Development and Functional Analysis for the Generalized Feeding Robot

Thumbnail Image
Mashrur, Tanzim
Journal Title
Journal ISSN
Volume Title
University of Guelph

A personal care robotic system has been developed that is able to provide feeding assistance to those suffering from upper limb impairment. The system introduces a novel approach for feeding that prioritized two ideas; generalized functionality to encompass multiple feeding tasks and seamless interaction with the user. A heavy emphasis was placed on these ideas because it maximized the chances of commercial success. Additionally, the system leveraged the latest advances in computer vision to incorporate functionality that was not seen within the literature. For the functional prototype, the system was comprised of an off the shelf 6-DOF robotic manipulator, cameras, and actuator. Furthermore, various tools used during operation were 3D printed. The system software has 3 main phases of operation; identification, acquisition, and delivery. One of the novel features of this system is that instead of attempting to identify the food, the robot will identify the method required for acquiring the food instead. From testing and validation, the system had minimal identification error, high success rates for acquisition and delivery, and a fast safety response time.

Rehabilitation Robotics, Feeding Robot, Computer Vision, Machine Learning