Algorithmic development of an optimal path computation model based on topographic map features

Abstract

Navigation applications have been at the heart of much GIS network analysis research. Path finding problems have attracted widespread research interests with different applications such as Logistics applications, Infrastructure Planning and Travel Demand analysis. To support such applications, specific data requirements are required. The conventional arc-node model is commonly and widely utilized to model the complexity of network elements in a logical way. The reason is clear. For example, in an arc-node road centerline network model, arcs correspond to segments that are the conduits for transportation and nodes correspond to intersections connecting arcs together. However, a road centerline is not a natural feature that can be found on the ground or map. The generation and maintenance of centerlines are difficult and tedious because these are only imaginary lines. Human judgment and manual digitization are often essential in generating these lines. To minimize the effort of producing and maintaining a network of additional linear features, an alternative approach is suggested. Path finding method is independent of any arc-node data structure. The network model for such computation is solely based on feature outlines as appeared on the topographic maps. In other words, outlines or symbology of relevant features like road margins, building outlines,and footbridges will directly be used to model the path finding network. This project therefore aims at investigating the rationale and logistics behind in developing such a model with pedestrian walking as an application example. Horizontal cell decomposition and triangulation algorithm are the main aspect of the study In the prototype, users are required to input the topographic map features available, specify the semantics of these features, and identify the origin and destination points. Without the need to generate extra lines of arc-node structure, this study will investigate the automatic generation of an optimal path in terms of computation.