Abstract:

I - Introduction
This paper presents a pedagogical project implemented at the University of Paris Descartes. The main idea is to involve undergraduate students in interdisciplinary projects using an attracive medium in order to increase their interest in computer science and to help them to think about their curriculum.
Computer science like most other scientific domains, suffers from young people's lack of motivation to study science. As a consequence, it is important to find, and elaborate pedagogical schemes that will be more attractive. A project-based approach can appear more challenging to them.
Robots have already been used in the learning process of artificial intelligence. For example : the MIT Robot Design Project in which students approached robotic basis through the construction of their own robot and they had to compete with each other's robots at the end.

II - Setup
During their studies, undergraduate students are given the opportunity to make a computer science project. Second and third year-students carry out their project during the second semester by groups. Each group is supervised weekly by a member of the pedagogical team to have a regular following up.
One of the objectives of the project is to let the students move away from the usual practical lessons and have the opportunity to develop applications within a team. This experience made them aware of the different stages in implementing a project by using a methodological approach often used in their future professional environment.
The students were asked to choose their project from a given list. Although all students are undergraduates, we had two different populations with different criterion according to the year of study. Projects involving robots were the most popular among students.

Students who managed to complete the semester were given a questionnaire focused on their feelings about many aspects of their project : achievements and difficulties as well as teamwork efficiency, etc. Two different populations were observed: those who participated in a robotics project and the others. These two populations are equivalent in term of prior academic experience; approximatively half of the students surveyed had already participated in this kind of project.

III - Observations
Students involved in robotics projects expressed the feeling that they had conducted their project better than the other students in terms of planning, achieving their objectives and teamwork efficiency. These feelings appear to be consistent with the free comments left by students on the questionnaires and confirms our observations that students of robotics projects have worked more to complete their objectives.

These observations corroborate those previously obtained in similar experiments of project-based learning. A greater investment of students involved in robotics projects may however be noted. Thus, using robots in project-based learning seems worthwhile within this kind of academic context.

IV – Conclusion
This experiment has been very encouraging. We were surprised to see the emotionnal implication of students who personalized and named their robots.
At this stage, it is not possible to rigourously evaluate the results of our projects. Indeed, it seems difficult to know who learnt what, when and how. Nevertheless, we can observe the progress achieved in various domains such as the employed methodology, students' motivation, …

Keywords:

Project-based learning, computer science, student's motivation.