Robots Understanding Their Actions by Imagining Their Effects
IMAGINE seeks to enable robots to understand how their environment behaves and how to interact with it. 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 and simulate the effects of its actions, before and during their execution. This allows the robot to choose actions based on their simulated performance, and to monitor them by matching observed to simulated behaviour.
IMAGINE will produce novel robotic systems with the ability to understand, apply and generalize actions learned from human demonstration based on a functional analysis of parts, resulting in substantial advances in autonomy, flexibility and robustness of interaction in the presence of changes in task and environment. The immediate application is the dismantling of electrical appliances in recycling, where the increase in human productivity and safety achieved by the project can have substantial environmental and economic impact.
KARLSRUHER INSTITUT FÜR TECHNOLOGIE
INSTITUT NATIONAL DES SCIENCES APPLIQUÉES DE RENNES
CONSEJO SUPERIOR DE INVESTIGACIONES CIENTÍFICAS
Coordinator: Justus Piater
Project website: https://imagine-h2020.eu/
Call H2020-ICT-26-2016 a
Duration 1 January 2017 – 31 December 2019
Project ID 731761