Abstract: A vision-aided inertial navigation system determines navigation solutions for a traveling vehicle. A navigation solution module analyzes image sensor poses and estimated inertial navigation solutions to produce a time sequence of system navigation solution outputs that represent changing locations of the traveling vehicle. A loop closure detector performs a two-threshold image classification to classify an incoming navigation image as novel, loop-closing, or neither, and the navigation solution module resets the navigation solution output to represent a previous location when the loop closure detector classifies an incoming navigation image as loop-closing.

Abstract: A navigation system provides pose, i.e., location and orientation, solutions using best available location information from two or more location systems. One of the location systems is a fiducial-based location system, which is accurate when a sufficient number of fiducials is recognized. However, when an insufficient number of fiducials is recognized, an odometry-based location system is used. Although the odometry-based location system is subject to drift, when a sufficient number of fiducials is recognized, the fiducial-based location system is used to correct the odometry-based location system. The navigation system provides robust, accurate and timely pose solutions, such as for augmented reality (AR) or virtual reality (VR) systems, without the time-consuming requirement to establish and localize many fiducials or the computational and memory requirements of pure fiducial-based location systems, and without the inherent drift of pure odometry-based location systems.

Abstract: A vision-aided inertial navigation system determines navigation solutions for a traveling vehicle. A navigation solution module analyzes image sensor poses and estimated inertial navigation solutions to produce a time sequence of system navigation solution outputs that represent changing locations of the traveling vehicle. A loop closure detector performs a two-threshold image classification to classify an incoming navigation image as novel, loop-closing, or neither, and the navigation solution module resets the navigation solution output to represent a previous location when the loop closure detector classifies an incoming navigation image as loop-closing.

Abstract: A vision-aided inertial navigation system determines navigation solutions for a traveling vehicle. A feature tracking module performs optical flow analysis of the navigation images based on: detecting at least one image feature patch within a given navigation image comprising a plurality of adjacent image pixels corresponding to a distinctive visual feature, calculating a feature track for the at least one image feature patch across a plurality of subsequent navigation images based on calculating a tracking residual, and rejecting any feature track having a tracking residual greater than a feature tracking threshold criterion that varies over time with changes in quantifiable characteristics of the at least one feature patch. A multi-state constraint Kalman filter (MSCKF) analyzes the navigation images and the unrejected feature tracks to produce a time sequence of estimated image sensor poses characterizing estimated position and orientation of the image sensor for each navigation image.