PROBE: Predictive Robust Estimation for Visual-Inertial Navigation

TL;DR

PROBE introduces a learning-based approach to weight visual features dynamically in a visual-inertial navigation system, significantly improving localization accuracy in challenging environments by adjusting feature influence based on predicted quality.

Contribution

The paper presents a novel method that learns to predict feature quality and adaptively weights observations in VINS, enhancing robustness and accuracy in complex scenarios.

Findings

Reduced localization error on KITTI dataset

Improved accuracy in indoor and outdoor driving tests

Effective feature quality prediction for robust navigation

Abstract

Navigation in unknown, chaotic environments continues to present a significant challenge for the robotics community. Lighting changes, self-similar textures, motion blur, and moving objects are all considerable stumbling blocks for state-of-the-art vision-based navigation algorithms. In this paper we present a novel technique for improving localization accuracy within a visual-inertial navigation system (VINS). We make use of training data to learn a model for the quality of visual features with respect to localization error in a given environment. This model maps each visual observation from a predefined prediction space of visual-inertial predictors onto a scalar weight, which is then used to scale the observation covariance matrix. In this way, our model can adjust the influence of each observation according to its quality. We discuss our choice of predictors and report substantial reductions in localization error on 4 km of data from the KITTI dataset, as well as on experimental datasets consisting of 700 m of indoor and outdoor driving on a small ground rover equipped with a Skybotix VI-Sensor.