reasoning. Experimental validation through several demonstrations in the context of the H2020 ACROBA project confirms the practical application and effectiveness of the outcomes of this PhD in modelling robotic processes, intelligent task planning, and human-robot collaboration scenarios.

keywords: