實作整合式室內定位

Abstract

近年來，智慧型手機最熱門的服務不外乎是適時地服務(Location-Based Service, LBS)，利用位置資訊給於相關服務，如導航、定位…等，於是要如何達到精準又快速的定位是LBS的最大要求。GPS室外定位的技術發展趨於成熟、穩定，若將該技術應用到室內，則容易因為環境問題導致訊號衰退，發生位置誤差。i-Loc是一套整合式室內定位系統，結合了行人軌跡系統(Pedestrian Trajectory System, PTS)、Wi-Fi定位系統以及相片定位系統，三個不同系統互相抵消誤差，達到即時又準確的位置回報。Wi-Fi定位容易受到環境遮蔽影響導致訊號衰退，產生定位結果發生飄移誤差。行人軌跡系統可藉由手機內建的感測器來偵測行人的步行資料，彌補掉Wi-Fi訊號所造成的定位問題，然而在長期使用下，感測器容易發生累積誤差，導致定位效果變差。i-Loc使用Kalman Filter結合行人軌跡系統和Wi-Fi定位系統，不僅降低PTS的累積誤差，也可減少因為訊號飄移導致的誤差。在部分區域，Wi-Fi訊號薄弱甚至完全收不到，以及磁力影響較大的地方，可以建置PoI環境來進行相片定位，校正當下的位置誤差。

A great variety of indoor positioning systems for mobile devices have different advantages and limitations. In this work, we propose an integrated indoor positioning system to integrate Wi-Fi positioning, Pedestrian Tracking System (PTS) and Photo positioning with one Kalman filter. By using our single intuitive positioning UI, the median positioning error of PTS+Wi-Fi positioning and Photo positioning are as low as 2.82m and 0.74m respectively. The Photo positioning can provide a high accuracy positioning service by investigating the geometric relations between the images of points of interest (POI) in the photo and their location in the real world based on the principle of photo imaging. A Photo positioning algorithm based on the geometric similarity in photo imaging is presented in this work, and an integrated indoor positioning system is developed for Android smartphone platforms.