Smart City Framework

The idea of using autonomous driving and smart cities as environments for intelligent robotics teaching is becoming quite popular. In the case of Robobo, we have developed of a full-scale model of a neighbourhood in a smart city, and its subsequent implementation in different simulation software. In this framework, students can face many challenges dealing with autonomous vehicles, like perception, control, computer vision, localization or path planning.

The model is a rectangular city layout of 3.5m x 4m. It is made up of an external two-way road, surrounding a central part that contains a roundabout, where four two-way road sections intersect. In order to build this model on your own, here are two links to basic elements:

• Plan of the model lines: https://github.com/mintforpeople/robobo-programming/wiki/Plane.dwg

• A zipped folder with the stl files to 3D print the signs: https://github.com/mintforpeople/robobo-programming/wiki/3dmodels.7z

SmartCity, simulation models

Three different models have been developed according to the three skill levels considered in the Robobo Project:

• The RoboboSim and intermediate (Python) level programmers.
• CoppeliaSim for intermediate (Python) level programmers.
• Gazebo for advanced (ROS) programmers.

Associated libraries

Some specific Python libraries have been developed to support the use of this framework at classes:

• Object recognition
• Lane detection
• Aruco
• Streaming