Browse Prior Art Database

Automatic Destination Prediction after Route Disruption

IP.com Disclosure Number: IPCOM000214456D
Publication Date: 2012-Jan-30
Document File: 2 page(s) / 50K

Publishing Venue

The IP.com Prior Art Database

Abstract

Pattern keys are an existing technology that allow prediction of a driver's destination based on fulfilment of a list of criteria from which the pattern key is generated. Problems arise when one of these criteria is not fulfilled (e.g. the driver deviates from a normal route due to traffic) but reaching the final destination of the pattern key is still the aim. Here, we propose a method whereby we can still predict the likely destination if the driver deviates from the expected route.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 59% of the total text.

Page 01 of 2

Automatic Destination Prediction after Route Disruption

Pattern keys are an existing technology that allow prediction of a driver's destination based on fulfilment of a list of criteria from which the pattern key is generated. Problems arise when one of these criteria is not fulfilled (e.g. the driver deviates from a normal route due to traffic) but reaching the final destination of the pattern key is still the aim. Here, we propose a method whereby we can still predict the

likely destination if the driver deviates from the expected route.

Known Related Art:
Real Time Destination Prediction Based On Efficient Routes - John Krumm

    This is based on the most efficient routes and takes no account for being diverted from the most likely or easiest route to a particular destination.

Travel destination prediction using frequent crossing pattern from driving history - Kostov, V

    This is based on past driving history and takes no account of taking an alternative route to a destination.

    Currently with pattern keys, each journey is broken down into a set of key decision nodes which form the criteria for that pattern key. When the pattern key is generated, a series of sub-nodes which bridge the gaps between each key decision node are discarded. We propose keeping a subset of the more significant sub-nodes in order to help determine when a pattern key has been deviated from (missing out one or more key decision nodes) and subsequently rejoined.

    The algorithm we propose waits fo...