LEArning-CONtrol tight interaction: a novel approach to robust execution of mobile manipulation tasks

Project Information

LEACON

Grant agreement ID: 659265

DOI 10.3030/659265

Project terminated on 7 February 2018

EC signature date 12 March 2015

Start date 7 March 2016

End date 6 March 2018

Funded under

• EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions

Total cost € 159 460,80

EU contribution € 159 460,80

Coordinated by

TECHNISCHE UNIVERSITAET MUENCHEN
Germany

Objective

One of the main challenges of roboticists is to take robots out of the factories and let them enter into unstructured environments, such as houses, hospitals, small manufacturers and dangerous area.
The objective of the project is to take a step towards the presence of robots in such environments.
Currently, there are still important obstacles to the massive diffusion of advanced mobile manipulation systems in the fields described above. First of all, programming mobile manipulators with the classical methods is still too expensive and time-consuming due to intrinsic complexity of mobile manipulation tasks.
A second limitation is that planning the robot motion completely off-line, as often happens in classical industrial applications, may likely bring to a failure of the assigned task, since a high degree of uncertainty is present and the environment can dynamically change. Such features may cause safety issues for humans potentially present in the workspace and for the external environment itself.
In order to tackle these limitations, the LEACON project has the objective to develop a framework that:
- allows robots to learn in a real world scenario manipulation skills from human demonstration
-exploits multimodal perception (tactile, proximity, visual, force sensors) to increase the robustness to unforeseen events and safety when manipulation tasks are executed.
To fulfill such objectives, a multidisciplinary approach that combines machine learning and perception-based control is proposed. The core of the proposed framework will provide two planning levels tightly connected: the high-level and low-level cognitive system.
To show the effectiveness of the developed architecture, the main use cases will be constituted by a robot that performs picking, manipulation, and placing operations in a dynamic, unstructured environment in presence of humans in its workspace. At the end of the project, the developed software will be released as open source code.

Fields of science

• natural sciences › computer and information sciences › artificial intelligence › machine learning › reinforcement learning
• social sciences › sociology › industrial relations › automation
• engineering and technology › electrical engineering, electronic engineering, information engineering › electronic engineering › robotics › autonomous robots
• engineering and technology › electrical engineering, electronic engineering, information engineering › electronic engineering › sensors
• engineering and technology › electrical engineering, electronic engineering, information engineering › electronic engineering › control systems › home automation

Keywords

• LEACON

Programme(s)

• H2020-EU.1.3. - EXCELLENT SCIENCE - Marie Skłodowska-Curie Actions Main Programme
• H2020-EU.1.3.2. - Nurturing excellence by means of cross-border and cross-sector mobility

Topic(s)

• MSCA-IF-2014-EF - Marie Skłodowska-Curie Individual Fellowships (IF-EF)

Call for proposal

H2020-MSCA-IF-2014

Funding Scheme

MSCA-IF - Marie Skłodowska-Curie Individual Fellowships (IF)

Coordinator

TECHNISCHE UNIVERSITAET MUENCHEN

• Net EU contribution: € 159 460,80
• Address: Arcisstrasse 21 ; 80333 Muenchen ; Germany
• Region: Bayern › Oberbayern › München, Kreisfreie Stadt
• Activity type: Higher or Secondary Education Establishments
• Other funding: € 0,00