readme~

Instructions on how to run multi-agent search an drescue simulations.

Modify the following paths in the following files to your preferred destination:

mylogging.py ---> line 6, 7, and 8.
simulation.py ---> line 297, 400, and 452.
environment.py ---> line 87, 89, and 93.
msg_PUnit.py ---> line 98, and 103.
agent_run.py ---> 296, 300, 318, 323, 382, and 384.

Put aamas-res.sh, run-loop.sh, and saso.sh in the catkin_ws folder. 
In order to run a batch of the simulations, run the following script:

./run-loop.sh $nr_of_trials saso.sh

saso.sh is the simulation script, to run the batch only once, execute run-loop as follows:
./run-loop.sh 1 saso.sh

Note that in the scripts the directory structure is assumed as follows:
~/catkin_ws/src/gitagent
Moreover the results shall be moved to the following path:
~/catkin_ws/results/sandr/w\=25\,v\=5/trial$1

Check both run-loop.sh and saso.sh in order to make sure the paths fit what you need.
Once the simulations finish -- the length of one trial can vary dependending on the platform it is being run on -- run the following scripts to get the results. (Make sure that the paths match, for minimal confusion the best would be to keep close to the already defined paths).

Put the bar2.sh and get_means.py script in the sandr folder. Open bar2.sh, and change the input argument related to the number of trials (the second to last argument). If you ran run-loop.sh with ./run-loop.sh 1 saso.sh, then set that number to 1 on all lines. Then run:

./bar2.sh