The abilities to self-erect and maneuver around moving obstacles are crucial steps towards the deployment of fully autonomous mobile robots acting in uncertain environments. With this perspective in mind, learning algorithms for robotics must be tested on real-world systems, as hardware raises a multitude of additional challenges. In the field of machine learning and control, the preferred testbed for algorithms are either rotorcrafts (quadrocopters) or stationary systems (robot arms, pendulums). While research on such testbeds led to breakthroughs in learning control and motion planning, analyzing learning algorithms for naturally unstable non-holonomic systems has been rarely investigated. With representatives such as motorcycles and airplanes - the development of a simple small-scale non-holonomic testbed is of great significance for research on learning control and finally feasible due to steady advances in technology.
In this project, a first prototype of a highly maneuverable robot that utilizes state-of-the-art hardware shall be developed. Core aspects of this project are the software development and hardware configuration. While most of the hardware components (Micro-controllers, Sensors, Brush-less DC Motors) are provided, open questions on the software architecture and system design must be answered. Optionally, the student is encouraged to transfer existing controllers from simulation to the real system. Different project types are available (Hiwi position, Bachelor, Master) and can be discussed based on the interest of the candidate.