This code allows p4 to be used to generate and display deceptive paths. It has been developed for and used in the following paper:
- Peta Masters and Sebastian Sardina. Deceptive path-planning. In Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), pages 4368--4375, Melbourne, Australia, 2017.
Deceptive path-planners (e.g. agent_ds1.py
, agent_ds2.py
, etc) are able to generate deceptive paths.
File deceptor.py
contains helper routines.
Pass extra possible goals in config file or as array of coordinate tuples in POSS_GOALS
command line parameter.
Call agent in from command line or menu and run in the usual way.
- For automated testing, generate problems sets exactly as for goal recognition (see
/src/GR
folder). Locatedpp.py
insrc/
folder as calling script: inputs problems from nominated GR file, runs strategies (i.e. deceptive path-planning agents), and outputs results to csv. Currently, you need to hard-code the agent names in the dpp script.