Speaker: Kevin Sebastian Luck
Robotics Seminar Series. Fourth Session – 17th September 2021, 14:00-15:00, via zoom (use this link): https://aalto.zoom.us/j/62124942899
Abstract: While much progress was made in recent years in both the evolutionary robotics and the deep learning community regarding the adaptation of robot morphology, no strong emphasis was and is placed on the data-efficiency of the developed approaches. Newly developed methodologies are primarily evaluated in simulation and make use of the current abundance of computational resources and mass-parallelization of simulation. This limits the applicability of these technologies in regards to the co-adaptation of robots in the real world.
I will present in this talk current and planned research efforts for the development of data-efficient learning strategies for the co-adaptation of robot behaviour and morphology. The central goal is to develop methodologies suitable for the co-adaptation of robots in the real world which do not require prior knowledge or simulators.