Methodology and Algorithms for Pedestrian Network Construction

With the advanced capabilities of mobile devices and the success of car navigation systems, interest in pedestrian navigation systems is on the rise. A critical component of any navigation system is a map database which represents a network (e.g., road networks in car navigation systems) and supports key functionality such as map display, geocoding, and routing. Road networks, mainly due to the popularity of car navigation systems, are well defined and publicly available. However, in pedestrian navigation systems, as well as other applications including urban planning and physical activities studies, road networks do not adequately represent the paths that pedestrians usually travel. Currently, there are no techniques to automatically construct pedestrian networks, impeding research and development of applications requiring pedestrian data. This coupled with the increased demand for pedestrian networks is the prime motivation for this dissertation which is focused on development of a methodology and algorithms that can construct pedestrian networks automatically. A methodology, which involves three independent approaches, network buffering (using existing road networks), collaborative mapping (using GPS traces collected by volunteers), and image processing (using high-resolution satellite and laser imageries) was developed. Experiments were conducted to evaluate the pedestrian networks constructed by these approaches with a pedestrian network baseline as a ground truth. The results of the experiments indicate that these three approaches, while differing in complexity and outcome, are viable for automatically constructing pedestrian networks