Pororobot: Child Tutoring Robot for English Education

The recent success of machine learning has lead to advancements in robot intelligence and human-robot interaction. It is reported that robots can well understand visual scene information and describe the scenes in language using computer vision and natural language processing methods. Image Question-Answering (QA) systems can be used for human-robot interaction. However, to achieve human-level artificial intelligence based on lifelong learning, a model must deal with real-world environments including dynamic, uncertain, and asynchronous properties based on lifelong learning. In this paper, we propose a prototype system for a video QA robot "Pororobot". The system uses the state-of-the-art machine learning system using dual memory model for implementing robot intelligence. In our scenario, a robot and a child plays a video QA game. Here we demonstrate preliminary results of our system. Keywords—videoQA; tutoring robot; human-robot interaction; robot intelligence; machine learning;