A CASP-based Solution for Traffic Signal Optimisation

Requirements

• Clingcon
• Java: to run enhsp and pps (compiled files are in ./bin)

Parse pddl domains into asp instances

python extract_instances.py [Directory_with_pddl_files]

Example

python extract_instances.py ./Test

Run clingcon

clingcon instance_fixed.lp enc_clingcon.lp [ASP_instance] --const horizon=[horizon] --const bound=[cars_bound] --config=crafty 

Example

clingcon instance_fixed.lp enc_clingcon.lp ./Test/p01[count=350].lp --const horizon=600 --const bound=1000000 --config=crafty 

Note: 1000000 stands for 10.00000

Run experiments

Task 1 [Remove optimisation statements from encoding]

task1_run_clingcon_with_bounds.sh 
task1_run_enhsp.sh [Directory]
task1_run_pps.sh [Directory]

Task 2

task2_run_clingcon.sh 
task2_run_enhsp.sh [Directory]
task2_run_pps.sh [Directory]

Extra

Run clingo-dl encoding

clingo-dl instance_fixed.lp enc_clingodl.lp ./Test/p01[count=350].lp --const horizon=900 --config=crafty --heuristic=Domain [--minimize-variable="counter(900,[link])"]

Run asp encoding

clingo instance_fixed.lp enc_asp.lp ./Test/p01[count=350].lp --const horizon=900 --config=crafty