relative navigation, GPS-denied, GPS-degraded, observability, consistency, navigation framework, state estimation, keyframe, banana distribution, sensor fusion, robocentric, stochastic cloning, multirotor, micro air vehicle, simultaneous localization and mapping, SLAM
State estimation for micro air vehicles (MAVs) often depends heavily on reliable global measurements such as GPS. When global measurements are unavailable, additional sensors, such as cameras or laser scanners, are commonly used to provide measurements of the MAV’s translation and rotation relative to a previously observed keyframe image or scan. With the use of only relative sensors, however, the global position and heading of the vehicle are unobservable and cannot be reliably reconstructed. Many existing approaches work with respect to a global reference frame, resulting in a loss of state observability. This article highlights that unobservability leads to inconsistency and a loss of optimality, which reduces estimation accuracy and robustness of the navigation solution. Relative navigation is presented as an alternative approach that maintains observability by always working with respect to a local coordinate frame. While still subject to global drift, relative navigation is shown through rigorous simulation and hardware validation to produce accurate and consistent state estimates when other approaches break down. By subtly restructuring the state estimation problem to a relative framework, many of the pitfalls prevalent in GPS-denied MAV navigation systems are inherently mitigated.
Original Publication Citation
D. O. Wheeler, D. P. Koch, J. S. Jackson, T. W. McLain and R. W. Beard, "Relative Navigation: A Keyframe-Based Approach for Observable GPS-Degraded Navigation," in IEEE Control Systems, vol. 38, no. 4, pp. 30-48, Aug. 2018, DOI: 10.1109/MCS.2018.2830079
BYU ScholarsArchive Citation
Wheeler, David O.; Koch, Daniel P.; Jackson, James S.; McLain, Timothy W.; and Beard, Randal W., "Relative Navigation: A Keyframe-Based Approach for Observable GPS-Degraded Navigation" (2018). All Faculty Publications. 1961.
Ira A. Fulton College of Engineering and Technology
Electrical and Computer Engineering
(c) 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.
Copyright Use Information