研究業績リスト
その他
Research and development of cloud-based pedestrian positioning system
作成日時 01/04/2019–31/03/2022
Offer Organization: Japan Society for the Promotion of Science, System Name: Grants-in-Aid for Scientific Research, Category: Grant-in-Aid for Scientific Research (C), Fund Type: -, Overall Grant Amount: - (direct: 3400000, indirect: 1020000)
In urban canyons, pedestrian positioning precision may be significantly degraded by the obstruction and reflection of roadside buildings. In our work, this problem is solved from two aspects. (i) When the line-of-sight (LoS) wave from a satellite is not received, but a reflected wave is used instead for positioning, the multipath error occurs. By exploiting the temporal/spatial correlation of multipath errors, the multipath error at a pedestrian is estimated and corrected. (ii) When the number of satellites is insufficient for positioning, vehicles near a pedestrian are used as anchors, and the spatial and temporal variation of channel state information is used to construct a virtual 2D antenna array and used to accurately measure the pedestrian-vehicle angle, which helps to greatly improve the pedestrian positioning precision.
その他
作成日時 01/04/2016–31/03/2019
Offer Organization: Japan Society for the Promotion of Science, System Name: Grants-in-Aid for Scientific Research, Category: Grant-in-Aid for Scientific Research (C), Fund Type: -, Overall Grant Amount: - (direct: 3500000, indirect: 1050000)
A pedestrian positioning method is studied, in which pedestrians use nearby vehicles and roadside units as anchors, besides GPS satellites.
The channel state information between vehicles/roadside-units and a pedestrian is obtained, and a non-linear regression function is learned for distance estimation. In the position computing, weighting is performed by considering the error characteristics of distances. The impact of time resolution of the receiver is also evaluated. In addition, we proposed a method for high precision pedestrian positioning by fusing the pedestrian's own moving speed, the distance between the pedestrian and vehicles, roadside-units, satellites by using a Kalman filter. Ray-tracing simulations have shown that in urban environments, the proposed method can significantly reduce the positioning error, compared to the present GPS positioning techniques.