Multi-Agent Systems & Social Robotics


FMI, Sofia University “St. Kliment Ohridski”

Elective course

Instructor: Trayan Iliev

Internet of Things (IoT) and Social Robotics are regarded by many as the next major technological challenge – a wave of change comparable in effect to emergence of Internet itself.

IoT is emerging as fusion of several technological tendencies in their inter-dependence:

  • Internet of Things & Mobility
  • Social Robotics
  • Big Data, Analytics and Cloud Computing
  • Web Services & SOA Infrastructure

Course Multi-Agent Systems & Social Robotics presents some of the above tendencies highlighting the areas of intelligent actors / agents / appliances and social robotics.Social robots are interacting and communicating with people and/or other physical agents by following social rules and behaviors in accordance with the social role played by the robot. The social robots should be physically embodied in order to be able to participate in social interactions.

Kaspar [1] developed by University of Hertfordshire is an example of social robot. He helps to children with autism disease to learn social behaviors using interactive games.

Project Goals

The main goals of this project are:

  1. Presentation of latest development and technologies in the areas of Multi-Agent Systems (MAS), Internet of Things (IoT) and Social Robotics:
    1. Three major robotics paradigms – deliberative, behavioral and hybrid
    2. Representative robotic architectures – STRIPS, Subsumption architecture, AIMA agents, INTERRAP
    3. Actor model for building resilient, concurrent, high-performance, extensible and distributed applications – Akka™ framework
    4. Belief–Desire–Intention (BDI) model of practical human-like reasoning – an event driven executable model of reactive and proactive behavior and Agent Communication Languages (ACLs)
    5. Knowledge representation formats and W3C open standards – rules, frames, semantic nets, Resource Description Framework (RDF), Web Ontology Language (OWL)
    6. Internet of Things (IoT), Service Oriented Architecture (SOA) & Event-driven architecture (EDA)
  2. Practical demonstration of interactive control and remote supervision of small robots, providing first-hand experience in discussed technologies;
  3. To facilitate stydy of available hardware and software platforms for implementation of small do-it-yourself (social) robots.

Fig.1: LeJaRo – the ultimate Lego® Java™* Robot

Used Technologies

Two main robotics hardware platforms will be used during the project:

However the course participants are free to use other embedded platforms as well. Particularly recommended are following platforms:

Project web robot control application will be implemented using following technologies: HTML 5, CSS 3 & HTML 5 JavaScript APIs – Canvas, Drag-and-Drop, Audio, Video, Geolocation, Local & Session Storage, Web Sosket, REST & AJAX, jQuery, Bootstrap/jQuery Mobile, Angular JS, JAX-RS Java REST API/ Node.js REST services.

Resources

  1. Byrne, P., Robot at Hertfordshire University aids autistic children (2011), BBC – http://www.bbc.co.uk/news/uk-england-beds-bucks-herts-16261057
  2. IPT Social Robotics & Internet of Things (IoT) page – http://iproduct.org/en/social-robotics-iot/
  3. Blockly JS library official website – https://developers.google.com/blockly/