Aarhus University Seal

Open Deep Learning Toolkit for Robotics - OpenDR

The aim of OpenDR is to develop a modular, open and non-proprietary toolkit for core robotic functionalities by harnessing deep learning to provide advanced perception and cognition capabilities, meeting in this way the general requirements of robotics applications in the applications areas of healthcare, agri-food and agile production. The term toolkit in OpenDR refers to a set of deep learning software functions, packages and utilities used to help roboticists to develop and test a robotic application that incorporates deep learning. OpenDR will provide the means to link the robotics applications to software libraries (deep learning frameworks, e.g., Tensorflow) and to link it with the operating environment (ROS). OpenDR focuses on the AI and Cognition core technology in order to provide tools that make robotic systems cognitive, giving them the ability to a) interact with people and environments by developing deep learning methods for human centric (WP3) and environment (WP4) active perception and cognition, b) learn and categorise by developing deep learning tools for training and inference in common robotics settings (WP3 and WP4), and c) make decisions and derive knowledge by developing deep learning tools for cognitive robot action and decision making (WP5). As a result, the developed OpenDR toolkit will also enable cooperative human-robot interaction as well as the development of cognitive mechatronics where sensing and actuation are closely coupled with cognitive systems thus contributing to another two core technologies beyond AI and Cognition. OpenDR will develop, train, deploy and evaluate deep learning models that improve the technical capabilities of the core technologies beyond the current state of the art. It will enable a greater range of robotics applications that can be demonstrated at TRL 3 and above, thus lowering the technical barriers within the prioritised application areas. OpenDR aims to an easily adopted methodology to adapt the provided tools in order to solve any robotics task without restricting it to any specific application.

OpenDR’s expected impact is to improve the technical capabilities in robotics by providing easily deployable, efficient and novel Deep Learning tools, as well as to lower the technical barriers by providing a modular and open platform for developing Deep Learning for Robotics tools. Concerning industry, OpenDR’s expected impact is to enable a greater range of applications in agri-food, healthcare robotics and agile production, as well as to strengthen the competitiveness of companies by lowering the cost to access robotics-oriented Deep Learning tools.

The OpenDR consortium consists of 5 top-ranked academic and research institutes: Aristotle University of Thessaloniki (Greece), Tampere University (Finland), Aarhus University (Denmark), Delft University of Technology (Netherlands), University of Freiburg (Germany) and 3 leading industry partners: CYBERBOTICS (Switzerland), PAL Robotics (Spain) and AgroIntelli (Denmark). The Aarhus University team is focusing on robotic vision tasks of human activity recognition, 2D/3D object detection and tracking, social signal analysis, and robot planning.

You can find more information and news about the OpenDR project on its website: http://opendr.eu/