Opt Express. 2023 Feb 27;31(5):7742-7752. doi: 10.1364/OE.469104.
Visible light positioning (VLP) can provide high positioning accuracy based on existing lighting facilities, which gives it great potential for intelligent transportation system (ITS). However, in real scenarios, the performance of visible light positioning is limited by outage problem caused by the sparse distribution of light emitting diode (LED) and time cost of the positioning algorithm. In this paper, a particle filter (PF) assisted single LED VLP (SL-VLP) and inertial fusion positioning scheme is proposed and experimentally demonstrated. It improves the robustness of VLP in sparse LED scenarios. In addition, the time cost and the positioning accuracy at different outage rates and speeds are investigated. The experimental results show that, by using the proposed vehicle positioning scheme, it can achieve the mean positioning error is 0.09 m, 0.11 m, 0.15 m and 0.18 m at 0, 5.5%, 11% and 22% of SL-VLP outage rate, respectively.
可见光定位（VLP）可以利用现有的照明设施提供高精度的定位，这使其在智能交通系统（ITS）中具有很大的潜力。然而，在实际场景中，由于发光二极管（LED）的稀疏分布和定位算法的时间成本，可见光定位的性能受到限制。在本文中，提出了一种基于粒子滤波（PF）辅助的单个 LED VLP（SL-VLP）和惯性融合定位方案，并进行了实验验证。该方案提高了 VLP 在稀疏 LED 场景中的鲁棒性。此外，还研究了不同中断率和速度下的时间成本和定位精度。实验结果表明，采用所提出的车辆定位方案，在 SL-VLP 中断率分别为 0、5.5%、11%和 22%的情况下，平均定位误差分别为 0.09m、0.11m、0.15m 和 0.18m。
