Manipulation Assist for Teleoperation in VR

Christian Michael Barentine, Andrew McNay, Ryan Pfaffenbichler, Addyson Smith, Eric Rosen, Elizabeth Phillips

Keywords: Virtual Reality, Robot, Predictive Control

TL;DR: Description of a system to let robots perform fine grain motor tasks while their operators specify high level tasks.

Abstract: VR interfaces are promising for robotics for several reasons, including that they may be suitable for resolving many of the human performance issues associated with traditional robot teleoperation interfaces used for robot manipulation. In this systems-focused paper, we introduce and document the development of a VR-based robot control paradigm with manipulation assist control algorithm, which allows human operators to specify larger manipulation goals while leaving the low-level details of positioning, manipulation and grasping to the robot itself. For the community, we also describe system design challenges to our progress thus far.

Track: extended abstract

Broader Impacts: VR interfaces are promising for allowing users to more effectively teleoperate robots, especially for manipulation tasks. In this paper, we presented system details for a VR-based robot control system with manipulation assist, which allows users to specify high level manipulation goals while leaving the low-level goals to the robot itself. For the community, we detail the individual components of the system (hardware, software, and algorithm details) and progress and challenges to its development to date, as well as planned future development details. By laying the foundations for effective Manipulation assist, we aim to further enable humans and robot to specialize in areas that they are singularly effective in. By combining Robotic capacity for precise manipulation with human planning, we hope to impact the approach taken towards teleoperation in the future.