LRIMa City: A Fog-Computing-Based Smart City with Integrated Self-Driving Cars

The LRIMa city is a continuous, expandable and polyvalent project focused on IoT and IA solutions for research and learning. It serves as a testbed for exploring IoT architectures, including cloud centralized and fog on-device computation. Additionally, the city enables the development of AI models for autonomous vehicle navigation. Our smart city implementation encompasses various embedded systems such as smart motorized vehicles, a speed radar, smart parking systems, adaptive streetlights, and a remotely controlled bridge. To facilitate autonomous driving, we have created a Convolutional Neural Network (CNN) based on NVIDIA's model for predicting steering angles from input images. Moreover, we have developed a YOLOv8-based traffic light detection model and a cascade classifier for stop sign detection. For other components, we have employed diverse AI solutions, ranging from license plate detection and identification to hand gesture recognition with Mediapipe. We hope that our city will serve as a valuable resource for researchers and newcomers to explore and develop innovative IoT and AI solutions, promoting experimentation and advancements in these fields.