Vision-Based Trainable Robotic Arm for Individuals with Motor Disability

Abstract

This paper presents a trainable robotic arm with vision-based guidance capability. The arm can be trained to perform object manipulation tasks such as picking up an object from a location. The vision-based guidance, that utilizes a low-cost integrated webcam, augments the trainable arm's capability to pick up objects; it assists in situations in which the object to be manipulated is displaced from the location the arm was originally trained to pick up. The experimental results from 10 trails demonstrate the vision-based trainable arm's potential to be utilized as a robotic assistant for individuals with physical functioning difficulties.