Research

Robots for Learning – R4L

Cellulo_windfieldIn 2016, we have launched a series of events aiming to develop the research community working on educational robots (see . I am currently working on two research projects dealing with robots in educational contexts: CoWriter and Cellulo.

References:

  • Özgür A., Johal W., Mondada F and Dillenbourg P.. Haptic-Enabled Handheld Mobile Robots : Design and Analysis.. ACM CHI Conference on Human Factors in Computing System (CHI). April 2017. Denver. USA
  • Özgür A., Johal W., Mondada F and Dillenbourg P.. Windfield : Learning Wind Meteorology with Handheld Haptic Robots.. ACM/IEEE International Conference on Human-Robot Interaction (HRI). March 2017. Vienna. Austria
  • Özgür A., Lemaignan S., Johal W., Mondada F. et al.. Cellulo : Versatile Handheld Robots for Education.. ACM/IEEE International Conference on Human-Robot Interaction (HRI). March 2017. Vienna. Austria
  • F. Mondada, E. Bonnet, S. Davrajh, W. Johal and R. Stopforth. R2T2 : Robotics to Integrate Educational Efforts in South Africa and Europe . accepted in International Journal of Advanced Robotic Systems, 2016.
  • W. Johal, A. D. Jacq, A. Paiva and P. Dillenbourg.. Child-Robot Spatial Arrangement in a Learning by Teaching Activity.. 25th IEEE International Symposium on Robot and Human Interactive Communication. August 2016. New York City, USA.

Projects:

Cellulo, CoWriter  

Robots as social companions

Companion robots for everyday life assistance should be versatile and be able to adapt to the social context. Within the MoCA research project, we defined several social roles for companion robots for children. A survey showed for instance that parents most often wish the robot to play several roles, but they have different opinions concerning which roles. (see on the right) We also explored different embodiment for companion robots, looking at its impact on role attribution.

References:

  • Johal W., Adam C., Fiorino H., Pesty, S., Duhaut D.. Acceptability of a companion robot for children in daily life situations. 5th IEEE Conference on Cognitive InfoCommunications. (CogInfoCom) Nov 2014. Vietri, Italy

Project:

ANR MoCA

Tools:

Questionnaire used for the Survey

Social adaptation of robot’s behaviour

Adaptability to the user and to the context is required for acceptability of social robots. There are various kind of adaption possible:
  • adaptation to the task
  • social adaptation (social context and user)
This adaption can be either triggered by the system itself (self-adaptation, dynamic, non-controlled by the user) or chosen by the user (customization, personalisation). In my research, I focus mainly on social adaptation and in particular in adaptation of the robot’s non-verbal behaviour. During my thesis, I explored the notion of behavioural styles, which enables parent users to choose a social attitude in a given role (social context) for a companion robot for children. Behavioural Styles: In HTML and CSS based webpages, the content (in HTML) is fixed but the look can change by modifying the style (CSS). Similarly we defined BSS (Behavioural Style Sheets) that contain information to modify predefined behaviours.

References:

  • Johal W.. Robots Interacting with Style. In Proceedings of the Tenth Annual ACM/IEEE International Conference on Human-Robot Interaction Extended Abstracts (HRI’15 Extended Abstracts). March 2015. Portland, USA. [pdf]
  • Johal W., Calvary G., Pesty S.. Toward Companion Robots Behaving with Style. In Proceedings of the 2014 IEEE International Symposium on Robot and Human Interactive Communication (ROMAN). Edinburgh, Scotland.[pdf]
  • Johal W., Calvary G. and  Pesty S.. A Robot with Style because you are Worth it!. In Proceeding of ACM CHI Conference on Human Factors in Computing Systems (CHI’14) . Toronto. 2014. Extended Abstract & Poster.[pdf]

Projects:

ANR MoCA, CoWriter

Evaluation of Child-Robot Interaction

References:

  • Vaufreydaz D., Johal W., Combe C. Starting engagement detection toward a companion robot using multimodal features . Robotics and Autonomous Systems, Jan.2015, http://dx.doi.org/10.1016/j.robot.2015.01.004 [pdf].
  • Ta V. C., Johal W.,Portaz M., Castelli E., Vaufreydaz D.. The Grenoble system for the social touch challenge at ICMI 2015. Proceedings of the International Conference on Multimodal Interaction (ICMI). November 2015. Seattle, USA.
  • Johal W., Calvary G. and Pesty S.. A Non-verbal Signals in HRI: Interference in Human Perception. International Conference on Social Robotics. October 2015. Paris, France.
  • Benkaouar (Johal) W., Vaufreydaz D.. Multi-sensors engagement detection with a robot companion in a home environment. Workshop on Assistance and Service Robotics in a Human Environment at IEEE International Conference on Intelligent Robots and Systems (IROS2012), Vilamoura, Algarve – Portugal, oct 2012.[pdf]

Projects:

PRAMAD, ANR MoCA, Cellulo, CoWriter

Affective Thinking for HRI

caio-figWorking on affective reasonning for robots, we proposed CAIO (Cognitive and Affective Interaction Oriented) architecture. Using BDI and logic-based reasonning, we proposed a two loops structure enabling deliberative and reactive cognitive process. Including emotional and dialogue reasonning, this new architecture aims to be modular, domain independent and deployable on several robotics platform.

References:

  • Adam C., Johal W., Pellier D., Fiorino H., and Pesty S.. Social Human-Robot Interaction: A New Cognitive and Affective Interaction-Oriented Architecture.. International Conference on Social Robotics (ICSR). November 2016. Kansas City.
  • Johal W., Pellier D., Adam C., Fiorino H., and Pesty S.. A Cognitive and Affective Architecture for Social Human-Robot Interaction. In Proceedings of the Tenth Annual ACM/IEEE International Conference on Human-Robot Interaction Extended Abstracts (HRI’15 Extended Abstracts). March 2015. Portland, USA. [pdf]

Project:

ANR MoCA

Comments are closed.