Visual-Inertial SLAM as Simple as A, B, VINS

TL;DR

AB-VINS introduces a deep learning-based visual-inertial SLAM system that estimates depth parameters and uses a novel memory tree structure, achieving high efficiency and robustness over traditional methods.

Contribution

It proposes a new SLAM approach using neural networks for depth estimation and a memory tree data structure for efficient loop closure handling.

Findings

Front-end motion tracking surpasses state-of-the-art filtering methods in efficiency.

Loop closure integration affects only a constant number of variables.

System demonstrates higher robustness despite slightly lower accuracy.

Abstract

We present AB-VINS, a different kind of visual-inertial SLAM system. Unlike most popular VINS methods which only use hand-crafted techniques, AB-VINS makes use of three different deep neural networks. Instead of estimating sparse feature positions, AB-VINS only estimates the scale and bias parameters (a and b) of monocular depth maps, as well as other terms to correct the depth using multi-view information, which results in a compressed feature state. Despite being an optimization-based system, the front-end motion tracking thread of AB-VINS surpasses the efficiency of a state-of-the-art filtering-based method while also providing dense depth. When performing loop closures, standard keyframe-based SLAM systems need to relinearize a number of variables which is linear with respect to the number of keyframes. In contrast, the proposed AB-VINS can incorporate loop closures while only affecting a constant number of variables. This is thanks to a novel data structure called the memory tree, where keyframe poses are defined relative to each other rather than all in one global frame, allowing for all but a few states to be fixed. While AB-VINS might not be as accurate as state-of-the-art VINS algorithms, it is shown to be more robust.