DSLs for Modelling, Coordinating and Programming Multi-Robot Systems

Software development for robotics applications is still a major challenge that becomes even more complex when considering a Multi-Robot System (MRS). Such a distributed software has to perform multiple cooperating tasks well-coordinatedly to avoid unsatisfactory emerging behavior. This thesis presents a high-level programming and modeling approach for MRSs using X-Klaim programming language and the Business Process Modeling No- tation (BPMN). The computation and communication model of X-Klaim, based on multiple distributed tuple spaces, permits coordinating with the same abstractions and mechanisms both intra- and inter-robot interactions of an MRS. This allows developers to focus on the MRS behavior, achiev- ing readable, reusable, and maintainable code. Alongside this, BPMN offers an effective mechanism for representing the high-level design of MRS mis- sions. Its graphical nature aids in depicting complex mission sequences and interactions in a way that is easily comprehensible, promoting clarity and interoperability. Our approach consolidates these two paradigms, enabling a smooth transition from the conceptual design to concrete implementation of MRS missions. We achieve this by providing a systematic mapping of BPMN elements to X-Klaim constructs, which preserves the logic and structure of the original design while adding necessary detail for execution in Robot Op- erating System (ROS) environment. ROS is a flexible framework for writing robot software and provides the tools, libraries and conventions to simplify the process. To validate the feasibility and effectiveness of our approach, we implemented and tested it across different MRS scenarios. The results showed that our approach scales effectively when applied to increasingly com- plex scenarios, allowing for code reusability. Further, our solution introduces a slightly higher but acceptable latency compared to traditional ROS im- plementations based on Python code, and it consumes less memory. This research contributes towards improving the efficiency and quality of MRS software development, bringing together the power of BPMN, X-Klaim and ROS in a synergistic and integrated design-to-implementation process. Our findings offer promising implications for future development of more complex, scalable, and effective MRS missions.