Ultrasound-based Indoor Collaborative Pedestrian Dead Reckoning System

• Main Authors: Ching-Chung Chen, 陳敬忠
• Other Authors: Min-Te Sun
• Format: Others
• Language: en_US
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/z235gc

碩士 === 國立中央大學 === 資訊工程學系 === 104 === Indoor localization has become a popular issue in recent years. Most of the indoor localization approaches either require the availability of an infrastructure or the additional training eorts. The traditional pedestrian dead reckoning (PDR) system can be implemented on mobile devices without additional cost and training. However, it accumulates errors quickly and leads to unacceptable results after a short period of time. To address this issue, we propose the ultrasound-based collaborative pedestrian dead reckoning (UCPDR) system. The main idea of UCPDR is to reduce the estimation error of a pedestrian's location by two steps: First,UCPDR estimates the pedestrian's location based on another nearby pedestrian's location information obtained by PDR and the distance information obtained by ultrasound signals exchanged by the two pedestrians; second, reduce the error estimation through the opportunistic Kalman lter by using the previous location estimation as a new measurement. In addition, the backward correction scheme is used to improve the accuracy of user's trajectory. To evaluate feasibility of UCPDR, a prototype is built on the iOS (iPhone Operating System) platform. The conducted experiment results show that UCPDR is able to limit the localization error within 2m (when number of steps less than 80) after a long period of time through the help of neighbors location information and ultrasound distance information. Our UCPDR system always achieves a better localization accuracy than the traditional PDR system.