Robotic systems have traditionally been controlled by handheld interfaces such as game controllers, joysticks, mice, and keyboards. While these interfaces are highly effective, there are situations where an operator may be unable to use their hands. The proposed method explores the use of an Intel RealSense depth sensing camera as a robotic teleoperation input method by tracking the operator's eye-gaze to the computer screen. It utilizes the eye-gaze vector and body position to determine operator intent and provide input to a robotic control system.

• This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.

Creator

• Hollander, Nicholas H.

Publisher

• Worcester Polytechnic Institute

Identifier

• 22121
• E-project-050521-015009

Keyword

• Python
• RGB+D
• Robotics
• Teleoperation
• Machine Learning
• Computer Vision

Advisor

• Li, Zhi

Year

• 2021

Date created

• 2021-05-05

Resource type

• Major Qualifying Project

Major

• Robotics Engineering

Rights statement

• In Copyright