next up previous
Next: Introduction

An Adaptive Interactive Agent
for Route Advice

Seth Rogers
Claude-Nicolas Fiechter
Pat Langley
DaimlerChrysler Research and Technology Center
1510 Page Mill Road, Palo Alto, CA 94304-1135
+1 650 845 2500
{rogers, fiechter, langley}@rtna.daimlerchrysler.com

Abstract:

Current route advice systems present a single route to the driver based on static evaluation criteria, with little or no recourse if the driver finds this solution unsatisfactory. In this paper, we propose a more flexible approach and its implementation in the Adaptive Route Advisor. Our system behaves more like a human travel agent, using driver preferences, when known, and working with him or her to find a satisfactory route. The route advisor predicts what route a driver will prefer based on a model of driver preferences, and, if the predicted route is unsatisfactory, it generates additional routes based on interaction with the driver. The route that the driver eventually selects serves as feedback to improve the preference model. We present a pilot study on using these route selections to construct a personalized model. As the preference model becomes more accurate, the need for interaction decreases and the driver receives better route advice.

Printable Postscript version (6551K, 9 pages)



 
next up previous
Next: Introduction
Seth Rogers
1999-09-10