A Hierarchical Path Planning Method Using the Experience of Taxi Drivers

Abstract

Effective path planning has been identified as an important requirement for route navigation in Intelligent Transportation Systems (ITS). However, the paths computed by the conventional path planning algorithms are usually not optimal because they ignore the drivers’ experience and the characteristics of the local road network. Routes chosen by taxi drivers are believed to be more representative, so it can be significant to take advantage of the taxi drivers’ experience to support the path planning. We present a hierarchical path planning method based on the experiential routes of taxis. The algorithm consists of three steps. Firstly, routes are extracted from original taxi trajectories. Secondly, all roads are categorized according to the track data and then the road network is classified into two grades using travel frequency for road segments. Thirdly, combined with Dijkstra algorithm, a hierarchical path planning method which searches paths by traversing the hierarchy is proposed. Guangzhou, China is chosen for the study because of the availability of taxi GPS data. This study compares the paths planned by the proposed approach with the conventional algorithm's results from the aspects such as travel time and route length. The experimental results show that the method proposed incurs much lower costs in travel time than the traditional ITS path planning using Dijkstra or hierarchical Dijkstra algorithms.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 13 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.