A Virtual Reality Interface for an Autonomous Spray Painting UAV

Abstract

PaintCopter is an autonomous unmanned aerial vehicle (UAV) capable of spray painting on complex three-dimensional (3D) surfaces. This letter aims to make PaintCopter more user-friendly and to enable more intuitive human–robot interaction. We propose a virtual reality interface that allows the user to immerse in a virtual environment, navigate around the target surface, and paint at desired locations using a virtual spray gun. A realistic paint simulator provides a real-time previsualization of the painting activity that can either be processed right away or stored to a disk for later execution. An efficient optimization based planner uses this information to plan the painting task and execute it. The proposed planner maximizes the paint quality while respecting the spray nozzle constraints and platform dynamics. Our experiments show that the interface allows the user to make precise modifications to the target surface. Finally, we demonstrate the use of virtual reality interface to define a painting mission, and then the PaintCopter carrying out the mission to paint a desired multicolored pattern on a 3D surface.