Balancio-Kit: An educational self balancing robot

In this paper Balancio-Kit is presented, a low-cost self balancing robot, targeted to robotics, systems control, and machine learning students and enthusiasts, at different knowledge levels. The platform comes with various additions, such as CAD files; a Bluetooth application to command the robot; a simulation, where the user can evaluate different control algorithms or train reinforcement learning agents; and a tested workflow to implement neural networks inside a low-cost microprocessor. It is designed to be low-cost and easy to assemble, simple to flash software and to be upgradable and remixable. Implemented using well-known software like Arduino IDE and open source tools like Pybullet and Open AI Gym.