Date of Award
2026
Document Type
Thesis
Degree Name
Master of Science in Engineering (MSE)
Department
Mechanical and Aerospace Engineering
Committee Chair
Farbod Fahimi
Committee Member
Chang-kwon Kang
Research Advisor
Howard Chen
Subject(s)
Inertial navigation systems--Design and construction, Multisensor data fusion, Drone aircraft
Abstract
This thesis investigated lever arm effects on distributed inertial measurement unit (IMU) array navigation and evaluated the performance of multiple fusion architectures with applications to uncrewed aerial systems (UAS). The Kalman Filters comprise of three filters: a naive baseline, a filter tracking lever arm as a filter state, and a filter constraining lever arm state estimates around known values. These filters feed into a virtual IMU fusing raw measurements and a federated filter combining output states. Each are evaluated with equal, static, and dynamic weighting based on lever arm and sensor noise. Experiments on indoor and outdoor UAS datasets reveal that lever arm compensation reduced position errors by up to 75%, weighted fusion is essential in vibration-corrupted environments, and federation outperforms virtual IMU fusion outdoors by approximately 30% in position, attributable to local filter buffering of sensor noise. IMU placement is identified as a primary performance driver across all architectures.
Recommended Citation
Neese, Jackson, "Spatially-distributed IMU array navigation : virtual IMU and federated approaches with applications to uncrewed aerial systems" (2026). Theses. 822.
https://louis.uah.edu/uah-theses/822