Home automation projects have been developed for some time, having evolved into the so-called smart environments. These environments are characterized by the presence of sets of sensors and actuators connected according to the concept of Internet of Things (IoT), in order to respond appropriately and proactively to different situations. The integration of intelligent environments with robots produces the concept of robotic ecologies, capable of performing tasks with greater flexibility and less mechanical complexity than traditional monolithic robots. This project aims to contribute to the construction of an architecture for robotics in intelligent environments, under development by the Robots Learning Lab (LAR) of ICMC-USP. Focus will be given to the evolution of machine learning techniques, mainly for human activity recognition. Different configurations will be considered, in order to produce an adaptable module. An application scenario will be implemented in a physical environment, so that the developed module could be evaluated in practical situations, including in terms of human-robot interaction. It is expected to evolve the state of the art both in the application of learning techniques and in the constructive aspects of architectures for robotics in intelligent environments.
News published in Agência FAPESP Newsletter about the scholarship: