Desenvolvimento orientado a modelos no domínio de robótica: uma revisão sistemática da literatura

The field of robotics has been applied in various contexts, such as the industrial, health and education. The robotic projects involve various fields of study such as computer vision, artificial intelligence, psychology, biology, among others. The Academic competitions have encouraged the construction of robots that explore environments, play soccer and perform tasks of various types. These robots are agents made up of multiple sensors and actuators working along with software that meets specific requirements, and the system responsible for the management of all components. In this sense, there is a subset of robots known as social robots that have the ability to interact among themselves or with humans. These in turn are able to recognize natural language through speech or writing, interpreting gestures and interact in social and affective way. However, the increased complexity of robots reflects similarly in more complex control software, leaving the task of development more challenging. Thus, researchers have pointed to the model-driven development as an alternative to assist in the reduction of complexity of software development in the field of robotics. The model-driven development is a promising paradigm that uses models as first order artifacts and seeks to promote the reuse of software components and fast code generation with quality, thus reducing the cost of development and effort. Therefore, this research performs an analysis of how the model-driven development has supported the field of robotics, pointing the available artifacts and semi or automatically generated contributions, techniques involved, the functional and non-functional requirements, paradigms involved in robot behavior and service for social issues. The data was extracted from 86 studies writing a systematic literature review in order to assist researchers in the basement for realization of new research activities.