Tightly Coupled UWB-INS Positioning With Passive Synchronization Using Continuous Clock Phase Tracking
This work evaluates the effectiveness of a novel one-way time of flight-based passive ultrawideband (UWB) inertial navigation solution for indoor positioning of a mobile platform and formulates a design criterion for choosing between passive systems through noise analysis. The proposed tightly coupled time of arrival (TC-TOA) approach achieves passive synchronization by keeping track of the local clock and its derivatives as states. Using these states and position estimate, the reception timestamps of the network messages are predicted. The errors in these predicted timestamps are then used to update both clock states and the position states through accurate modeling of coupled interactions. The design results in a tightly coupled estimator formulation achieved using an error state Kalman filter with right quaternion error parameterization. The proposed method is evaluated using a MATLAB simulation environment and on a dataset acquired by flying a quadcopter in an indoor environment, with ground truth obtained from a motion capture system. Decawave DWM 1000-DEV hardware with custom firmware was used in the measurement acquisition process. Simulation results demonstrate that the proposed tightly coupled system outperforms time difference of arrival (TDOA)-based methods, especially when the network transmission gap becomes large or in the presence of communication interruptions, which can happen in large-scale networks. Experimental validation resulted in RMS-position errors around 25 cm, with increasing differences between the TC-TOA versus TDOA methods as the number of anchors drops. Furthermore, the proposed method can accommodate measurement updates even when the connection with the network is interrupted down to one anchor.
Authors
- Nushen M. Senevirathna (ORCID: https://orcid.org/0000-0002-7937-777X)
- Raymond G. Gosine (ORCID: https://orcid.org/0000-0002-8990-115X)
- Oscar De Silva (ORCID: https://orcid.org/0000-0002-6255-5728)
- George K. I. Mann (ORCID: https://orcid.org/0000-0002-1211-3374)
Institutions
- Memorial University of Newfoundland (CA)
Publication Details
- Journal
- IEEE Internet of Things Journal
- Published
- 2024-11-29
- DOI
- https://doi.org/10.1109/jiot.2024.3509352
- Citations
- 2
- Primary Topic
- Ultra-Wideband Communications Technology
- Type
- article
- Field-Weighted Citation Impact
- 0.42
Funders
- Natural Sciences and Engineering Research Council of Canada