Drittmittelprojekt | beendet | 01.02.2014
- 31.01.2018
COgnitive Development for Friendly RObots and Rehabilitation (CODEFROR)
Projektleitung
Koordinierende Einrichtung
Zweck
Forschung
Förderung
Mittelgeber:
Europäische Union
Förderprogramm:
Drittmittel EU und sonst. internat. Organisationen
The objective of the joint exchange project is to investigate aspects of human cognitive development with the double goal of developing robots able to interact with humans in a friendly way and of designing and testing protocols and devices for sensory and motor rehabilitation of disabled children. The methodology we intend to follow will combine science driven investigation of human cognitive development and engineering based implementation of devices and protocols.The intended focus is on social interaction and how the knowledge of this aspect of development could lead to robots able to communicate with humans in a natural and “biological way” (friendly robots), and/or give rise to training and rehabilitation techniques for children with sensory, motor and cognitive disabilities.Social interaction is a bidirectional process based on a shared representation of actions and on mutual understanding and its study will help discovering how infants develop the understanding of actions, intentions and emotions to progressively improve their social behaviours. In addition, implementing models derived from humans studies on robots provides an additional constructive approach to investigate cognitive developments and could benefit both robotics (better robots) and neuroscience, providing a test-bed for the proposed theories.To be successful this multidisciplinary program calls for a wide range of expertise both in terms of scientific communities (developmental psychology, robotics, sensory and motor rehabilitation), and in relation to engineering implementation (robots as well rehabilitation devices) and social exploitation (sensory and motor rehabilitation). The exchange program proposed has the goal of joining the forces and expertises of the participating partners and of helping the formation and establishment of an international community of young researchers that shall effectively bridge the involved groups and their expertise in order to be effective in the long term.
- Queißer J, Steil JJ. Bootstrapping of parameterized skills through hybrid optimization in task and policy spaces. Frontiers in Robotics and AI. 2018;5:49.
- Queißer J, Ishihara H, Hammer B, Steil JJ, Asada M. Skill Memories for Parameterized Dynamic Action Primitives on the Pneumatically Driven Humanoid Robot Child Affetto. Presented at the International Conference on Development and Learning and on Epigenetic Robotics 2018 (ICDL-EPIROB2018), Tokyo .
- Schulz A, Queißer J, Ishihara H, Asada M. Transfer Learning of Complex Motor Skills on the Humanoid Robot Affetto. Presented at the International Conference on Development and Learning and on Epigenetic Robotics 2018 (ICDL-EPIROB2018), Tokyo (In Press).
- Malekzadeh M, Queißer J, Steil JJ. Imitation learning for a continuum trunk robot. In: Verleysen M, ed. Proceedings of the 25. European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. ESANN 2017. Louvain-la-Neuve: Ciaco; 2017.
- Malekzadeh M, Queißer J, Steil JJ. Learning the end-effector pose from demonstration for the Bionic Handling Assistant robot. Presented at the 9th Int. Workshop on Human-Friendly Robotics, Genoa.
The objective of the joint exchange project is to investigate aspects of human cognitive development with the double goal of developing robots able to interact with humans in a friendly way and of designing and testing protocols and devices for sensory and motor rehabilitation of disabled children. The methodology we intend to follow will combine science driven investigation of human cognitive development and engineering based implementation of devices and protocols.The intended focus is on social interaction and how the knowledge of this aspect of development could lead to robots able to communicate with humans in a natural and “biological way” (friendly robots), and/or give rise to training and rehabilitation techniques for children with sensory, motor and cognitive disabilities.Social interaction is a bidirectional process based on a shared representation of actions and on mutual understanding and its study will help discovering how infants develop the understanding of actions, intentions and emotions to progressively improve their social behaviours. In addition, implementing models derived from humans studies on robots provides an additional constructive approach to investigate cognitive developments and could benefit both robotics (better robots) and neuroscience, providing a test-bed for the proposed theories.To be successful this multidisciplinary program calls for a wide range of expertise both in terms of scientific communities (developmental psychology, robotics, sensory and motor rehabilitation), and in relation to engineering implementation (robots as well rehabilitation devices) and social exploitation (sensory and motor rehabilitation). The exchange program proposed has the goal of joining the forces and expertises of the participating partners and of helping the formation and establishment of an international community of young researchers that shall effectively bridge the involved groups and their expertise in order to be effective in the long term.
- Queißer J, Steil JJ. Bootstrapping of parameterized skills through hybrid optimization in task and policy spaces. Frontiers in Robotics and AI. 2018;5:49.
- Queißer J, Ishihara H, Hammer B, Steil JJ, Asada M. Skill Memories for Parameterized Dynamic Action Primitives on the Pneumatically Driven Humanoid Robot Child Affetto. Presented at the International Conference on Development and Learning and on Epigenetic Robotics 2018 (ICDL-EPIROB2018), Tokyo .
- Schulz A, Queißer J, Ishihara H, Asada M. Transfer Learning of Complex Motor Skills on the Humanoid Robot Affetto. Presented at the International Conference on Development and Learning and on Epigenetic Robotics 2018 (ICDL-EPIROB2018), Tokyo (In Press).
- Malekzadeh M, Queißer J, Steil JJ. Imitation learning for a continuum trunk robot. In: Verleysen M, ed. Proceedings of the 25. European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. ESANN 2017. Louvain-la-Neuve: Ciaco; 2017.
- Malekzadeh M, Queißer J, Steil JJ. Learning the end-effector pose from demonstration for the Bionic Handling Assistant robot. Presented at the 9th Int. Workshop on Human-Friendly Robotics, Genoa.