170 / 2023-08-30 23:37:29

A Robust Particle Filtering-Based UWB Positioning Method for Indoor Complicated Environment

Non-Line-of-Sight (NLOS); particle filtering; particle impoverishment; Ultrawide-Band (UWB); indoor positioning; genetic algorithm

The signal Non-Line-of-Sight (NLOS) propagation caused by the complicated indoor environments is the main reason for accuracy degradation of the Ultra-Wide Band (UWB)-based indoor positioning. Mitigating the effects of NLOS ranges on positioning is important for positioning accuracy improvement. In this paper, a robust particle filtering (PF)-based positioning algorithm is proposed to address this problem. The proposed algorithm consists of two phases, rough positioning and accurate positioning. In the rough positioning phase, an improved residual weighting algorithm (IRwgh) is proposed to obtain the target rough position, and the positioning result is used for NLOS identification and range reconstruction with the aid of the error function. In the accurate positioning phase, the identified NLOS ranges are used for constrained particle sampling to achieve accurate position estimation. Meanwhile, regarding the particle impoverishment problem in the traditional resampling step, a genetic algorithm-based particle distribution optimization strategy is implemented in the algorithm to enhance the particle diversity. The experimental test result shows that under the indoor NLOS environment, the proposed robust PF algorithm not only can obviously improve the positioning accuracy but also can meet the requirement of the real-time applications. Compared with the standard PF algorithm, the robust PF algorithm improves the positioning accuracy by about 36.5%.