Skip to content
/ rmader Public

Decentralized Multiagent Trajectory Planner Robust to Communication Delay

License

Notifications You must be signed in to change notification settings

mit-acl/rmader

Repository files navigation

Robust MADER: Decentralized and Asynchronous Multiagent Trajectory Planner Robust to Communication Delay

Accepted to IEEE Robotics and Automation Letters (RA-L)

Accepted to 2023 IEEE International Conference on Robotics and Automation (ICRA)

Awarded Best Poster Award at 2023 IEEE International Conference on Robotics and Automation (ICRA) CAMRS Workshop

4 agents with 2 dynamic obstacles 6 agents with 2 dynamic obstacles
Hardware experiment: 4 agent with 2 dynamic obstacles Hardware experiment: 6 agent with 2 dynamic obstacles
10 agents under 300ms comm. delay 10 agents with 10 dynamic obstacles under 50ms comm. delay
Simulation: 10 agents under 300ms comm. delay Simulation: 10 agents with 10 dynamic obstacles under 50ms comm. delay

Citation

(RA-L preprint) Robust MADER: Decentralized Multiagent Trajectory Planner Robust to Communication Delay in Dynamic Environments (pdf, video):

@article{kondo2023robust,
  title={Robust MADER: Decentralized Multiagent Trajectory Planner Robust to Communication Delay in Dynamic Environments},
  author={Kondo, Kota and Figueroa, Reinaldo and Rached, Juan and Tordesillas, Jesus and Lusk, Parker C and How, Jonathan P},
  journal={arXiv preprint arXiv:2303.06222},
  year={2023}
}

(ICRA Paper) Robust MADER: Decentralized and Asynchronous Multiagent Trajectory Planner Robust to Communication Delay (pdf, video):

@INPROCEEDINGS{10161244,
  author={Kondo, Kota and Tordesillas, Jesus and Figueroa, Reinaldo and Rached, Juan and Merkel, Joseph and Lusk, Parker C. and How, Jonathan P.},
  booktitle={2023 IEEE International Conference on Robotics and Automation (ICRA)}, 
  title={Robust MADER: Decentralized and Asynchronous Multiagent Trajectory Planner Robust to Communication Delay}, 
  year={2023},
  volume={},
  number={},
  pages={1687-1693},
  doi={10.1109/ICRA48891.2023.10161244}}

General Setup

RMADER has been tested with Ubuntu 20.04/ROS Noetic

Not Using Docker

The backend optimizer is Gurobi. Please install the Gurobi Optimizer, and test your installation typing gurobi.sh in the terminal. Have a look at this section if you have any issues.

Then simply run this commands:

cd ~/ && mkdir ws && cd ws && mkdir src && cd src
git clone https://github.com/mit-acl/rmader.git
cd ..
bash src/rmader/install_and_compile.sh      

The script install_and_compile.sh will install CGAL v4.12.4, GLPK and other ROS packages (check the script for details). It will also compile the repo. This bash script assumes that you already have ROS installed in your machine.

Using Docker

Install Docker using this steps, and remove the need of sudo following these steps. Then follow these steps:

cd ~/ && mkdir ws && cd ws && mkdir src && cd src
git clone https://github.com/mit-acl/rmader.git

For Gurobi, you need to download gurobi.lic file from Gurobi Web License Manager (more info here). A gurobi.lic not obtained through WLS will not work on docker. Place your gurobi.lic in docker folder and execute these commands:

cd ./rmader/rmader/docker
docker build -t rmader . #This will probably take several minutes

Once built,

docker run --volume=$PWD/gurobi.lic:/opt/gurobi/gurobi.lic:ro -it rmader
Useful Docker commands
docker container ls -a  #Show a list of the containers
docker rm $(docker ps -aq) #remove all the containers
docker image ls #Show a lis of the images
docker image rm XXX #remove a specific image

### lambda machine simulation
docker build -f rmader/rmader/docker/Dockerfile -t rmader .
docker run --cpus=48 --volume=/home/kkondo/rmader_project/rmader_ws/src/rmader/rmader/docker/gurobi.lic:/opt/gurobi/gurobi.lic:ro --volume=/home/kkondo/data:/home/kota/data -it rmader

Running Multiagent Simulations with 10 agents

Note: For a high number of agents, the performance of RMADER improves with the number of CPUs available in your computer.

roscd rmader && cd script && python run_rmader.py

(if you want to modify the parameters, you can do so in rmader.yaml).

Issues when installing Gurobi:

If you find the error:

“gurobi_continuous.cpp:(.text.startup+0x74): undefined reference to
`GRBModel::set(GRB_StringAttr, std::__cxx11::basic_string<char,
std::char_traits<char>, std::allocator<char> > const&)'”

The solution is:

cd /opt/gurobi800/linux64/src/build  #Note that the name of the folder gurobi800 changes according to the Gurobi version
sudo make
sudo cp libgurobi_c++.a ../../lib/

Credits:

This package uses some C++ classes from the DecompROS repo (included in the thirdparty folder).

Approval for release: This code was approved for release by The Boeing Company in [NEED TO FILL].