Intelligent Navigation System for Ambulance Vehicles: Incorporating Expert Knowledge

Abstract

Real traffic time data acquisition , currently used in vehicle navigation systems , can be very expensive , and also inaccurate and biased . Therefore , researchers have been motivated to develop alternative systems that are affordable by many countries , with acceptable accuracy especially in emergency medical services

This research , which has been developed in GIS , presents an intelligent navigation system for ambulance drivers , with an aim of finding the least travel time route that is independent of using real time traffic data . This system helps the drivers to overcome decision making errors due

to time pressure stress , about which route to follow . Here , the specific focus is on building a set of rules from the knowledge and experiences of various ambulance drivers , and is associated with factors that might affect the response time to reach the incident locations This is done by weight roads according to such factors in to calculate relatively accurate travel times along these roads . In addition , this system also considers the time of the day and locations

The system in this research was implemented as a set of scenarios using ArcView 's network analyst extension , in to calculate the fastest route from any hospital (as the dispatch centre ) to any incident in the city of Leicester , UK

Research

Introduction

In Vehicles Navigation Systems (IVNS ) - mounted in ambulance vehicles are used to guide the drivers by following the quickest path from the dispatch location to the incident location . This navigation is presently supported by real time data of the current traffic conditions of the roads . The real time traffic data inputs are collected using traffic sensors (detectors , which are either mounted on specially equipped moving vehicles or situated on the road sides . These sensors are either buried under the roads surfaces or are camera-based (Fisher , 2004

However , real-time data collecting equipments are very expensive which also includes the relatively high operating costs for each survey (Nual et al . 2002 , Balke et al . 2005 . Moreover , real time data collected from vehicles ' sensors depend on fixed number of vehicles , thus the data collected only cover limited number of roads (Thompson , 2003 . In addition , the results of the real time traffic collection can also be biased about the traffic condition . For instance , buried and road side sensors can only provide the information about the spots that they are situated in , thus the data collected does not represent the real traffic state on the entire stretch of the road (Haas et al , 2001 . Moreover real time data collection is labour intensive and requires trained technicians (Thompson , 2003 . Finally , it is also difficult to record the change of traffic conditions throughout the day (Borri and Cera 2005

Many Intelligent transportation systems (ITS ) specialists and technology scientists have been working towards the problem , by either developing ways to overcome the problems mentioned above or by developing alternative methods to collect real time traffic data . For example Thompson (2003 ) proposes using a relatively inexpensive technique that provides more accurate results by Integrating PDA...