Hands-on Workshops
Learn practical skills directly from AI experts
Responsible AI for Sustainable Development in Low-Resource and Multilingual Societies
📝 What You'll Learn:
The workshop on Responsible AI for Sustainable Development in Low-Resource and Multilingual Contexts will convene researchers, practitioners, policymakers, industry partners, and community stakeholders to examine how AI can be designed, evaluated, and deployed responsibly in settings where linguistic resources, data, compute, and infrastructure are limited. The workshop will focus on AI applications that advance sustainable development across areas such as education, health, agriculture, climate resilience, public services, accessibility, and cultural heritage. It will give particular attention to multilingual and culturally grounded AI for Arabic, African languages, French, and other relevant languages and varieties, while addressing key issues in fairness, privacy, transparency, accountability, community participation, and environmental sustainability. By bringing together technical, policy, and community perspectives, the workshop aims to support AI systems that are not only effective, but also inclusive, locally relevant, sustainable, and useful to the communities they are meant to serve.
👨🏫 Presenters:
Prof. Muhammad Abdul-Mageed and Dr. Abdellah Elmekki, University of British Columbia (UBC), Canada.
Embedded AI for Object Detection in Autonomous Drones
📝 What You'll Learn:
This workshop explores the integration of artificial intelligence at the edge for real-time object detection aboard autonomous drone systems. Participants will gain hands-on experience building a complete pipeline that combines the Pixhawk flight controller, a robust, open-source autopilot platform, with a Raspberry Pi serving as the onboard edge computing unit for AI inference and video acquisition. The workshop covers key topics including camera interfacing, video streaming optimization for low-latency capture, and deploying lightweight deep learning models such as YOLOv8-nano or MobileNet SSD, optimized through quantization and pruning techniques to meet the computational constraints of embedded hardware. Attendees will learn how to establish MAVLink communication between the Raspberry Pi and Pixhawk, enabling AI-driven decisions, such as object tracking or obstacle avoidance, to influence flight behavior in real time. The session also addresses power consumption trade-offs, thermal management, and model performance benchmarking on ARM-based processors. By the end of the workshop, participants will have assembled and tested a functioning drone perception module capable of detecting and classifying objects from live aerial footage. This workshop is designed for engineers, researchers, and enthusiasts with a background in embedded systems, robotics, or computer vision who wish to bridge the gap between AI research and real-world UAV deployment.
👨🏫 Presenters:
Prof. Eng. Mohamed Ould-Elhassen AOUEILEYINE, Innov'COM Laboratory, SUPCOM, University of Carthage