Cel The goal of SKILLS4ROBOTS is to develop a autonomous skill learning system that enables humanoid robots to acquire and improve a rich set of motor skills. This robot skill learning system will allow scaling of motor abilities up to fully anthropomorphic robots while overcoming the current limitations of skill learning systems to only few degrees of freedom. To achieve this goal, it will decompose complex motor skills into simpler elemental movements – called movement primitives – that serve as building blocks for the higher-level movement strategy and the resulting architecture will be able to address arbitrary, highly complex tasks – up to robot table tennis for a humanoid robot. Learned primitives will be superimposed, sequenced and blended. Four recent breakthroughs in the PI’s research will make this project possible due to successes on the representation of the parametric probabilistic representations of the elementary movements, on probabilistic imitation learning, on relative entropy policy search-based reinforcement learning and on the modular organization of the representation. These breakthroughs will allow create a general, autonomous skill learning system that can learn many different skills in the exact same framework without changing a single line of programmed code. To accomplish this goal, our skill learning system will autonomously extract the necessary movement primitives out of observed trajectories, learn to generalize these primitives to different situations and select, sequence or combine them such that complex behavior can be synthesized out of the primitive building blocks. We will evaluate our autonomous learning framework on a real humanoid robot platform with 60 degrees of freedom and show that it can learn a large variety of new skills. Dziedzina nauki natural sciencescomputer and information sciencesartificial intelligencemachine learningreinforcement learningengineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringroboticsautonomous robots Program(-y) H2020-EU.1.1. - EXCELLENT SCIENCE - European Research Council (ERC) Main Programme Temat(-y) ERC-StG-2014 - ERC Starting Grant Zaproszenie do składania wniosków ERC-2014-STG Zobacz inne projekty w ramach tego zaproszenia System finansowania ERC-STG - Starting Grant Instytucja przyjmująca TECHNISCHE UNIVERSITAT DARMSTADT Wkład UE netto € 1 405 572,50 Adres KAROLINENPLATZ 5 64289 Darmstadt Niemcy Zobacz na mapie Region Hessen Darmstadt Darmstadt, Kreisfreie Stadt Rodzaj działalności Higher or Secondary Education Establishments Linki Kontakt z organizacją Opens in new window Strona internetowa Opens in new window Uczestnictwo w unijnych programach w zakresie badań i innowacji Opens in new window sieć współpracy HORIZON Opens in new window Koszt całkowity € 1 405 572,50 Beneficjenci (1) Sortuj alfabetycznie Sortuj według wkładu UE netto Rozwiń wszystko Zwiń wszystko TECHNISCHE UNIVERSITAT DARMSTADT Niemcy Wkład UE netto € 1 405 572,50 Adres KAROLINENPLATZ 5 64289 Darmstadt Zobacz na mapie Region Hessen Darmstadt Darmstadt, Kreisfreie Stadt Rodzaj działalności Higher or Secondary Education Establishments Linki Kontakt z organizacją Opens in new window Strona internetowa Opens in new window Uczestnictwo w unijnych programach w zakresie badań i innowacji Opens in new window sieć współpracy HORIZON Opens in new window Koszt całkowity € 1 405 572,50