Skip to main content

Goal-based Open-ended Autonomous Learning Robots

Deliverables

Science communication

Communication activities, supported by services from Robohub within the collaboration with them, and Blogs, in particular within the Digital Europe Agenda system.

Project web-site and logo

First version of project web-site. Logo of project. The maintenance of the web-site is an activity that will last during the whole project. The consortium, under the lead of CNR, will work to design the web-site structure and logo, and to populate it with contents related to the project. CNR will pay some professionals for support on technical aspects contributing to the realisation of the web-site.

Professional video on GOAL2 Demo

Professional video on the second demo GOAL2 .

Action Check Meeting 2

Draft agenda and list of presentations of Check Meeting 2.

Action Check Meeting 1

Draft agenda and list of presentations of Check Meeting 1.

Dissemination and Exploitation Plan: first version

Plan of dissemination and exploitation: first compilation. CNR will lead the whole plan compilation. As stated in the submitted proposal, TUDa will lead the exploitation section of the plan.

Goal-based skill learning: first results

Algorithms allowing autonomous robots to learn large repertoires of flexible skills based on goals.

Scientific publications and conferences

Publications in journals, and presentations at conferences and other similar scientific events, on the project results.

GOAL in infants and its models: first results

Report on how contingencies and goals are discovered and drive autonomous motor learning in infants, and formalisation of discovered mechanisms in computational models.

Self goal-generation: first results

Mechanisms for goal formation, and architectures/mechanisms to support open-ended learning of skills in robots.

Dissemination and Exploitation Plan: second version

Plan of dissemination and exploitation: revision leading to the second version of the plan. CNR will lead the whole plan compilation. As stated in the submitted proposal, TUDa will lead the exploitation section of the plan.

GOAL1 Demonstrator

GOAL1 demonstrator will show: (a) the set-ups used in the Demos; (b) the operation of the basic robot architectures supporting open-ended learning of GOAL2/GOAL3; (c) the successful accomplishment of initial open-ended learning processes.

GOAL2 Demonstrator

GOAL2 demonstrator will show robots that can learn several different skills in an open-ended fashion.

Plan for Open Data Pilot: first version

First document describing the data management plan to comply to the Open Data Pilot initiative.

Plan for Open Data Pilot: second version

Revised second document describing the data management plan to comply to the Open Data Pilot initiative.

First ethical implications workshop

Organisation of the first workshop on the possible implications of open-ended learning for society.

First open-ended learning competition

First open-ended learning competitions related to open-ended learning, supporting the definition of a benchmark for open-ended learning.

International Summer School

International Summer School on the project topics and approach to open-ended learning.

Industry fair: first participation

First participation to an industry fair and contacts with companies for the possible exploitation of the project outcomes.

Searching for OpenAIRE data...

Publications

The architecture challenge: Future artificial-intelligence systems will require sophisticated architectures, and knowledge of the brain might guide their construction

Author(s): Gianluca Baldassarre, Vieri Giuliano Santucci, Emilio Cartoni, Daniele Caligiore
Published in: Behavioral and Brain Sciences, Issue 40, 2017, ISSN 0140-525X
DOI: 10.1017/S0140525X17000036

Which limb is it? Responses to vibrotactile stimulation in early infancy

Author(s): Eszter Somogyi, Lisa Jacquey, Tobias Heed, Matej Hoffmann, Jeffrey J. Lockman, Lionel Granjon, Jacqueline Fagard, J. Kevin O'Regan
Published in: British Journal of Developmental Psychology, 2017, ISSN 0261-510X
DOI: 10.1111/bjdp.12224

Know Your Body Through Intrinsic Goals

Author(s): Francesco Mannella, Vieri G. Santucci, Eszter Somogyi, Lisa Jacquey, Kevin J. O'Regan, Gianluca Baldassarre
Published in: Frontiers in Neurorobotics, Issue 12, 2018, ISSN 1662-5218
DOI: 10.3389/fnbot.2018.00030

Acceptability of the Transitional Wearable Companion “+me” in Typical Children: A Pilot Study

Author(s): Valerio Sperati, Beste Özcan, Laura Romano, Simone Scaffaro, Tania Moretta, Giada Turturo, Maria Nicoletta Aliberti, Vincenzo Guidetti, Gianluca Baldassarre
Published in: Frontiers in Psychology, Issue 10, 2019, ISSN 1664-1078
DOI: 10.3389/fpsyg.2019.00125

Action Observation With Dual Task for Improving Cognitive Abilities in Parkinson’s Disease: A Pilot Study

Author(s): Daniele Caligiore, Magda Mustile, Alissa Fineschi, Laura Romano, Fabrizio Piras, Francesca Assogna, Francesco E. Pontieri, Gianfranco Spalletta, Gianluca Baldassarre
Published in: Frontiers in Systems Neuroscience, Issue 13, 2019, ISSN 1662-5137
DOI: 10.3389/fnsys.2019.00007

Bio-Inspired Model Learning Visual Goals and Attention Skills Through Contingencies and Intrinsic Motivations

Author(s): Valerio Sperati, Gianluca Baldassarre
Published in: IEEE Transactions on Cognitive and Developmental Systems, Issue 10/2, 2018, Page(s) 326-344, ISSN 2379-8920
DOI: 10.1109/tcds.2017.2772908

Dysfunctions of the basal ganglia-cerebellar-thalamo-cortical system produce motor tics in Tourette syndrome

Author(s): Daniele Caligiore, Francesco Mannella, Michael A. Arbib, Gianluca Baldassarre
Published in: PLOS Computational Biology, Issue 13/3, 2017, Page(s) e1005395, ISSN 1553-7358
DOI: 10.1371/journal.pcbi.1005395

Fetal Origin of Sensorimotor Behavior

Author(s): Jaqueline Fagard, Rana Esseily, Lisa Jacquey, Kevin O’Regan, Eszter Somogyi
Published in: Frontiers in Neurorobotics, Issue 12, 2018, ISSN 1662-5218
DOI: 10.3389/fnbot.2018.00023

Stage-Wise Learning of Reaching Using Little Prior Knowledge

Author(s): François de La Bourdonnaye, Céline Teulière, Jochen Triesch, Thierry Chateau
Published in: Frontiers in Robotics and AI, Issue 5, 2018, ISSN 2296-9144
DOI: 10.3389/frobt.2018.00110

Intrinsic motivation and mental replay enable efficient online adaptation in stochastic recurrent networks

Author(s): Daniel Tanneberg, Jan Peters, Elmar Rueckert
Published in: Neural Networks, Issue 109, 2019, Page(s) 67-80, ISSN 0893-6080
DOI: 10.1016/j.neunet.2018.10.005

A Reinforcement Learning Architecture that Transfers Knowledge between Skills when Solving Multiple Tasks

Author(s): Paolo Tommasino, Daniele Caligiore, Marco Mirolli, Gianluca Baldassarre
Published in: IEEE Transactions on Cognitive and Developmental Systems, 2019, Page(s) 1-1, ISSN 2379-8920
DOI: 10.1109/tcds.2016.2607018

Inverse reinforcement learning via nonparametric spatio-temporal subgoal modeling

Author(s): A. Šošić, E. Rueckert, J. Peters, A. M. Zoubir, H. Koeppl
Published in: The Journal of Machine Learning Research, 2018, ISSN 1532-4435

Autonomous Development of Active Binocular and Motion Vision Through Active Efficient Coding

Author(s): Alexander Lelais, Jonas Mahn, Vikram Narayan, Chong Zhang, Bertram E. Shi, Jochen Triesch
Published in: Frontiers in Neurorobotics, Issue 13, 2019, ISSN 1662-5218
DOI: 10.3389/fnbot.2019.00049

Robot End Effector Tracking Using Predictive Multisensory Integration

Author(s): Lakshitha P. Wijesinghe, Jochen Triesch, Bertram E. Shi
Published in: Frontiers in Neurorobotics, Issue 12, 2018, ISSN 1662-5218
DOI: 10.3389/fnbot.2018.00066

Intrinsically Motivated Discovered Outcomes Boost User’s Goals Achievement in a Humanoid Robot

Author(s): Kristsana Seepanomwan, Vieri Giuliano Santucci, Gianluca, Baldassarre
Published in: Development and Learning and Epigenetic Robotics (ICDL-EpiRob), 2017 Joint IEEE International Conference on, 2017

Workshop On Ethical Issues of Open Ended-Learning in Autonomous Robots

Author(s): Daniele Caligiore, Vieri Giuliano Santucci, Gianluca Baldassarre
Published in: Development and Learning and Epigenetic Robotics (ICDL-EpiRob), 2017 Joint IEEE International Conference on, 2017

Learning of Active Binocular Vision in a Biomechanical Model of the Oculomotor System

Author(s): Lukas Klimmasch, Alexander Lelais, Alexander Lichtenstein, Bertram Emil Shi, Jochen Triesch
Published in: Development and Learning and Epigenetic Robotics (ICDL-EpiRob), 2017 Joint IEEE International Conference on, 2017

Online Learning with Stochastic Recurrent Neural Networks using Intrinsic Motivation Signals

Author(s): Tanneberg D., Peters J., Rueckert E.
Published in: Proceedings of Machine Learning Research: Conference on Robot Learning (CoRL), 2017

Learning Inverse Dynamics Models in O(n) time with LSTM networks

Author(s): Rueckert E., Nakatenus M., Tosatto S., Peters, J
Published in: Proceedings of the International Conference on Humanoid Robots (HUMANOIDS), 2017

Efficient Online Adaptation with Stochastic Recurrent Neural Networks

Author(s): Tanneberg D., Peters J., Rueckert E.
Published in: Proceedings of the International Conference on Humanoid Robots (HUMANOIDS), 2017

A Comparison of Distance Measures for Learning Nonparametric Motor Skill Libraries

Author(s): Stark S., Peters J., Rueckert E.
Published in: Proceedings of the International Conference on Humanoid Robots (HUMANOIDS), 2017

Simulation of the underactuated Sake Robotics Gripper in V-REP

Author(s): Thiem S., Stark S., Tanneberg D., Peters J., Rueckert E.
Published in: Workshop at the International Conference on Humanoid Robots (HUMANOIDS), 2017

Development of reaching to the body in early infancy: From experiments to robotic models

Author(s): Matej Hoffmann, Lisa K. Chinn, Eszter Somogyi, Tobias Heed, Jacqueline Fagard, Jeffrey J. Lockman, J. Kevin O'Regan
Published in: 2017 Joint IEEE International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob), 2017, Page(s) 112-119
DOI: 10.1109/devlrn.2017.8329795

Action-outcome contingencies as the engine of open-ended learning: computational models and developmental experiments

Author(s): G. Baldassarre, F. Mannella, V.G. Santucci,E. Somogyi, L. Jacquey, M. Hamilton, J.K. O’Regan
Published in: Proceedings of The 8th IEEE International Conference on Development and Learning and Epigenetic Robotics (ICDL-EpiRob2018), 2018

Learning to Categorize Bug Reports with LSTM Networks

Author(s): K.D. Gondaliya, J. Peters, E. Rueckert
Published in: International Conference on Advances in System Testing and Validation Lifecycle, 2018

An Active Efficient Coding Model of Binocular Vision Development Under Normal and Abnormal Rearing Conditions

Author(s): Lukas Klimmasch, Johann Schneider, Alexander Lelais, Bertram E. Shi, Jochen Triesch
Published in: From Animals to Animats 15 - 15th International Conference on Simulation of Adaptive Behavior, SAB 2018, Frankfurt/Main, Germany, August 14-17, 2018, Proceedings, Issue 10994, 2018, Page(s) 66-77
DOI: 10.1007/978-3-319-97628-0_6

The development of reaching and grasping: towards an integrated framework based on a critical review of computational and robotic models

Author(s): D. Caligiore, G. Baldassarre
Published in: 2019

Recurrent Connections Aid Occluded Object Recognition by Discounting Occluders

Author(s): Markus Roland Ernst, Jochen Triesch, Thomas Burwick
Published in: Artificial Neural Networks and Machine Learning – ICANN 2019: Image Processing - 28th International Conference on Artificial Neural Networks, Munich, Germany, September 17–19, 2019, Proceedings, Part III, Issue 11729, 2019, Page(s) 294-305
DOI: 10.1007/978-3-030-30508-6_24

Learning Hierarchical Integration of Foveal and Peripheral Vision for Vergence Control by Active Efficient Coding

Author(s): Zhetuo Zhao, Jochen Triesch, Bertram E. Shi
Published in: From Animals to Animats 15 - 15th International Conference on Simulation of Adaptive Behavior, SAB 2018, Frankfurt/Main, Germany, August 14-17, 2018, Proceedings, Issue 10994, 2018, Page(s) 78-89
DOI: 10.1007/978-3-319-97628-0_7

Autonomous discovery of the goal space to learn a parameterized skill

Author(s): E. Cartoni, G. Baldassarre
Published in: 2018

Autonomous Open-Ended Learning of Interdependent Tasks

Author(s): V.G. Santucci, E. Cartoni, B. Castro da Silva, G. Baldassarre
Published in: 2019

Learning to touch objects through stage-wise deep reinforcement learning

Author(s): de La Bourdonnaye , François; Teulière , Céline; Triesch , Jochen; Chateau , Thierry
Published in: IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Issue 1, 2019

Autonomous Reinforcement Learning of Multiple Interrelated Tasks

Author(s): Santucci V. G., Baldassarre G. and Cartoni E.
Published in: arXiv, 2019

Recurrent Connectivity Aids Recognition of Partly Occluded Objects

Author(s): Ernst, Markus Roland and Triesch, Jochen and Burwick, Thomas
Published in: arXiv, 2019