Research Project

MdM: Unit of Excellence María de Maeztu


National Project

Start Date


End Date


Project Code


Project illustration


Project Description

> PhD Student - Fractionary systems in mechanical and fuel cell systems
> PhD Student - Action planning and synchronization for human-robot collaboration
> PhD Student - Supervision and control in human-centered robots
> PhD Student - Human movement and action estimation for Robot-Human Collaboration

The "María de Maeztu" scientific excellence seal has been awarded to IRI by the Spanish State Research Agency for the period 01/07/2017-30/06/2021. Beneficiary entities are both CSIC and UPC as a joint research unit.

A 4-year Strategic Research Programme will be developed in HUMAN-CENTERED ROBOTICS


The concept of “human-centered robotics” is meant to include the many different situations in which robots are in close contact and interact with humans. These include, but is not limited to, social robotics with significant human-robot interaction, collaborative robotics in which humans and robots work together to achieve a common task, or assistive robotics in which robotic technologies are exploited to help the elderly or the impaired. The 7 key core challenges identified apply to all these domains, and translate into this 7 Specific Scientific Objectives of significant relevance and novelty:

  • Emphatic natural human robot interaction and collaboration: Robots need to be accessible to the non-expert. Research is needed to find ways to allow a robot to interact with people in a natural and intuitive way, combining multiple sensing modalities to understand gestures, and to produce socially acceptable gestures and motion. Moreover, research is needed to develop models of robot behavior that guarantee that a task can be completed in cooperation with a human, understanding and anticipating human intentions, being a true collaborator.

  • Robust localization and mapping: Localization and mapping is a key component of any assistive robotics solution, and it is perhaps one of the most studied topics in robotics. Whereas there is today a general consensus on the mathematical formulation of the problem and on the computational tools needed for its solution, the objective now is to provide robustness, reliability, and scalability.

  • Dexterous textile manipulation: Textile manipulation would open a whole range of possibilities, from increased autonomy of the elderly, to housekeeping or industrial applications. In our approach to the problem, we seek to develop a theory of cloth manipulation using tools from computational topology and machine learning, and to develop the adequate end effectors and tools to handle cloth with dexterity.

  • Robot learning using natural communication: A new paradigm for robot learning will be studied. Exploiting recent advances in the area of visual object recognition, natural language processing, and their combination by means of machine learning tools, we seek to advance the way a human can instruct a robot using solely natural language and standard gestures.

  • Energy supply and optimization: Energy autonomy is also a key factor for the penetration of robust mobile systems in assistive robotics applications. We seek to develop an innovative fuel cell system tailored to assistive robotics applications with improved efficiency, reliability and lifetime, addressing the problems of space (high energy density) and weight (high specific energy). Additionally, we seek to explore the possibility of obtaining the fuel (hydrogen or hydrogen rich compounds) from different alternative means.

  • Supervision and control of complex dynamic systems: Robots need to react in real time and reliable alongside humans. Hence, the objective is to work on controllers that can change parameters in real time, efficiently handle constraints and disturbances, and detect faults and reconfigure to maintain functionality.

  • Ethical, regulatory and philosophical aspects of social robotics: Many risks and regulatory issues still need to be resolved prior to the deployment of helper robots. These issues have passed until now overwhelmingly unattended, concentrating research efforts in developing the technologies and the prototypes, but they can no longer be obviated. A collaborative effort with experts in legal frontiers of technology and privacy issues will be pursued, as well as with experts that can analyze the ethical perspective.


IRI’s María de Maeztu Strategic Research Program on Human-Centered Robotics is organized in 7 Work Packages, consistent with the 7 Research Objectives outlined above.

WP1 Emphatic natural human robot interaction and collaboration
- 1.1 Human action recognition
- 1.2 Socially aware path planning, navigation and path execution
- 1.3 Compliant and safe interaction between humans and robots
PI Leader: A. Sanfeliu
Other PIs involved: JM Porta , V Puig

WP2 Robust localization and mapping
- 2.1 Sensory data fusion for localization
- 2.2 SLAM and Deep Learning
- 2.3 Robustness/reliability/scalability/loop closure detection
PI Leader: J. Andrade
Other PIs involved: F. Moreno, A Sanfeliu

WP3 Dexterous textile manipulation
- 3.1 Perception of deformable objects
- 3.2 High level encoding of object primitives using computational topology
- 3.3 Design and construction of dexterous end effectors and grippers for robot design
PI Leader: C. Torras
Other PIs involved: F. Moreno, JM Porta

WP4 Robot learning using natural communication
- 4.1 Joint annotation of visual and textual data
- 4.2 Inference techniques (dynamic neural networks) to predict semantic tuples and actions from object detection in input images and video
- 4.3 Learning from demonstration manipulation skills
PI leader: F Moreno
Other PIs involved: J Andrade, C Torras

WP5 Energy supply and management
- 5.1 Design of autonomous energy systems based on PEM fuel cells
- 5.2 Dynamic modelling and development of diagnosis tools
- 5.3 Design and implementation of controllers for optimal efficiency and low degradation of the energy generation system
PI leader: V Puig
Other IPs involved: G Cembrano, A Sanfeliu

WP6 Advanced supervision and control
- 6.1 Modelling complex dynamic systems
- 6.2 Real-time supervision in complex dynamic systems
- 6.3 Model-based and data driven control in complex dynamic systems
PI leader: G Cembrano
Other PIs involved: J Andrade, V Puig

WP7 Ethical, regulatory and philosophical aspects of social robotics
- 7.1 Risk/contingency management
- 7.2 Liability /regulations
- 7.3 Privacy issues
PI Leader: A Sanfeliu
Other PIs involved: C Torras

Project Publications

Journal Publications

  • A. Agudo and F. Moreno-Noguer. Force-based representation for non-rigid shape and elastic model estimation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, to appear.

    Open/Close abstract Abstract Info Info pdf PDF
  • A. Santamaria-Navarro, G. Loianno, J. Solà, V. Kumar and J. Andrade-Cetto. Autonomous navigation of micro aerial vehicles: State estimation using fast and low-cost sensors. Autonomous Robots, 2018, to appear.

    Open/Close abstract Abstract Info Info pdf PDF
  • J. Vallvé, J. Solà and J. Andrade-Cetto. Graph SLAM sparsification with populated topologies using factor descent optimization. IEEE Robotics and Automation Letters, 3(2): 1322-1329, 2018.

    Open/Close abstract Abstract Info Info pdf PDF
  • C. Torras. Assistent(e)s robòtiqu(e)s: Un punt de confluència entre tecnociència i humanitats. Quadern de les idees, les arts i les lletres, 38(209): 19-21, 2017.

    Open/Close abstract Abstract Info Info pdf PDF

Conference Publications

  • V. Vaquero, I. del Pino, F. Moreno-Noguer, J. Solà, A. Sanfeliu and J. Andrade-Cetto. Deconvolutional networks for point-cloud vehicle detection and tracking in driving scenarios, 8th European Conference on Mobile Robots, 2017, Paris, France, pp. 1-7.

    Open/Close abstract Abstract Info Info pdf PDF
  • J. Vallvé, J. Solà and J. Andrade-Cetto. Factor descent optimization for sparsification in graph SLAM, 8th European Conference on Mobile Robots, 2017, Paris, France, pp. 1-6.

    Open/Close abstract Abstract Info Info pdf PDF
  • I. del Pino, V. Vaquero, B. Masini, J. Solà, F. Moreno-Noguer, A. Sanfeliu and J. Andrade-Cetto. Low resolution lidar-based multi object tracking for driving applications, 3rd Iberian Robotics Conference, 2017, Seville, in Robot 2017: Third Iberian Robotics Conference, Vol 694 of Advances in Intelligent Systems and Computing, pp. 287-298, Springer.

    Open/Close abstract Abstract Info Info pdf PDF
  • A. Hernandez Ruiz, L. Porzi, S. Rota and F. Moreno-Noguer. 3D CNNs on distance matrices for human action recognition, 25th ACM Conference on Multimedia, 2017, Mountain View, CA. USA., pp. 1087-1095.

    Open/Close abstract Abstract Info Info pdf PDF
  • A. Garrell Zulueta, L.A. Garza, M. Villamizar, F. Herrero and A. Sanfeliu. Aerial social force model: A new framework to accompany people using autonomous flying robots, 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2017, Vancouver, Canada, pp. 7011-7017.

    Open/Close abstract Abstract Info Info pdf PDF
  • L. Porzi, A. Penate-Sanchez, E. Ricci and F. Moreno-Noguer. Depth-aware convolutional neural networks for accurate 3D pose estimation in RGB-D images, 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems, 2017, Vancouver, Canada, pp. 5777-5783.

    Open/Close abstract Abstract Info Info pdf PDF