ROJul 6, 2017

Embodied Flight with a Drone

arXiv:1707.01788v130 citations
Originality Incremental advance
AI Analysis

This work addresses the problem of improving human-robot interaction for drone control, particularly for unskilled users, though it is incremental as it builds on existing flight simulator technology.

The paper tackles the challenge of controlling non-anthropomorphic robots like drones by developing an embodied interaction system that maps human body movements to drone flight, resulting in a more natural and immersive experience for unskilled users compared to conventional remote controllers.

Most human-robot interfaces, such as joysticks and keyboards, require training and constant cognitive effort and provide a limited degree of awareness of the robots state and its environment. Embodied interactions, instead of interfaces, could bridge the gap between humans and robots, allowing humans to naturally perceive and act through a distal robotic body. Establishing an embodied interaction and mapping human movements and a non-anthropomorphic robot is particularly challenging. In this paper, we describe a natural and immersive embodied interaction that allows users to control and experience drone flight with their own bodies. The setup uses a commercial flight simulator that tracks hand movements and provides haptic and visual feedback. The paper discusses how to integrate the simulator with a real drone, how to map body movement with drone motion, and how the resulting embodied interaction provides a more natural and immersive flight experience to unskilled users with respect to a conventional RC remote controller.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes