Rigby

Rigby is a system comprised primarily by a cloud platform and a robotic companion/assistant. It is aimed at primary and secondary school students that struggle in their home environment with their learning, homework, exams and other school activities.

The system will try to help the kid to fulfill all their activities by reminding him of their homework and due dates, gathering at the same time data about the kid's study habits. Along with data provided by the teacher and parents; with the help of behavioral psychology and AI techniques it will try to identify behavior patterns and recommend strategies to the teacher, parents and the robotic companion, in order to help the child improve his performance.

The cloud platform

Provides the teacher with a web based interface to fill in data about the subjects she is teaching in class; lessons, homework, activities, due dates and details about grading rubrics, along with suggested bibliography and references. At the same time the teacher can fill information and evaluations about the kid's performance in class, along with recommendations for him and their parents. All the information the teacher provides is systematically acquired, in order to apply behavioral psychology and AI techniques to learn about the struggles and needs of the child; to further assess him, the parents and the teacher.

The cloud platform also provides access for the parents in order to reach information the teacher provides, the robotic companions gathers, and the cloud app generates using AI.

The robotic companion

The robotic companion is a sociable robot with a GUI and voice controlled interface. It talks to the kid in an amenable manner about their lessons, homework, due dates and general stuff he has to do; in order to fully take advantage of their lessons and learning. By applying behavioral psychology techniques, the companion also gathers critical information about the kid's struggles: what topics he didn't understand well, why and under what circumstances he tends to procrastinate, what days and hours he favors for doing (or not) homework, how much time does he spend on leisure activities in comparison to doing homework, and why, etc.

The premise here is that, under some circumstances kids will bond with a robotic companion more easily than with real adult people (parents, teachers, etc.) So, in the end, the system is a bridge between the kid and parents/teachers; that helps the latter know and understand better the learning habits of the child, by accessing information that normally the child naturally refuses to share with older people.

Scope of the project's first phase

In this first phase (until October, 29), a first robotic companion prototype will be implemented using Android Things with a Raspberry Pi 3 and a cloud based interface for teacher and parents where the teacher will provide data about subjects, homework, due dates and grades. The robotic companion will download this data in order to generate a "conversation strategy" for every day to assess the child and gather relevant data.

For this first phase, a prototype of an AI app will be developed to learn specifically the days and hours preferred for the child to his homework. If training of the AI is needed, artificially generated data will be used to test the AI prototype.