A Data Mining Approach For Adaptive Path Planning On Large Road Networks

Estimating fastest paths on large networks is a crucial problem for dynamic route guidance systems. This chapter presents a data mining based approach for approximating fastest paths on urban road networks. The traffic data consists of input flows at the entry nodes, system state of the network or the number of cars present in various roads of the roads, and the paths joining the various origins and the destinations of the network. To find out the relationship between the input flows, arc states and the fastest paths of the network, we developed a data mining approach called hybrid clustering. The objective of hybrid clustering is to develop IF-THEN based decision rules for determining fastest paths. Whenever a driver wants to know the fastest path between a given origin-destination pair, he/she sends a query into the path database indicating his/her current position and the destination. The database then matches the query data against the database parameters. If matching is found, then the database provides the fastest path to the driver