Care-O-bot II—Development of a Next Generation Robotic Home Assistant

  • Authors:

  • Birgit Graf;Matthias Hans;Rolf D. Schraft

  • Affiliations:

  • Fraunhofer Institute Manufacturing Engineering and Automation (IPA), Nobelstr. 12, D-70569 Stuttgart, Germany. birgit.graf@ipa.fraunhofer.de www.care-o-bot.de;Fraunhofer Institute Manufacturing Engineering and Automation (IPA), Nobelstr. 12, D-70569 Stuttgart, Germany. www.care-o-bot.de;Fraunhofer Institute Manufacturing Engineering and Automation (IPA), Nobelstr. 12, D-70569 Stuttgart, Germany. www.care-o-bot.de

  • Venue:
  • Autonomous Robots
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

Technical aids allow elderly and handicapped people to live independently and supported in their private homes for a longer time. As a contribution to such technological solutions, two demonstrator platforms for a robotic home assistant—called Care-O-bot—were designed and implemented at Fraunhofer IPA, Stuttgart. Whereas Care-O-bot I is only a mobile platform with a touch screen, Care-O-bot II is additionally equipped with adjustable walking supporters and a manipulator arm. It has the capability to navigate autonomously in indoor environments, be used as an intelligent walking support, and execute manipulation tasks. The control software of Care-O-bot II runs on two industrial PCs and a hand-held control panel. The walking aid module is based on sensors in the walking aid handles and on a dynamic model of conventional walking aids. In “direct mode”, the user can move along freely with the robot whereas obstacles are detected and avoided. In “planned mode”, he can specify a target and be lead there by the robotic assistant. Autonomous planning and execution of complex manipulation tasks is based on a symbolic planner and environmental information provided in a database. The user input (graphical and speech input) is transferred to the task planner and adequate actions to solve the task (sequence of motion and manipulation commands) are created. A new method for sensor based manipulation using a tilting laser scanner and camera integrated in the head of the robot has been implemented. Additional sensors in the robot hand increase the grasping capabilities. The walking aid has been tested with elderly users from an assisted living facility and a nursery home. Furthermore, the execution of fetch and carry tasks has been implemented and tested in a sample home environment.