Conference Paper

Faster Batched Shortest Paths in Road Networks.

Conference: ATMOS 2011 - 11th Workshop on Algorithmic Approaches for Transportation Modeling, Optimization, and Systems, Saarbrücken, Germany, September 8, 2011
Source: DBLP

ABSTRACT We study the problem of computing batched shortest paths in road networks efficiently. Our focus is on computing paths from a single source to multiple targets (one-to-many queries). We perform a comprehensive experimental comparison of several approaches, including new ones. We conclude that a new extension of PHAST (a recent one-to-all algorithm), called RPHAST, has the best performance in most cases, often by orders of magnitude. When used to compute distance tables (many-to-many queries), RPHAST often outperforms all previous approaches.

  • Source
    • "P i , . . . P |p| } ordered by hubs [16]. For each such hub u, those backward labels-to-many is an array of pairs (P i , d(u, P i ). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Quite recently, the algorithmic community has focused on solving multiple shortest-path query problems beyond simple vertex-to-vertex queries, especially in the context of road networks. Unfortunately, this research cannot be generalized for large-scale graphs, e.g., social or collaboration networks, or to efficiently answer Reverse k-Nearest Neighbor (RkNN) queries, which are of practical relevance to a wide range of applications. To remedy this, we propose ReHub, a novel main-memory algorithm that extends the Hub Labeling technique to efficiently answer RkNN queries on large-scale networks. Our experimentation will show that ReHub is the best overall solution for this type of queries, requiring only minimal preprocessing and providing very fast query times.
  • Source
    • "For location services that depend on a query source s, a query target t, and a set of predefined POIs P, single-hub indexing is not good enough. For example, consider the best via point problem [6] [30] [11]: assume you want to go from s to t but need to stop at a post office on the way while minimizing your overall travel time. Formally, you want the post office p that minimizes dist(s, p) + dist(p, t). "
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper introduces HLDB, the first practical system that can answer exact spatial queries on continental road networks entirely within a database. HLDB is based on hub labels (HL), the fastest point-to-point algorithm for road networks, and its queries are implemented (quite naturally) in standard SQL. Within the database, HLDB answers exact distance queries and retrieves full shortest-path descriptions in real time, even on networks with tens of millions of vertices. The basic algorithm can be extended in a natural way (still in SQL) to answer much more sophisticated queries, such as finding the ten closest fast-food restaurants. We also introduce efficient new HL-based algorithms for even harder problems, such as best via point, ride sharing, and point of interest prediction. The HLDB framework makes it easy to implement these algorithms in SQL, enabling interactive applications on continental road networks.
  • Source
    • "PHAST can be extended to a one-to-many scenario, where one must compute shortest paths from a query vertex s to a (fixed) set T of vertices. The resulting algorithm, called RPHAST [9], introduces a target selection phase between preprocessing and query, which extracts from G ↓ the smallest subgraph that is necessary to compute distances to all t ∈ T . "
    [Show abstract] [Hide abstract]
    ABSTRACT: We study the problem of route planning on mobile devices. There are two current approaches to this problem. One option is to have all the routing data on the device, which can then compute routes by itself. This makes it hard to incorporate traffic updates, leading to suboptimal routes. An alternative approach outsources the route computation to a server, which then sends only the route to the device. The downside is that a user is lost when deviating from the proposed route in an area with limited connectivity. In this work, we present an approach that combines the best of both worlds. The server performs the route computation but, instead of sending only the route to the user, it sends a corridor that is robust against deviations. We define these corridors properly and show that their size can be theoretically bounded in road networks. We evaluate their quality experimentally in terms of size and robustness on a continental road network. Finally, we introduce several algorithms to compute corridors efficiently. Our experimental analysis shows that our corridors are small but very robust against deviations, and can be computed quickly on a standard server.
Show more


Available from