Toward GPS-Denied, Multi-Vehicle, Fixed-Wing Cooperative Localization

Gary Ellingson, Brigham Young University

Session 2

Description

This paper describes a vision and proposes a method for multiple, small, fixed-wing aircraft cooperatively localizing in GPS-denied environments. Recent work has focused on the development of a monocular, visualinertial odometry for fixed-wing aircraft that accounts for fixed-wing flight characteristics and sensing requirements. The odometry was developed to be a front-end for novel methodology called relative navigation, which has been developed in prior work. This paper describes how the front-end could enable a back-end where odometry from multiple vehicles and inter-vehicle measurements could be used in a single, global, back-end, graph-based optimization. The inter-vehicle measurements over constrain the graph and allow the optimization to remove accumulated drift for more accurate estimates. The goal of this work is to show that many, small, potentially-lower-cost vehicles could collaboratively localize better than a single, more-accurate, higher-cost GPS-denied system.

 
May 17th, 9:10 AM

Toward GPS-Denied, Multi-Vehicle, Fixed-Wing Cooperative Localization

Orbital ATK Conference Center

This paper describes a vision and proposes a method for multiple, small, fixed-wing aircraft cooperatively localizing in GPS-denied environments. Recent work has focused on the development of a monocular, visualinertial odometry for fixed-wing aircraft that accounts for fixed-wing flight characteristics and sensing requirements. The odometry was developed to be a front-end for novel methodology called relative navigation, which has been developed in prior work. This paper describes how the front-end could enable a back-end where odometry from multiple vehicles and inter-vehicle measurements could be used in a single, global, back-end, graph-based optimization. The inter-vehicle measurements over constrain the graph and allow the optimization to remove accumulated drift for more accurate estimates. The goal of this work is to show that many, small, potentially-lower-cost vehicles could collaboratively localize better than a single, more-accurate, higher-cost GPS-denied system.