项目作者: RobustFieldAutonomyLab

项目描述 :
Autonomous Exploration Under Uncertainty via Deep Reinforcement Learning on Graphs
高级语言: C++
项目地址: git://github.com/RobustFieldAutonomyLab/DRL_graph_exploration.git
创建时间: 2020-07-21T17:04:56Z
项目社区:https://github.com/RobustFieldAutonomyLab/DRL_graph_exploration

开源协议:BSD 3-Clause "New" or "Revised" License

下载


Autonomous Exploration Under Uncertainty via Deep Reinforcement Learning on Graphs

This repository contains code for robot exploration under uncertainty that uses graph neural networks (GNNs) in conjunction with deep reinforcement learning (DRL), enabling decision-making over graphs containing exploration information to predict a robot’s optimal sensing action in belief space. A demonstration video can be found here.


drawing


drawing

Dependency

  • Python 3
  • PyTorch
  • PyTorch Geometric
  • gtsam (Georgia Tech Smoothing and Mapping library)
    1. git clone -b emex --single-branch https://bitbucket.com/jinkunw/gtsam
    2. cd gtsam
    3. mkdir build && cd build
    4. cmake ..
    5. sudo make install
  • pybind11 (pybind11 — Seamless operability between C++11 and Python)
    1. git clone https://github.com/pybind/pybind11.git
    2. cd pybind11
    3. mkdir build && cd build
    4. cmake ..
    5. sudo make install

Compile

You can use the following commands to download and compile the package.

  1. git clone https://github.com/RobustFieldAutonomyLab/DRL_graph_exploration.git
  2. cd DRL_graph_exploration
  3. mkdir build && cd build
  4. cmake ..
  5. make

Please use the following command to add the build folder to the python path of the system

  1. export PYTHONPATH=/path/to/folder/DRL_graph_exploration/build:$PYTHONPATH

Issues

There is an unsolved memory leak issue in the C++ code. So we use the python subprocess module to run the simulation training. The data in the process will be saved and reloaded every 10000 iterations.

How to Run?

  • To run the saved policy:
    1. cd DRL_graph_exploration/scripts
    2. python3 test.py
  • To show the average reward during the training:
    1. cd DRL_graph_exploration/data
    2. tensorboard --logdir=torch_logs
  • To train your own policy:
    1. cd DRL_graph_exploration/scripts
    2. python3 train.py

Cite

Please cite our paper if you use any of this code:

  1. @inproceedings{chen2020autonomous,
  2. title={Autonomous Exploration Under Uncertainty via Deep Reinforcement Learning on Graphs},
  3. author={Chen, Fanfei and Martin, John D. and Huang, Yewei and Wang, Jinkun and Englot, Brendan},
  4. booktitle={2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  5. pages={6140--6147},
  6. year={2020},
  7. organization={IEEE}
  8. }

Reference