Robotic Healthcare Teleoperation of a Vision-guided Virtual Tunneling Control for Oral Feeding Assistance
Elizabeth Kirwan
School of Engineering
Faculty Supervisor: David Quintero
Assistive robotics have a wide range of applications in the medical field, particularly in rehabilitation and telemedicine. In the field of telemedicine, assistive robotics can perform some of the tasks traditionally done by nurses in person, such as medication management, feeding, and hygiene. They can provide an alternative for those who cannot afford the cost of a full-time at-home nurse. With teleoperation, the burden of traveling to a patient's home is alleviated for nurses, making at-home care more accessible. An intuitive shared control system for a teleoperated robotic arm will be developed. The control system will feature an autonomous mode and a semi-autonomous mode that will allow a nurse to bring objects, such as food, close to a patient's mouth. The semi-autonomous mode will use tunneling control with a variable width that adjusts based on the estimated distance between the robotic gripper and its target location, which will be tracked using computer vision. The robot's movement will be constrained by the width of the tunnel. As the gripper approaches its desired location, the radius of movement becomes progressively more restricted, allowing precise and controlled interaction with the patient. This system will improve operator performance compared to a manually operated telerobot.