Efficient fastest-path computations for road maps

Abstract In the age of real-time online traffic information and GPS-enabled devices, fastest-path computations between two points in a road network modeled as directed graph, where each edge is weighted by “travel time” value, are becoming standard feature many navigation-related applications. To support this, very efficient computation these paths large networks critical. Fastest may be computed minimal-cost but traditional path algorithms based on variants classical Dijkstra algorithm do not scale well, worst case they traverse entire graph. A common improvement, which can dramatically reduce number graph vertices traversed, A* algorithm, requires good heuristic lower bound minimal cost. We introduce simple, effective, function small values assigned to vertex. The separators efficiently preprocessing stage. present experimental results demonstrating that our provides estimates cost superior those other heuristics. Our experiments show when used this traversed an order magnitude compared