Voice-Controlled Vision-based Semi-Autonomous Assistive Robotic Manipulation Assistance

Abstract

People with severe physical disabilities have found it difficult or impossible to independently use assistive robotic manipulators (ARM) due to their lack of access to the conventional control methods and the cognitive/physical workload associated with operating the ARMs. This study aims to develop and evaluate an alternative control method for ARMs that enables these individuals to independently and efficiently operate such devices. This method termed as AROMA (assistive robotic manipulation assistance) automatically recognizes user speech and intelligently plans a path/motion to the object of interest using a low-cost 3D depth-sensing camera.