Who we are?
The Aristotle University of Thessaloniki (AUTH) is the largest university in Greece, established in 1925. Today the main campus extends over an area of 429 metric acres in the heart of the city and comprises 10 Faculties organized in a total number of 40 Schools that cover the full range of scientific disciplines. More than 70.000 undergraduate and 8.000 postgraduate students are working towards their degrees at AUTH, while more than 2.200 Faculty members are employed as Teaching and Research personnel.
The Computational Intelligence and Deep Learning research group (CIDL) belongs to the Artificial Intelligence and Information Analysis Lab, Dept. of Informatics of the Aristotle University of Thessaloniki. The group is led by Prof. Anastasios Tefas, it consists of highly educated scientists, dedicated to the research in the fields of computational intelligence, deep learning, robotics, pattern recognition, statistical machine learning, digital signal and image analysis and retrieval and computer vision.
The premises where the CIDL group carries out significant scientific research is equipped with powerful, state-of-the-art computing nodes, providing unique research opportunities.
AUTH will coordinate the project and lead WP1 and WP9, taking care of financial, technical and organisational management, as well as organizing dissemination activities and ensuring that the necessary steps for efficient and effective exploitation are taken. AUTH main contributions will be in WP3, where it will contribute on deep person/face/body part active detection/recognition and pose estimation, deep person/face/body part tracking, human activity recognition, as well as social signal (facial expression, gesture, posture, etc) analysis and recognition and multi-modal human centric perception and cognition. AUTH will also lead task T4.1 (Object detection/recognition and semantic scene segmentation and understanding) and contribute to 2D/3D Object localization and tracking and 3D scene reconstruction. It will also contribute in the evaluation and benchmarking of the toolkit in WP8.
Tampere University (TAU) was created on 1 January 2019 as a merger of Tampere University of Technology (TUT) and the University of Tampere (UTA). The multidisciplinary, foundation-based Tampere University is Finland’s second-largest university with 20.000 students and 330 professors. TAU conducts scientific research in technology, health and society and provides high quality education within these fields.
In 2017 the total budget of the two universities was 322 M€ of which 42% was external research funding. TAU ranks fourth among all Finnish participants in H2020 funding. The university researcher community and support services have long experience in EU funding. TAU has an EU support team dealing with legal, financial and administrative matters.
Both TUT and UTA have been awarded with the European Commission’s “HR Excellence in Research” logo.
Prior to the merger, the Times Higher Education World University Ranking ranked TUT 11th in the world and 4th in Europe in innovation indicators based on industry collaboration. This ranking indicates how much companies are involved in and invest time in the active research area of the institution. In a global ranking, TUT was the world’s best for international student satisfaction with facilities and educational technology.
The Department of Computing Sciences (CS) includes 48 professors and over 380 other staff members. CS is responsible of over 140 research projects annually and has significant impact in higher education in Finland. CS has eight Finnish and international bachelor and master degree programs with 290 courses and 2100 enrolled students in 2018. The research areas cover the full spectrum of information technology research from human technology interaction to computer hardware. Research results are disseminated in over 580 scientific publications in 2018.
The Department of Automation Technology and Mechanical Engineering (ATME) is the leading university level educational and research institution in Finland in the field of automation, robotics and mechanical engineering, having over 20 professors and more than 100 researchers. The research areas relevant to OpenDR are robotics, production engineering and signal processing.
TAU will lead WP3, as task leader on T.3.4, undertaking the research on deep speech and biosignals analysis and recognition, and contributing on deep person/face/body part active detection/recognition and pose estimation and multi-modal human centric perception and cognition. TAU will also contribute on WP2 on defining the agile production use case requirements and specifications, WP4 on 2D/3D object localization and tracking, as well as sensor information fusion, and WP5 on deep navigation, human-robot interaction, deep action and control, and deep planning. Furthermore, TAU will work on integration of OpenDR to the specific agile production use case on WP7, as well as on the evaluation on WP8
Aarhus University (AU) was founded in 1928. It has four faculties – Faculty of Arts, Faculty of Science and Technology, Faculty of Health Sciences and School of Business & Social Sciences. The four faculties cover the entire research spectrum – basic research, applied research, strategic research and research based advice to the authorities. AU has 39,000 students; about 1,800 PhD students – of which one in four has a foreign nationality – and close to 900 postdoctoral scholars together with 11,500 employees. As such, internationalisation is a key part of AU’s mission and it continuously works to strengthen the international profile of the University through a series of initiatives, which will increase international research partnerships and the number of international students. AU has been establishing itself as a university for cutting-edge research, and has been moving up the most important university ranking lists. In 2018, the university is ranked at number 111 at the Leiden Ranking, number 141 at the QS World University Ranking, and number 123 of 17,000 universities on the Times Higher Education World University Ranking. AU has participated in 295 FP7 and 255 H2020 projects – 21 and 26, respectively, as coordinator – and has hosted/is currently hosting 51 ERC projects. AU has a very successful track record of managing both individual fellowships and large international projects and of hosting visiting researchers of all career stages for both training and knowledge transfer purposes.
AU will lead T4.2 on 2D/3D Object localization and tracking and will work on sensor information fusion, as well as contributing to object detection/recognition and semantic scene segmentation and understanding. AU will also contribute on WP3 working on deep person/face/body part active detection/recognition and pose estimation, deep person/face/body part tracking, human activity recognition, social signal analysis and recognition and multi-modal human centric perception and cognition. AU will also work on WP5 on deep planning, deep navigation and deep action and control.
Delft University of Technology (TUD) is the oldest and largest technical university in the Netherlands, with over 23,000 students and 3,300 scientific staff. Its high quality teaching standards and experimental facilities are renowned, placing it among the 6 top universities in Europe and top 18 in the world in the Engineering and Technology field (Times Higher Education 2017). It is a member of the IDEA League, a strategic alliance of five of Europe’s leading universities of technology. TU Delft comprises eight faculties: among these is the 3mE Faculty (Mechanical, Maritime and Materials Engineering), which hosts the Cognitive Robotics (CoR) department. In 2012, TU established the Delft Robotics Institute, in which CoR takes part. The Institute unites all the university’s research in the field of robotics, bringing together more than 150 scientific staff from six TU Delft faculties. The aim is to get robots and humans to work together effectively in unstructured environments, and real-world settings. Both the ‘hard’ robotics disciplines (mechatronics, embedded systems, control and AI) and the ‘soft’ ones (human-machine interaction, user interaction, architecture, ethics and design) are represented. A part of the Institute is the TU Delft Digital Innovation Hub called RoboValley (https://robovalley.com) established to foster collaboration with companies and technology transfer.
TUD will lead WP5, organizing the research activities on deep action and control, deep planning, as well as deep navigation. Furthermore, TUD will also lead and undertake the research activities on T5.4 on human robot interaction. TUD will also lead and organize the toolkit evaluation and benchmarking activities on WP8.
The Albert Ludwigs University Freiburg (ALU-FR), with more than 25,000 students and over 5,000 academic employees, is one of Germany’s leading research institutions with an international reputation in many fields. It is one of the oldest independent academic institutions in Germany and has been a comprehensive university since its founding (1457), and its diversity of disciplines provides an ideal environment for innovative interdisciplinary studies. It laureated in the German “Excellence” competitions (2007 for research and teaching, 2009 for instruction), and it is a member of LERU (League of European Research Universities).
ALU-FR has always been committed to provide state-of-the-art research facilities such as the Autonomous Intelligent Systems (AIS) lab, who is a leading laboratory in the areas of perception, state estimation and navigation for autonomous robots. This lab is especially known for its probabilistic state estimation techniques for simultaneous localization and mapping (SLAM).
The AIS lab has about 25 male and about 10 female members. It participated in a series of national and international projects including a transregional research center on spatial cognition and a research training group on embedded microsystems, as well as several European Projects. The AIS lab also heads the German cluster of excellence BrainLinks-BrainTools. The lab collaborates with numerous universities worldwide. ALU-FR will bring in its internationally renowned experts in robotics as well as results of several years of research in perception, manipulation, state estimation, localization and mapping and machine and deep learning.
ALU-FR will lead WP4 and also T4.3 on Deep SLAM and 3D scene reconstruction, as well as T5.2 on deep navigation. ALU-FR will also contribute on developing methodologies on deep planning on WP5.
Cyberbotics S.à r.l. (CYB) is a spin-off company from EPFL (Swiss Federal Institute of Technology in Lausanne) which has been developing the Webots robot simulator since 1998. It currently employs 6 people in Lausanne, Switzerland to continuously develop Webots according to customers needs. Cyberbotics provides consulting on both industrial and academic research projects and delivers open-source software solutions to its customers. It also provides user support and training to the users of the Webots software.
CYB will lead WP2 on requirements and specifications, working mainly on toolkit general requirements and toolkit general specifications, as well as WP6, working on developing simulation environments and data collection. CYB will also lead T7.1 on toolkit integration by collecting and integrating all the OpenDR modules developed by the partners.
PAL Robotics provides robotic products and services which can become an integral part people’s daily life, as well as research platforms as base for further innovation. It builds robot parts, personalized robot platforms, and robots for different service industries. The company also offers a range of robotic components – including actuation modules, robotic arms and mobile bases along with software i.e. navigation and manipulation software.
PAL Robotics expertise is based in the integral construction of robots, both from the mechanically and electronic point of view, as well as the development of the software are designed and implemented within the company. PAL Robotics robots are constructed in a modular way, so it is possible to change the different components of the robot. They operate in crowded environments, regularly tested in shopping malls, museums and conference centres, ensuring its interaction with hundreds of people. Safety considerations are included in the design of the robots and movements definition. Currently, PAL Robotics has almost 40 employees specialized in a range of disciplines, such as mechanical and electronic design, software HRI, movement control, vision and navigation, amongst others. PAL Robotics´ expertise is rooted in the integral approach adopted in the design and development of advanced robotics platforms, which main features are:
Modular design both in hardware and in software
Active and passive safety
Hardware abstraction layer with open interfaces
ROS (Robot Operating System) framework is fully supported
Realistic 3D dynamic simulation
Whole body control capabilities
Human Robot Interaction
Extensive documentation and tutorials
PAL will lead WP7 organizing and co-ordinating the toolkit integration, as well as the use case integration activities. PAL will also contribute on WP2 on defining the healthcare robotics use case requirements and specifications. Furthermore, PAL will work on integration of OpenDR to the specific healthcare robotics use case on WP7, as well as on the evaluation on WP8.
Within the last years, Agrointelli has developed an autonomous tool-carrier, Robotti, designed specifically for agricultural field operations such as mechanical weeding, seeding and spraying. In addition, the mechanical design includes a standard three-point linkage that allow for mounting a wide range of agricultural implements. The robot is four wheel driven and the high-level control consisting of route planning and navigation are carried out using ROS. The ability to mount a wide range of tools and implements results in numerous scenarios of different usage of the robot that we need to consider in the development of the machine, of where the OpenDR toolkit will be applicable.
AGI will lead WP8 organizing and coordinating the toolkit evaluation, as well as the use case specific toolkit evaluation activities. AGI will also contribute on WP2 on defining the agri-food use case requirements and specifications. Furthermore, AGI will work on deep SLAM and 3D scene reconstruction (WP4). Furthermore, AGI will work on integration of OpenDR to the specific agri-food use case on WP7, as well as on the evaluation on WP8.