Duckiebot Autonomy Stack
Educational autonomy stack for a miniature vehicle: ROS2, perception, lane following, and basic planning. Built for student projects.
Mobile Robotics & Computer Vision
Researcher and lecturer building reliable autonomous systems and teaching modern computer science.
About
I work at the intersection of robotics, computer vision, and machine learning, with a strong interest in making autonomous systems practical and accessible. For me, robotics is not only about algorithms but about creating systems that can sense, reason, and act in the real world. This passion drives both my research (autonomous driving, neural networks on edge devices) and my teaching - where I help students move from theory to working prototypes. I enjoy building tools and frameworks that make complex technologies approachable: from robotics simulators and benchmarking utilities to course materials that turn programming into hands-on discovery.
Projects
TPU vs GPU/CPU Benchmark
Cross‑device inference benchmark with a lightweight GUI; measures latency and throughput across accelerators.
GitHub classroom to Google Spreadsheets integration
Github action that allow automatic push of students' results in GitHub calssroom to Google Spreadsheet document
Hardware simulator
Utility to emulate robotics hardware. It can be usefull for debugging robotics SDKs. Powered by Gazebo Simulator
The project is still under development, link will be available soon
Publications
- (Poster) Efficient Lane Marking Recognition for Duckietown for Varying Lighting Conditions — 2025 21st International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT). PDF
Teaching
- Programming Fundamentals with C & Linux
- Robotics & Computer Vision
- Operating Systems