We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
An Indoor Localization Algorithm of UWB and INS Fusion based on Hypothesis Testing.
- Authors
Long Cheng; Yuanyuan Shi; Chen Cui; Yuqing Zhou
- Abstract
With the rapid development of information technology, people's demands on precise indoor positioning are increasing. Wireless sensor network, as the most commonly used indoor positioning sensor, performs a vital part for precise indoor positioning. However, in indoor positioning, obstacles and other uncontrollable factors make the localization precision not very accurate. Ultra-wide band (UWB) can achieve high precision centimeter-level positioning capability. Inertial navigation system (INS), which is a totally independent system of guidance, has high positioning accuracy. The combination of UWB and INS can not only decrease the impact of non-line-of-sight (NLOS) on localization, but also solve the accumulated error problem of inertial navigation system. In the paper, a fused UWB and INS positioning method is presented. The UWB data is firstly clustered using the Fuzzy C-means (FCM). And the Z hypothesis testing is proposed to determine whether there is a NLOS distance on a link where a beacon node is located. If there is, then the beacon node is removed, and conversely used to localize the mobile node using Least Squares localization. When the number of remaining beacon nodes is less than three, a robust extended Kalman filter with M-estimation would be utilized for localizing mobile nodes. The UWB is merged with the INS data by using the extended Kalman filter to acquire the final location estimate. Simulation and experimental results indicate that the proposed method has superior localization precision in comparison with the current algorithms.
- Subjects
INERTIAL navigation systems; WIRELESS sensor networks; INFORMATION technology; KALMAN filtering; POSITION sensors
- Publication
KSII Transactions on Internet & Information Systems, 2024, Vol 18, Issue 5, p1317
- ISSN
1976-7277
- Publication type
Article
- DOI
10.3837/tiis.2024.05.010