Untitled

Abstract

In mobile navigation services, on-road path planning is a basic function that finds a route between a queried start location and a destination. While on roads, a path planning query may be issued due to dynamic factors in various scenarios, such as a sudden change in driving direction, unexpected traffic conditions, or lost ofGPS signals. In these scenarios, path planning needs to be delivered in a timely fashion. The requirement of timeliness is even more challenging when an overwhelming number of path planning queries is submitted to the server, e.g.,during peak hours. As the response time is critical to user satisfaction with personal navigation services, it is a mandate for the server to efficiently handle the heavy workload of path planning requests. To meet this need, we propose a system, namely, Path Planning by Caching (PPC), that aims to answer a new path planning query efficiently by caching and reusing historically queried paths (queried-paths in short). Unlike conventional cache-based path planning systems where a cached query is returned only when it matches completely with a new query, PPC leverages partially matched queried-paths in cache to answer part(s) of the new query. As a result, the server only needs to compute the unmatched path segments, thus significantly reducing the overall system workload.