This repository contains the code for the tool JEDAI. System Demonstration of JEDAI is accepted at AAMAS 2022, and will appear as:
JEDAI: A System for Skill-Aligned Explainable Robot Planning.
Naman Shah*,
Pulkit Verma*,
Trevor Angle, and
Siddharth Srivastava.
21st International Conference on Autonomous Agents and MultiAgent Systems, 2022.
The recommended way of using JEDAI is to use pre-configured Virtual Machine image that is available here: https://bit.ly/2WccU4K
To setup the system manually, you can use the steps given below:
Use this command to start JEDAI from the JEDAI source directory (~/JEDAI/ in VM Image).
./start_jedai.sh
Alternatively execute this command:
python3 manage.py runserver
The output of this command includes a link to the development server hosting the frontend.
You can stop the execution anytime using this command from the JEDAI source directory (~/JEDAI/ in VM Image):
./stop_jedai.sh
- Ubuntu 18.04
- Python 2 and 3
- Validate: https://github.com/KCL-Planning/VAL
-
Retrieve and enter the repo:
git clone https://github.com/KCL-Planning/VAL
cd VAL
-
Build the binary:
./scripts/linux/build_linux64.sh all Release
- This will put
Validate
in<PARENT_DIR>/VAL/build/linux64/Release/bin
- This will put
-
NOTE: JEDAI is tested extensively with Chromium (including Edge, Vivaldi, and Google Chrome). Support on other browsers is not guaranteed.
-
Retrieve the
TMP_Merged
submodule by running the following in the project root (unless you already haveTMP_Merged
somewhere else on your system and want to use that, in which case you can try a symlink):git clone https://github.com/AAIR-lab/ATAM.git TMP_Merged
-
You must then install the dependencies for the submodule (this will probably take a while):
bash TMP_Merged/install_tmp_dependencies.sh
-
Also make sure to check out the correct branch of the submodule:
cd TMP_Merged
git checkout origin/TMP_JEDAI
-
-
Install the web framework:
pip3 install django
-
Install the YAML library:
pip3 install PyYAML
-
Install the PDDL library:
pip3 install pddlpy
-
If you get an error while running the code about a missing module named
__builtin__
in theantlr4
library, then running this should help:pip3 install antlr4-python3-runtime==4.7
-
-
Install the imaging library:
pip3 install Pillow
-
Check that
PYTHON_2_PATH
andVAL_PATH
inconfig.py
are pointing to the corresponding binaries on your system.
You are required to submit a domain and problem file, as well as a .dae environment file. See the test_domains
directory for examples.
After installing its dependencies, the TMP submodule should work out of the box, with environments popping up and giving a demonstration of successful plans. If you get any strange import errors from TMP despite packages seeming to be installed correctly, double-check your all your environment variables (especially if using an IDE like PyCharm).
Naman Shah
Pulkit Verma
Trevor Angle
Kiran Prasad
Amruta Tapadiya
Kyle Atkinson
Chirav Dave
Judith Rosenke
Rushang Karia
Siddharth Srivastava
@inproceedings{shah_2022_jedai,
author = {Shah, Naman and Verma, Pulkit and Angle, Trevor and Srivastava, Siddharth},
title = {{JEDAI: A System for Skill-Aligned Explainable Robot Planning}},
booktitle = {Proceedings of the 21st International Conference on Autonomous Agents and MultiAgent Systems},
year={2022}
}