Learning in Robotics

Learning is the gathering of knowledge as well as of mental and physical abilities. Learning can be seen as a systematic changing of behaviour due to attained and elaborated information by examining changes in environment.

To enable autonomous robots to operate in unknown environments, they have to be able to adapt their behaviour their own – they have to learn.

So learning and adaptation are important paradigms in the field of current robotics research. Especially when a system cannot be completely modelled, data-driven methods enhance model-based approaches. We use different approaches to adaptation. Many yet not all of them are based on neural networks, often being inspired by solutions found in biological systems.

Driven by the problems that we encounter in robotics, we continue to develop learning and bio-inspired approaches to solve those tasks at hand.