Hello everyone,
My name is Morgan Jousset. I am currently studying Mechanical Engineering at SIGMA Clermont in France, where I am in my second year of engineering studies, equivalent to a Master’s level or four years of higher education. My specialization is structures and materials, where I have developed skills in mechanical engineering, structural analysis, and material science.
I am currently doing my internship at the University of Agder in Grimstad, Norway, in collaboration with NORCE. I am applying my skills in artificial intelligence and machine learning to my internship project with the CARM Centre. My current work focuses on using AI techniques for engineering applications, particularly for the analysis and detection of objects using 3D data.
About Me
Outside of my studies, I enjoy outdoor sports, drawing, and artistic activities. These activities allow me to maintain a good balance between my studies and my personal life. Outdoor sports help me stay active, motivated, and constantly push myself beyond my limits. Drawing and artistic activities give me a more creative way to express myself and allow me to relax and improve my well-being. Overall, these activities bring me motivation, creativity, discipline, and a sense of personal achievement, which are also qualities that I try to apply to my studies and professional life.
My work at CARM
During my internship, I am working on the development of an AI-based approach for the detection and analysis of printed circuit boards (PCBs).
The main goal of my project is to explore how artificial intelligence, 3D data, and robotics can be combined to improve the identification and handling of electronic components meant for recycling. My work involves collecting and processing data, preparing datasets, annotating PCBs, and developing machine learning models capable of recognizing objects from 3D scans.
This project is a great opportunity to connect the knowledge I have gained during my engineering studies with new technologies such as machine learning, computer vision and robotics. One of the main challenges is making the AI models reliable and accurate enough to work with real-world data, where the position, shape, and appearance of objects can vary.
Learning Through Research and Collaboration
Working within the CARM environment has also allowed me to discover a multidisciplinary research environment, where engineering, artificial intelligence, robotics, and sustainability come together. Throughout the project, I have gained practical experience with 3D scanning, point cloud processing, Python, machine learning, and AI-based object detection. I have also learned how important data preparation and experimentation are when developing an AI system for a real engineering application.
What I find particularly interesting is the connection between technology and sustainability. By improving the automated identification and sorting of electronic components, AI and robotics could contribute to more efficient recycling and recovery of critical raw materials. This internship has therefore given me the opportunity not only to apply what I have learned during my engineering studies, but also to discover new technologies and approaches that could play an important role in the circular economy and the transition towards more sustainable resource management.
A Glimpse of My Work
The first image shows a 3D scan of a PCB, which represents the raw data collected during the acquisition process. This point cloud provides the 3D information that is then processed and used as input for the AI model.
The second image shows the final detection result, where the trained model has identified the PCB and other relevant objects by placing bounding boxes around them. It illustrates the final step of the process, transforming raw 3D data into meaningful information through AI-based object detection.
Together, these two images give a glimpse of the workflow behind my project: from 3D data acquisition to automated object detection.
Looking Ahead
As I move forward with my internship, I will continue working on improving the accuracy and robustness of the AI model. A key objective will be to collect and annotate more data, allowing the model to better handle the variety of PCBs and different scanning conditions.
Beyond the technical aspects, I look forward to continuing to learn and to gaining a better understanding of how AI and robotics can be applied to real-world sustainability challenges.