What is R4A?

R4A (Robotics For All) is a robotics team focused on delivering enhanced functionality in/with robots, in a fast and easy manner. The team is manned from Faculty members and Researchers at the Electrical and Computer Engineering School, Aristotle University of Thessaloniki, Greece that have been extensively working on middleware/software robotics. The RAPP EU and the PANDORA projects are two of the main assets of the R4A members' portfolio.

The main novelty of the R4A approach is the design and implementation of a middleware architecture and the respective tools that allow for rapid prototyping of robot-agnostic applications. The R4A architecture supports the collaboration of robots with any connected device or "thing" (in the IoT concept), as well as with robotic cloud services, this way supporting a wide range of applications, even with low cost robots. The R4A approach has already been used in realistic environments and tested by developers in heterogeneous consumer robotic devices.

 
 

Lines of research

Robotics middleware

Our main line of research is the development of robot-agnostic middleware through Model-Driven Engineering approaches. This way we envision that any roboticist or developer will be capable of building properly desinged robotics software in a fast and easy manner. The R4A middlewares offer a set of web-based APIs, via which developers can fully manipulate a robotic device with minimum coding effort.

Robotic applications

Using the robotics middleware we have developed, we are able of designing and implementing robot-agnostic applications, which can be executed both in or out of a robot. Thus we treat robots as platforms, which can download and execute robotic applications in a seamless way. Tracking, surveillance and age friendly applications are typical cases developed.


Robotic algorithms

Apart from application development, the R4A group has extensive experience in traditional robotics algorithmic development also. We conduct ongoing scientific research in topics such as SLAM, navigation and path planning, exploration and coverage, robot perception and data fusion, desicion making and evolutionary/deep learning.

Robotic connectivity

Towards connecting robots to already established solutions, we develop APIs that connect robotic applications with IoT frameworks, this way exchanging information with the environment. We also provide a mechanism for consuming cloud-based functionality, this way reducing the computational burden from our robots.

Our vision



Our vision lies within the "Robotics 4 All" concept. We argue that robotic software development can be fast and easy, provided the suitable tools exist. Thus, our research efforts are oriented towards creating the appropriate software, which will support the rapid-prototyping concept in the robotics field.

In order to do so, we have developed the R4A middleware framework that supports the creation of robot-agnostic applications, that could potentially be downloaded as-is to heterogeneous robots and executed automatically. Furthermore, we fully support the utilization of the ROS ecosystem for having access to low-level robotic algorithms, as well as treating robots as "things", providing application-level access to the IoT world. Finally, we promote the cloud robotics concept, by creating cloud-based web-services that are ready-to-be-used by any robotic application.

Our ultimate vision includes a web-based workbench, for anyone to graphically create robotic software and interconnect it with existent robotic modules, services or other devices.

What our team offers

Robotics applications

We can uptake the task of bringing to life envisioned robotic workflows or tasks, in the form of robotic applications, using our enhanced middleware framework. These may include simple social interaction behaviors up to complex multi-robot applications. Furthermore, we can create the appropriate software towards supporting a new robot to our infrastructure, aiming at being able to create applications for it.

Robotic SW/HW solutions

We can create tailored software robotic solutions by using both high-level components (the R4A framework), as well as custom-made or existent robotic algorithms, which include mapping, navigation, exploration, speech-to-text or image processing functionalities. Furthermore, we can implement hardware robotic solutions using embedded HW modules and a variety of sensors and effectors.

Robots

Our laboratory is equipped with various robots, HW modules, sensors and effectors. Some of them are 3 custom made UGVs (examples 1, 2, 3), Turtlebot and Turtlebot 2 robots, one NAO robot, two AR.Drone 2.0 robots, several sensors including Lidars, Sonars, 2D/3D cameras (Kinect 1, 2 and Asus Xtion Pro), CO2, Sound, Thermal arrays and cameras, IMUs, compasses, a variety of effectors (DC brushless motors, servos, linear actuators), several embedded HW modules (Raspberry PI 2/3, NodeMCU, Odroid-XU4, FPGAs) and others.

Projects

The RAPP EU Project

RAPP provides an open source software platform to support the creation and delivery of robotic applications. These robotic applications are in turn expected to increase the versatility and utility of service and assistive robots. The emphasis of the RAPP project was on applications that will enable robots to understand and respond to the intentions and needs of people at risk of exclusion.

PANDORA Robotics

The PANDORA project (Program for the Advancement of Non-Directed Operating Robotic Agents) is focused on developing a robotic platform for room exploration and victim identification. PANDORA has repeatedlly participated in the RoboCup-Rescue competition, achieving in winning the 2nd place in the Autonomous vehicles contest twice (2013, 2015).

REMEDES System

The R4A group is currently developing a low cost and high accuracy system for recording reflex measurements using optical or acoustical stimulation. REMEDES is suitable for medical applications (diagnosis, rehabilitation or tracking neurological, neurokinetic, psychiatric or opthalmological conditions), as well as for sports applications (modeling the reaction of athletes or tracking their reflexes profile)

Indicative publications



Journals

  • A. Thallas, E.G. Tsardoulias, L. Petrou "Topological based scan matching – odometry posterior sampling in RBPF under kinematic model failures" Journal of Intelligent & Robotic Systems (2017).
  • Zieliński, Cezary, Maciej Stefańczyk,..., Emmanouil Tsardoulias, Konstantinos Panayiotou, Andreas Symeonidis et al. "Variable structure robot control systems: The RAPP approach." Robotics and Autonomous Systems (2017).
  • Christoforos Zolotas, Themistoklis Diamantopoulos, Kyriakos Chatzidimitriou and Andreas Symeonidis "From requirements to source code: a Model-Driven Engineering approach for RESTful web services" Automated Software Engineering, pp. 1-48, 2016 Sep
  • Tsardoulias, Emmanouil G., Athanassios M. Kintsakis, Konstantinos Panayiotou et al. "Towards an integrated robotics architecture for social inclusion–The RAPP paradigm." Cognitive Systems Research (2016).
  • Reppou, Sofia E., Emmanouil G. Tsardoulias, Athanassios M. Kintsakis, Andreas L. Symeonidis et al. "RAPP: A Robotic-Oriented Ecosystem for Delivering Smart User Empowering Applications for Older People." International Journal of Social Robotics (2016): 1-14.
  • Zolotas, Christoforos, Themistoklis Diamantopoulos, Kyriakos C. Chatzidimitriou, and Andreas L. Symeonidis. "From requirements to source code: a Model-Driven Engineering approach for RESTful web services." Automated Software Engineering (2016): 1-48.
  • Tsardoulias, Emmanouil, Aristeidis G. Thallas, Andreas L. Symeonidis, and Pericles A. Mitkas. "Improving Multilingual Interaction for Consumer Robots through Signal Enhancement in Multichannel Speech." Journal of the Audio Engineering Society 64, no. 7/8 (2016): 514-524.
  • Tsardoulias, E. G., A. Iliakopoulou, A. Kargakos, and L. Petrou. "Cost-Based Target Selection Techniques Towards Full Space Exploration and Coverage for USAR Applications in a Priori Unknown Environments." Journal of Intelligent & Robotic Systems (2016): 1-28.
  • Tsardoulias, E. G., A. Iliakopoulou, A. Kargakos, and L. Petrou. "A Review of Global Path Planning Methods for Occupancy Grid Maps Regardless of Obstacle Density." Journal of Intelligent & Robotic Systems (2016): 1-30.
  • Tsardoulias, E. G., A. T. Serafi, M. N. Panourgia, A. Papazoglou, and L. Petrou. "Construction of Minimized Topological Graphs on Occupancy Grid Maps Based on GVD and Sensor Coverage Information." Journal of Intelligent & Robotic Systems 75, no. 3-4 (2014): 457-474.
  • Tsardoulias, Emmanouil, and Loukas Petrou. "Critical Rays Scan Match SLAM." Journal of Intelligent & Robotic Systems 72, no. 3-4 (2013): 441-462.

Conferences

  • Vasilis N. Remmas, Konstantinos Panayiotou, Emmanouil Tsardoulias, Andreas L. Symeonidis "SRCA - The Scalable Robotic Cloud Agents Architecture" In International Conference on Cloud and Robotics 2017, Saint Quentin, France, 2017 (Best Paper Award)
  • Panagiotis Doxopoulos, Konstantinos Panayiotou, Emmanouil Tsardoulias, Andreas L. Symeonidis "Creating an extrovert robotic assistant via IoT networking devices" In International Conference on Cloud and Robotics 2017, Saint Quentin, France, 2017 (Finalist for Best Paper Award)
  • Konstantinos Panayiotou, Sofia E. Reppou, George Karagiannis, Emmanouil Tsardoulias, Aristeidis G. Thallas, Andreas L. Symeonidis "Robotic applications towards an interactive alerting system for medical purposes" In 30th IEEE International Symposium on Computer-Based Medical Systems - IEEE CBMS, Thessaloniki, 2017
  • Thallas, Aristeidis G., Konstantinos Panayiotou, Emmanouil Tsardoulias, Andreas L. Symeonidis, Pericles A. Mitkas, and George G. Karagiannis. "Relieving Robots from Their Burdens: The Cloud Agent Concept (Short Paper)." In Cloud Networking (Cloudnet), 2016 5th IEEE International Conference on, pp. 188-191. IEEE, 2016.
  • Thallas, Aristeidis, Emmanouil Tsardoulias, and Loukas Petrou. "Particle filter—Scan matching hybrid SLAM employing topological information." In 2016 24th Mediterranean Conference on Control and Automation (MED), pp. 226-231. IEEE, 2016.
  • Thallas, Aristeidis, Emmanouil Tsardoulias, and Loukas Petrou. "Particle filter—Scan matching SLAM recovery under kinematic model failures." In 2016 24th Mediterranean Conference on Control and Automation (MED), pp. 232-237. IEEE, 2016.
  • Christoforos Zolotas and Andreas Symeonidis "Towards an MDA Mechanism for RESTful Services Development" The 18th International Conference on Model Driven Engineering Languages and Systems, Ottawa, Canada, 2015 Oct
  • Tsardoulias, E. G., A. L. Symeonidis, and P. A. Mitkas. "An automatic speech detection architecture for social robot oral interaction." In Proceedings of the Audio Mostly 2015 on Interaction With Sound, p. 33. ACM, 2015.
  • Zolotas, Christoforos, and Andreas L. Symeonidis. "Towards an MDA Mechanism for RESTful Services Development." In CloudMDE@ MoDELS, pp. 31-36. 2015.
  • Gkiokas, Alexandros, Emmanouil G. Tsardoulias, and Pericles A. Mitkas. "Hive Collective Intelligence for Cloud Robotics: A Hybrid Distributed Robotic Controller Design for Learning and Adaptation." In Progress in Automation, Robotics and Measuring Techniques, pp. 65-78. Springer International Publishing, 2015.
  • Tsardoulias, Emmanouil G., Cezary Zieliński, Włodzimierz Kasprzak, Sofia Reppou, Andreas L. Symeonidis, Pericles A. Mitkas, and George Karagiannis. "Merging Robotics and AAL Ontologies: The RAPP Methodology." In Progress in Automation, Robotics and Measuring Techniques, pp. 285-297. Springer International Publishing, 2015.

The team

Loukas Petrou

Associate professor, School of Electrical and Computer Engineering, Aristotle University of Thessaloniki

Scientific responsible on Hardware/Middleware Engineering

Main research interests include Microprocessor Applications, Autonomous Mobile robots, Intelligent Control, Design of reconfigurable embedded systems, Fuzzy logic and Neural Network control systems.

   loukas@eng.auth.gr

Andreas Symeonidis

Assistant professor, School of Electrical and Computer Engineering, Aristotle University of Thessaloniki

Scientific responsible on Middleware/Software Engineering

Main research interests include Software Engineering (SE) Processes, data mining and knowledge extraction from Software repositories, Model-driven Engineering, Software quality and Software analytics.

   asymeon@eng.auth.gr

Dr Eng Manos Tsardoulias

PostDoc researcher, School of Electrical and Computer Engineering, Aristotle University of Thessaloniki.

Technical Leader

His research interests are focused in Autonomous Robots, SLAM and navigation, multi-robot exploration, robotic architectures oriented for the creation of robotic applications and robot-agnostic RESTful APIs.

   etsardou@eng.auth.gr

MSc Eng Konstantinos Panayiotou

School of Electrical and Computer Engineering, Aristotle University of Thessaloniki.

Middleware developer

His research interests include robotics middleware, cloud-based architectures, IoT platforms, deep learning (DNN/CNN), embedded modules and hardware-oriented development, among others.

   klpanagi@ee.auth.gr

PhD Student Christoforos Zolotas

School of Electrical and Computer Engineering, Aristotle University of Thessaloniki.

Model Driven Engineering researcher

His research interests are focused in Automated Software Engineering Techniques (e.g. Model Driven Engineering), Robotics, RESTful Web Services and Logic in Computing.

   christopherzolotas@issel.ee.auth.gr

Msc. Eng. Marina Stamatiadou

School of Electrical and Computer Engineering, Aristotle University of Thessaloniki.

Business and product development

She is responsible for the business and product development as well as the marketing strategy of the R4A team focusing also on the organisational and operational procedures and processes.

   marina.stamatiadou@gmail.com

Mr Ilias Ouzοunidis

R4A Affiliate, Graphics designer and Production Manager

Branding design and Digital Promotion

A graphics and marketing enthusiast with more than 15 years of experience in branding, re-branding and disseminating products and ideas. Assigned with the exploitation and promotion of R4A products to interested stakeholders and relevant media.

   ilias@ouzounidis.com