PhD - Memory and adaptative behaviour for companion robot
BLUE FROG ROBOTICS & BUDDY
Paris, France
il y a 16j

Key Words : Artificial Intelligence, Human-Robot Interaction, Social and Cognitive Robotics, Long-term Interaction, Natural Language, Learning.

DESCRIPTION

Robots were first limited to specific task achievement in the industry. However, robots will be more and more used in human centric-

environments and we will need to interact with them either at the working place or in our household.

It is in this context that Blue Frog Robotics is developing Buddy, the first companion robot intended for the whole family.

In few months, this robot will be among the first one to be deployed in thousands of customers’ house. The Buddy robot, being a companion robot, will not only need to understand its environment, but also to understand the users.

In this context, the robot will need to exhibit social behaviors in a long-term interaction. In daily social interactions, the memory of events (episodic memory) and the ability to link them to different users make humans able to create special bonds with each other in a long-

term relationship. This shared memory has a direct influence on the interacting modalities (e.g. using first name). The perception of others in an agent’s belief state is changing during the relation and leads to a change of the agent’s behavior.

Therefore, this pool of shared memories is a key feature for a robot to enhance its social skills and to bond with its users.

It will also be needed for the robot to detect the contextual cues and to analyze the length of the interactions and their evolution in order to change its behavior and adapt to its environment.

This information could then be shared between the whole set of each Buddy to draw general rules for appropriate of the companion robot’s behavior.

This PhD research work aims to develop a system for the robot’s memory to improve and adapt the behavior of the robot to its different users he is interacting with.

Moreover, it is also expected the development of an off-line system capable of analyzing collected data by each Buddy in order to improve the learning system on temporal and contextual factors which leads to the adaptation of the robot’s behavior.

An estimation of the user satisfaction upon the robot’s behavior adaptation can be used to evaluate the process. This PhD is a shared position between Blue Frog Robotics and the Autonomous Systems and Robotics Lab at ENSTA ParisTech.

REQUIRED QUALIFICATIONS

nbsp;Relevant degree in Computer Science, or related area,

nbsp;Strong programming skills in object oriented (C#, C++ or Java)

  • Proficient oral and written skills in English
  • Basic oral skills in French
  • IT WOULD BE A PLUS

  • Experience with Unity3D / Android
  • Knowledge in C / Matlab
  • Experience with Data management (database, bigdata )
  • TO APPLY

    Contact Dr. Grégoire Milliez (gm bluefrogrobotics.com) with a single PDF file including :

  • Cover Letter describing background, experience, and motivation,
  • CV including references + transcripts of the applicant’s bachelor and / or master studies.
  • Closing date for applications : until the position is filled.

    Postuler
    Postuler
    Mon email
    En cliquant sur « Continuer », je consens au traitement de mes données et à recevoir des alertes email, tel que détaillé dans la Politique de confidentialité de neuvoo. Je peux retirer mon consentement ou me désinscrire à tout moment.
    Continuer
    Formulaire de candidature