Research Project

IMAGINE: Robots understanding their actions by imagining their effects

Type

European Project

Start Date

01/01/2017

End Date

28/02/2021

Project Code

H2020-ICT-2016-1-731761

Project illustration

Staff

Project Description

Today´s robots are good at executing programmed motions, but they do not understand their actions in the sense that they could automatically generalize them to novel situations or recover from failures. IMAGINE seeks to enable robots to understand the structure of their environment and how it is affected by their actions. "Understanding" here means the ability of the robot (a) to determine the applicability of an action along with parameters to achieve the desired effect, and (b) to discern to what extent an action succeeded, and to infer possible causes of failure and generate recovery actions.

The core functional element is a generative model based on an association engine and a physics simulator. "Understanding" is given by the robot´s ability to predict the effects of its actions, before and during their execution. This allows the robot to choose actions and parameters based on their simulated performance, and to monitor their progress by comparing observed to simulated behavior.

This scientific objective is pursued in the context of recycling of electromechanical appliances. Current recycling practices do not automate disassembly, which exposes humans to hazardous materials, encourages illegal disposal, and creates significant threats to environment and health, often in third countries. IMAGINE will develop a TRL-5 prototype that can autonomously disassemble prototypical classes of devices, generate and execute disassembly actions for unseen instances of similar devices, and recover from certain failures. For robotic disassembly, IMAGINE will develop a multi-functional gripper capable of multiple types of manipulation without tool changes.

IMAGINE raises the ability level of robotic systems in core areas of the work programme, including adaptability, manipulation, perception, decisional autonomy, and cognitive ability. Since only one-third of EU e-waste is currently recovered, IMAGINE addresses an area of high economical and ecological impact.

Consortium:
Universitaet Innsbruck, Austria (Coordinator)
Georg-August-Universität Göttingen, Germany
Karlsruher Institute fuer Technology, Germany
Institut National des Sciences Appliquées de Rennes, France
Institut de Robòtica i Informàtica Industrial -CSIC-, Spain
Bogazici Universitesi, Turkey
Electrocycling GmbH, Germany

Project Publications

Journal Publications

  • A. Suárez, C. Torras and G. Alenyà. Practical resolution methods for MDPs in robotics exemplified with disassembly planning. IEEE Robotics and Automation Letters, 4(3): 2282-2288, 2019.

    Open/Close abstract Abstract Info Info pdf PDF
  • D. Martínez, G. Alenyà, T. Ribeiro, K. Inoue and C. Torras. Relational reinforcement learning for planning with exogenous effects. Journal of Machine Learning Research, 18(78): 1-44, 2017.

    Open/Close abstract Abstract Info Info pdf PDF

Conference Publications

  • A. Suárez, J. Segovia, C. Torras and G. Alenyà. Online action recognition, 35th AAAI Conference on Artificial Intelligence, 2021, (Virtual), pp. 11981-11989.

    Open/Close abstract Abstract Info Info pdf PDF
  • A. Suárez, A. Andriella, A. Taranović, J. Segovia, C. Torras and G. Alenyà. Automatic learning of cognitive exercises for socially assistive robotics, 30th IEEE International Symposium on Robot and Human Interactive Communication, 2021, Vancouver, Canada, pp. 139-146.

    Open/Close abstract Abstract Info Info pdf PDF
  • J. Segovia, S. Jiménez and A. Jonsson. Generalized planning with positive and negative examples, 34th AAAI Conference on Artificial Intelligence, 2020, New York, USA, pp. 9949-9956.

    Open/Close abstract Abstract Info Info pdf PDF
  • A. Suárez, T. Gaugry, J. Segovia, A. Bernardin, C. Torras, M. Marchal and G. Alenyà. Leveraging multiple environments for learning and decision making: a dismantling use case, 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2020, Las Vegas, NV, USA, pp. 6902-6908.

    Open/Close abstract Abstract Info Info pdf PDF
  • A. Suárez, J. Segovia, C. Torras and G. Alenyà. STRIPS action discovery, 2020 AAAI 2020 workshop on Generalization in Planning, 2020, New York, pp. 1-9.

    Open/Close abstract Abstract Info Info pdf PDF
  • A. Suárez, G. Alenyà and C. Torras. Interleaving hierarchical task planning and motion constraint testing for dual-arm manipulation, 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2018, Madrid, Spain, pp. 4061-4066.

    Open/Close abstract Abstract Info Info pdf PDF