Cooperative Navigation within UAV Swarms via Centralized-Unscented Kalman Filtering to Reduce C-SWAP Requirements

Recent advancements in autonomous systems have spurred research in cooperative navigation. Inspired by the maneuvering abilities of a marching band, the unmanned aerial vehicles (UAVs) are equipped with a sensor array system comprising of an INS, GPS, and ranging device, ensuring comprehensive observability in both absolute and relative positioning. This paper adopts an error-space centralized-unscented Kalman filter architecture for swarm communication to estimate positions, velocities, and attitudes (PVAs), in order to enhance the overall positioning of a swarm of UAVs. The proposed algorithm is evaluated through Monte Carlo simulations, demonstrating improved swarm performance whilst traveling along a figure-eight trajectory while maintaining a pyramidal configuration. The study assesses individual and collaborative performance, revealing potential benefits from cooperative aiding that may reduce the cost, size, weight, and power (C-SWAP) requirements of the overall swarm by utilizing a leader to enhance the relative distancing and absolute positioning of its followers. The presented algorithm shows promise for reducing the C-SWAP requirements for real-world applications in swarm navigation.