Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Rubicon aal testbed erf workshop rovereto 2014

613 views

Published on

Published in: Science, Technology, Business
  • Be the first to comment

  • Be the first to like this

Rubicon aal testbed erf workshop rovereto 2014

  1. 1. RUBICON is a three-year project, funded by the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement No.269914 AAL Testbed Arantxa Renteria (TECNALIA) – Mauro Dragone (Univ. College Dublin) European Robotics Forum, Rovereto. March 12-14th, 2014
  2. 2. RUBICON Goals Rubicon minimizes the need for programming and human supervision of robotic ecologies The Rubicon ecology is able to adapt to changes of its application requirements and to its environment
  3. 3. High-Level RUBICON Architecture Control Layer Cognitive Layer LearningLayer (Supervision) Wiring, Feedback, Training Executionstatus Goals (Env / System) Events, State Sensordata Actions Sensordata Messages (Env / System) Events, State, confidence (Novelty Detection) Feedback,Training Sensing and Actuation (WSN/Robots/Human Input) Action Quality Communication Layer 3
  4. 4. AAL Scenario (Homelab at Tecnalia) Actuators for blind, front door, TV on/off & channel selection
  5. 5. AAL Scenario (Homelab at Tecnalia) Turtlebot moving around, interacting with user, capturing data Sound recognition: door bell, faucet, microwave oven, music, TV…) Sensors on users (acceleremoter, localization)
  6. 6. Data acquisition • PDA app to label activities • Video recording of experiments • Scripts of daily activities: preparing lunch, setting up table, eating, resting, sleeping, listen to music…
  7. 7. Results Achieved • Implementation a distributed general-purpose learning system for robotic ecologies – Adaptive and reconfigurable – Incremental learning – Designed for computationally constrained devices • Experimental assessment on real-world Rubicon scenarios – Ambient assisted living (AAL) – Localization by Received Signal Strength (RSS) • Learning to adapt to environmental/user changes – Distributed intelligence for adaptive configuration planning – Innovative application through Control-Learning Layer integration 7
  8. 8. Results of AAL Testbed 8
  9. 9. Novelty of work • RUBICON robot ecology applied to a domotic environment • Example of AAL services: supporting activities (rehabilitation…), switching on/off of devices, house configuration (open/close blinds…) • Significant data coming from sensors (6 GB), labeled using a PDA based app. Made publicly available for research purposes
  10. 10. RUBICON AAL Testbed Thank you!!

×