Visual-Inertial Teach & Repeat for Aerial Robot Navigation

Abstract

This paper presents a Teach & Repeat (T&R) algorithm from stereo and inertial data, targeting Unmanned Aerial Vehicles with limited on-board computational resources. We propose a tightly-coupled, relative formulation of the visual-inertial constraints that fits the T&R application. In order to achieve real-time operation on limited hardware, we constraint it to motion-only visual-inertial Bundle Adjustment and solve for the minimal set of states. For the repeat phase, we show how to generate a trajectory and smoothly follow it with a constantly changing reference frame. The proposed method is validated with the sequences of the EuRoC dataset as well as within a simulated environment, running on a standard laptop PC and on a low-cost Odroid X-U4 computer.