R help - Jun 2010 - Need help in multi-dimensional clustering

Hi All , 

For an academic project I am trying to do the following 

Step 1 )     Draw and cluster a N ( lets say 3 ) column dataset by dbscan 
algorithm using  R-project’s fpc package   ( let say they are "training
clusters" ) ,
                   Using dbscan as number of clusters are not know before hand  
Step 2 )     Once that is done i want to spread some new 
data points in the above plot space ( lets say these are "test points"
)
Step 3 )     Find out which "test points" are lying within a boundary 
( kinda convex hull ) of any above discovered training clusters

I am done with Step1 but pretty much stuck after that , 
Anyone knows how to proceed on the above 

Below is the code i am using for Step 1 ) 
---------------
library('fpc')
x <- read.table("file.dat" , sep=",")
d <- dbscan(x,eps = 1.9, MinPts=3,showplot = 2);
Plot(d);
---------------
  
I am fairly  new to R , started a week back , but covered quite a few online
resources without success
If anyone knows of a better approach of what i am trying to achieve then  please
let me know

Thanks & Regards
Kapil 


	[[alternative HTML version deleted]]