Assignment notes: stage 2
CSE3301: SAR (C.Twardy)
April 9, 2001
Target date: Friday

April
The main goal of stage 2 is to find the best allocation of existing
resources. If your resources are just a set number of searcher-hours
for identical searchers, the allocation should be uniformly
optimal, as outlined in Frost. However, I'm going to complicate
things slightly, and move to a simulation environment for
testing.
That said, the first requirement is to handle the case of
one resource type for a bunch of identical equal-area
segments.
Overall, stage 2 should do three things which strongly extend it towards
stage 3.
- First, it should provide optimal allocation of mixed
resources (specified by speed and sweep width).
- Second, it should
provide a robust interactive user interface for non-batch mode.
- Third, it should provide a batch mode with standard
output, and error (plus return code) for being called from a
simulator environment with command-line arguments.
- General requirements: run in interactive or batch
mode.
- Interactive mode should allow users to enter files or values,
or modify existing values. If ``
'' is not given on the
command line, the program should go to interactive mode, even if
filenames are specified. (It should still read the files.) At
minimum it should allow users to specify-load POA, view initial
and current POA, add-delete-modify-view-load resources, request
resource allocation advice, apply advised allocation, specify
resource allocation (with advised values displayed or defaulted),
and save current state.
- Batch mode should expect command-line input, and output
to STDOUT and STDERR. It should have three modes:
: accept -Aareafile -Rresourcefile. Output
first the assignments and then the predicted POAs in the format
noted below.
: accept -Aareafile -Rresourcefile -Sassignfile. Output
first the
, in the format noted below.
: As far as STDOUT, runs like -advice. It's
here because originally I thought -advice would only generate
the assignment. But then I realized most users would want the
predicted POA as well. If you automatically save the state of the
search before exiting, save the updated POA table for
-auto, but not for -advice.
:
ASSIGN
the assignments in the assignment file
format
/ASSIGN
. Put
after each
line, including the tags.
POA
the new POAs in area file format
/POA
. Put
after each line, including the tags.
- all errors to STDERR
- a return value of 0 for success. Other codes are up to you.
- Area file: On each LINE of the file, separated by
commas:
- Area number: an integer from 1-100 with the special values -1
for RestOfWorld and -2 for Base Camp.
- Location: an (x,y) pair of 4-digit integer strings 0000 - 9999
specifying the center of the area. The integers will be UTM
coordinates: 4 digits by 4 digits.
- Polygon Boundaries: ((Corner-1-UTM-x, UTM y)(corner-2)...(corner-n))
For stage 2, I'll stick with
North-south rectangles, and I'll make sure the areas are well-defined and
precisely adjacent. Rectangles won't all be the same size.
- Travel time to area, in minutes. Integer.
- Altitude in meters. An integer. (No change.)
- Terrain: an integer. (no change)
- Vegetation: an integer between 0 and 100, representing density
of plant cover. (no change)
- Initial POA (float 0..100).
- Resources file
- Resource number (positive integer)
- Resource type (unquoted string)
- POD (float 0..100). In case we use it later.
- ESW in meters (positive float)
- Speed in km per hour (positive float)
- Number of resource-hours available (positive integer)
- (Correction table)(unspecified as yet; can ignore for now)
I will test with files with single resources and files with several
resources of different abilities.
- Assignment file
resourceNum, area, searcher-hours allocated
- Advanced Task Scheduling:
Still not implemented.
- Handing it in:
- cse3301-sar.stage2.[AGM
VLPR
Lee].tar
- compile the executable to
.
- include ``Makefile'' so just typing ``make'' gets a working
allocate.
- Email it to me by April 27. With luck I'll test it over the
weekend, but probably it will take 9 days.
- Testing
As soon as you have the input and output formats working, send me
the code and I will test it on the simulation script. That way we
can both be sure everything is working!
This document was generated using the
LaTeX2HTML translator Version 2K.1beta (1.49)
Copyright © 1993, 1994, 1995, 1996,
Nikos Drakos,
Computer Based Learning Unit, University of Leeds.
Copyright © 1997, 1998, 1999,
Ross Moore,
Mathematics Department, Macquarie University, Sydney.
The command line arguments were:
latex2html -split 0 assign2.tex
The translation was initiated by Charles Twardy on 2001-04-09
Charles Twardy
2001-04-09