Chapter

Computing with Spatial Trajectories

pp 213-241

Date:

Trajectory Analysis for Driving

  • John KrummAffiliated withMicrosoft Research, Microsoft Corpration Email author 

* Final gross prices may vary according to local VAT.

Get Access

Abstract

This chapter discusses the analysis and use of trajectories from vehicles on roads. It begins with techniques for creating a road map from GPS logs, which is a potentially less expensive way to make up-to-date road maps than traditional methods. Next is a discussion of map matching. This is a collection of techniques to infer which road a vehicle was on given noisy measurements of its location. Map matching is a prerequisite for the next two topics: location prediction and route learning. Location prediction works to anticipate where a vehicle is going, and it can be used to warn drivers of upcoming traffic situations as well as give advertising and alerts about future points of interest. Route learning consists of techniques for automatically creating good route suggestions based on the trajectories of one or more drivers.