Robots equipped with cameras and sensors that drive autonomously have become a reality. Meet the Duckiebots – the result of research conducted by Asst. Prof. Dr. Octavian Machidon from the Faculty of Computer and Information Science (FRI).

He recently led a demonstration project in which he tested autonomous robots equipped with a full spectrum of cutting-edge technologies, including computer vision, sensors, artificial intelligence, machine learning, and robotics. As part of the project, Dr. Machidon presented the so-called Duckietown at FRI – an innovative ecosystem of robotics and artificial intelligence designed for both educational and research purposes.

At the center of Duckietown are autonomous vehicles – Duckiebots – equipped with an NVIDIA Jetson Nano mini-computer, cameras, and sensors. The Duckiebots navigate a model city composed of roads, obstacles, traffic signals, and intersections. The Duckietown platform serves as a valuable resource for researchers and students.

Dr. Machidon plans to use the platform as a foundation for further development of context-aware and efficient deep learning on mobile devices. With this project, he builds upon his previous research, which explored the use of drones in smart agriculture as part of the European project AgriAdapt.

Duckietown provides students with a hands-on environment for learning about robot autonomy – how to enable robots to make independent decisions – and for experimenting with various technologies for self-driving cars, such as lane detection and tracking, vehicle recognition, obstacle avoidance, traffic sign recognition, and path planning. Through this innovative initiative, researchers and students can actively engage in practical applications and empirical research in the field of autonomous transportation.