Abstract:

Traditional teaching methodologies often fail to capture the attention and interest of the new generation of students. In a rapidly evolving world characterized by technological advancements and diverse learning preferences, students require different stimuli to encourage and enhance their learning experiences. To effectively engage and motivate these learners, it is crucial to adopt innovative approaches that incorporate technology, real-world applications, collaborative projects, and personalized learning experiences.

Our Educational Innovation Project titled "DIRASEI - Design and Implementation of new tools and Resources for the application of the Challenge-Based Learning methodology to the teaching and learning of Intelligent Electronic Systems Design" aims to leverage the potential of challenge-based learning to teach AI skills in electronics laboratory subjects. Specifically, this abstract focuses on the development of new artificial intelligence-based resources to recognize human activities that could enhance the student motivation through a challenge-based learning methodology. The proposed resources encompass the entire pipeline of a project, offering comprehensive modules to facilitate seamless project progress. Each system integrates various stages, including data acquisition, signal processing, training and classification, and deployment. By addressing each step of the project lifecycle, the proposed resources provide a robust framework for students to gain hands-on experience in every aspect of their learning journey.
In particular, the intelligent electronic systems proposed in this work are focused on Human Activity Recognition. One of those is based on inertial sensors to extract acceleration signals and recognize physical activities that the students perform. Another system is focused on recognizing the hand pose of the student using images. The initial proposals use state-of-the-art deep learning approaches such as Convolutional Neural Networks or Long short-term memory Networks to model the human activities. However, both resources are open platform designed to record and model different activities using other deep learning approaches, which could encourage the students to create their own custom system while learning. These systems have been developed using Tensorflow and Keras libraries in Python, with the possibility to deploy them over Google Colab or a Raspberry Pi platform.
The resources proposed in this work have broad applicability across a range of bachelor's and master's programs related to computer science, telecommunications, electronic engineering, and artificial intelligence. The interdisciplinary nature of the resources allows students from various disciplines to benefit from the innovative tools and methodologies presented, from managing new sensors to proposing new signal processing techniques or new artificial intelligence models to classify the activities.

The Educational Innovation Project "DIRASEI" not only enhances the teaching and learning experience in AI but also contributes to the broader goals of emphasizing new trends and experiences in education and research. Through our hands-on approach and practical implementation, we strive to prepare students for the demands of the ever-evolving AI landscape and cultivate a generation of curious learners and innovators.

Keywords:

Deep learning, human activity recognition, challenge-based learning methodology, student motivation.