Formula Derivation and Analysis of the VINS-Mono

TL;DR

This paper derives and analyzes the core equations of VINS-Mono, a robust monocular visual-inertial state estimator suitable for various platforms, highlighting its mathematical foundations and potential applications.

Contribution

It provides a detailed derivation and analysis of the main equations underlying VINS-Mono, enhancing understanding of its mathematical framework.

Findings

Derivation of IMU pre-integration equations

Analysis of visual-inertial co-initialization

Tightly-coupled nonlinear optimization framework

Abstract

The VINS-Mono is a monocular visual-inertial 6 DOF state estimator proposed by Aerial Robotics Group of HKUST in 2017. It can be performed on MAVs, smartphones and many other intelligent platforms. Because of the excellent robustness, accuracy and scalability, it has gained extensive attention worldwide. In this manuscript, the main equations including IMU pre-integration, visual-inertial co-initialization and tightly-coupled nonlinear optimization are derived and analyzed.