A UAV-Assisted Search and Localization Strategy in Non-Line-of-Sight Scenarios.

Bingbing Yuan,Ruisi He,Bo Ai,Ruifeng Chen,Gongpu Wang,Jianwen Ding,Zhangdui Zhong
DOI: https://doi.org/10.1109/jiot.2022.3188557
IF: 10.6
2022-01-01
IEEE Internet of Things Journal
Abstract:Recently, unmanned aerial vehicle (UAV)-assisted ground targets localization is widely used in search and rescue (SAR) scenes. In this article, we propose a UAV search and localization strategy, which can search and locate unknown number of victims. The proposed strategy uses the UAV as a mobile anchor to measure time of arrival (TOA) and can effectively mitigate localization errors caused by non-line-of-sight (NLOS) propagation. Generally speaking, the strategy is divided into two parts: 1) trajectory planning and 2) localization. By selecting waypoints and planning a suitable search trajectory, UAV can obtain all the measurement information in deployment area and ensure that each target has at least three anchors for localization even in the NLOS propagation environment. In the localization stage, a new estimator is proposed based on maximum-likelihood estimation (MLE), which estimates average NLOS bias together with the coordinates of target and can be well solved by the particle swarm algorithm. The simulation results show that the proposed strategy outperforms for mitigating NLOS error compared with other methods, and can effectively ensure high localization accuracy in the NLOS propagation environment.
What problem does this paper attempt to address?