Robotic ADaptation to Humans Adapting to Robots: Overview of the FP7 project RADHAR

Abstract

This paper presents the research objectives and current state of the semi-autonomous navigation components that are being developed in the FP7 project RADHAR (www.radhar.eu). RADHAR proposes a framework to fuse the inherently uncertain information from both environment perception and a wheelchair driver's steering signals by estimating the trajectory the wheelchair should execute in that environment, and to adopt this fused information for providing safe navigation assistance. In order to provide intuitive, bilateral navigation assistance a haptic joystick prototype was developed and control algorithms were developed and tested for this haptic joystick. The key components for this semi-autonomous navigation assistance in the RADHAR framework are explained and experimental results are shown.