A robotool plug-in for roboworld

• Ano de defesa: 2023
• Autor(a) principal: RODRIGUES JÚNIOR, Francisco Wilson
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Pernambuco; UFPE; Brasil; Programa de Pos Graduacao em Ciencia da Computacao
• Palavras-chave em Português: Engenharia de software; Robótica; Linguagem natural controlada
• Link de acesso: https://repositorio.ufpe.br/handle/123456789/53689

Developing robotic systems is a challenging task due to the inherently underlying complexity. Additionally, the lack of customised techniques and tools means that the current practice of software engineering for robotics is outdated. Therefore, model-driven software engineering, as opposed to simulation- and code-centric approaches, has been advocated for robotics. Considering this, the RoboStar framework, via its associated tool (RoboTool), provides a collection of domain-specific languages tailored for different aspects of the development of robotic systems. One of these languages is RoboWorld: a controlled natural language (CNL) for specifying operational requirements. In this work, we develop a RoboTool plug-in dealing with RoboWorld. This plug-in has a graphical user interface that enables the customisation of project-specific dictionaries. Moreover, it provides surface and structural editors of RoboWorld documents. Integration with underlying natural-language processing techniques and tools is transparent, and, thus, hidden from the end user. From an intermediate representation of RoboWorld documents, which is derived automatically, the plug-in also has automatic support for checking well-formedness conditions, and generating a formal CyPhyCircus semantics. The RoboTool plug-in for RoboWorld was validated considering three case studies: a rescue drone, a ranger robot, and a foraging robot.