Adaptive Modular Architecture for Rich Motor Skills
Bielefeld University, EPFL, Graz University of Technology, IRCCS Fondazione S. Lucia, Ghent University, University of Tubingen, University of Zurich, IIT, Jacobs University Bremen, Weizmann Institute.
Compared to animals and humans, the motor skills of today’s robots still must be qualified as poor. Their behavioral repertoire is typically limited to a narrow set of carefully engineered motor patterns that operate a rigid mechanics and lack situated adaptivity, learnability and dynamical fusion of motor primitives into complex, task-oriented behavioral patterns.The AMARSi Integrated Project aims at a qualitative jump toward biological richness of robotic motor skills. By richness we mean the systemic integration of motor primitives into a large repertoire of motor behaviors, ranging across the entire hierarchy from simple periodic and aperiodic motions to complex, task-oriented interaction sequences between a robot and a human caretaker. To achieve this goal, a number of innovative scientific concepts and interdisciplinary research methods will be implemented:
Progress will become manifest in a series of robotic demonstrations of increasing richness, based on (and further developing toward compliant mechanics) the human-infant-like iCub robot as shown in the upper right and the compliant quadruped Cheetah platform, a prototype is depicted in the lower left. In a final scenario, the robots will engage in an interaction with a human caretaker at the level of a young child playing an open-ended ball game with a parent in a cluttered and rough environment.
Hardware and software solutions will be made publicly available following an open source policy.
The consortium brings together internationally leading groups from the fields of biorobotics, robot engineering, compliant mechanics, morphological computing, human motor research & biomechanics,
theoretical biology, machine learning, neural networks & reservoir computing. The consortium can boast dense and long-standing collaboration ties.
Acquiring rich motor skills will change the role of robots in our human’s society in two fundamental ways. First – obviously – such robots will be much more versatile than today, with greatly expanded ranges of practical usages. And second – less obviously but even more importantly – the naturalness
and compliance of their motor behavior will make them blend into the everyday routines of human society, physically safe and psychologically acceptable.
Project budget 9.208.448,0000
Duration 01/03/2010-28/02/2010
For further information please contact
Darwin Caldwell (ADVR Department)
Bielefeld University, EPFL, Graz University of Technology, IRCCS Fondazione S. Lucia, Ghent University, University of Tubingen, University of Zurich, IIT, Jacobs University Bremen, Weizmann Institute.
Compared to animals and humans, the motor skills of today’s robots still must be qualified as poor. Their behavioral repertoire is typically limited to a narrow set of carefully engineered motor patterns that operate a rigid mechanics and lack situated adaptivity, learnability and dynamical fusion of motor primitives into complex, task-oriented behavioral patterns.The AMARSi Integrated Project aims at a qualitative jump toward biological richness of robotic motor skills. By richness we mean the systemic integration of motor primitives into a large repertoire of motor behaviors, ranging across the entire hierarchy from simple periodic and aperiodic motions to complex, task-oriented interaction sequences between a robot and a human caretaker. To achieve this goal, a number of innovative scientific concepts and interdisciplinary research methods will be implemented:
- Coordinated and simultaneous development of compliant mechanics, pervasive learning and dynamical-systems based control architectures, centered on the concept of adaptive modules.
- Mutually informing research in human motor behavior and robotics.
- Reliance on compliant mechanics and morphological computing for flexibility, computational and motoric speed, safety and damage-robust learning.
- Novel learning paradigms (unsupervised, reinforcement and imitation) drawing from principles of reservoir computing.
- Control architectures based on dynamical (neural) systems throughout, also on the higher cognitive levels.
- A unified framework for locomotion and manipulation behaviors.
Progress will become manifest in a series of robotic demonstrations of increasing richness, based on (and further developing toward compliant mechanics) the human-infant-like iCub robot as shown in the upper right and the compliant quadruped Cheetah platform, a prototype is depicted in the lower left. In a final scenario, the robots will engage in an interaction with a human caretaker at the level of a young child playing an open-ended ball game with a parent in a cluttered and rough environment.
Hardware and software solutions will be made publicly available following an open source policy.
The consortium brings together internationally leading groups from the fields of biorobotics, robot engineering, compliant mechanics, morphological computing, human motor research & biomechanics,
theoretical biology, machine learning, neural networks & reservoir computing. The consortium can boast dense and long-standing collaboration ties.
Acquiring rich motor skills will change the role of robots in our human’s society in two fundamental ways. First – obviously – such robots will be much more versatile than today, with greatly expanded ranges of practical usages. And second – less obviously but even more importantly – the naturalness
and compliance of their motor behavior will make them blend into the everyday routines of human society, physically safe and psychologically acceptable.
Project budget 9.208.448,0000
Duration 01/03/2010-28/02/2010
For further information please contact
Darwin Caldwell (ADVR Department)
