| Summary: | 碩士 === 國立交通大學 === 電機資訊國際學程 === 103 === Digital maps play an increasingly important role in a wide range of applications such as digital navigation systems, web mapping services, e-commerce, social media, education and several others. Similarly, representations of maps can vary from simple printed evacuation maps at the entrance of buildings to powerful web mapping services. Maps can be classified depending on the zones they represent, i.e., pedestrian zones or car zones, indoor maps or outdoor maps. This distinction is relevant due to the fact that different methodologies and technologies are employed depending on the zones these maps represent. Despite the wide adoption of digital maps, indoor mapping still represent a significant gap. With no clear standards or leading providers for indoor environments, it is a big opportunity for innovation that is taken as motivation for this research. Another important factor to the increasing adoption of digital maps is the wide usage of smartphones, equipped with a variety of sensors such as accelerometer, magnetometer, gyroscope and others. In order to contribute to a faster adoption of indoor maps, this research propose a new methodology to create digital maps for pedestrian areas using smartphones as the main tool for this purpose. To achieve this goal we start by implementing a localization and tracking system based on the dead reckoning process and employing the accelerometer and magnetometer embedded in the smartphone to collect the required data. Accelerometer data is used to calculate the displacement of the user whereas magnetometer data is used to calculate the heading direction. Employing this localization and tracking system we then record all the paths walked by the user while holding the phone. Finally we enable the user to add relevant points of interest throughout its walk, completing with this a meaningful representation of the area visited by the user. The current research was conducted by collecting accelerometer and magnetometer data under the desired settings, followed by a thorough analysis of this information in order to identify traits and potential issues. As a result an algorithm was developed, capable of accurately measure displacement and heading direction.
|
|---|
