Thanks to visit codestin.com
Credit goes to github.com

Skip to content

CWOA/gtrf

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

GTRF (Grid-based Target Recovery Framework)

This repository provides a framework for rapidly prototyping, verifying and evaluating algorithmic approaches to the problem of distributed static and moving target recovery when given limited environment visibility.

This repository accompanies the paper to be published as part of the 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) proceedings entitled "Deep Learning for Exploration and Recovery of Uncharted and Dynamic Targets from UAV-like Vision". Therefore, for implementation details, results, etc. see the published paper. All source code used to generate results and data within the paper is given here for public use, further work, replicability and transparency. The video accompanying the paper -- providing an overview of the problem definition itself, the implemented dual-stream paradigm as well as performance results across a variety of experiments -- can be watched at: https://vimeo.com/280747562

Installation

  1. Simply clone this repository: https://github.com/CWOA/gtrf (if using ROS, place repository in your catkin workspace)
  2. Install the dependencies below as necessary

Essential dependencies

Non-essential dependencies

ROS dependencies

If using ROS; you want to operate on realistic visual input from the simulator, the following dependencies must be installed:

Getting started

  • Run the file "src/NavigationModelMultiple.py"
  • To add your own algorithm, edit the file "YourAlgorithm.py"
  • Experiment configurations, constants, etc. can be found in "src/Constants.py"
  • To run via ROS and Gazebo, run the command "roslaunch gtrf gtrf.launch" which spawns a simulated UAV, the environment itself (containing 5 target cow models) and runs the file "src/NavigationModelMultiple.py".
  • Edit the base directory constant (BASE_DIR) in "src/Constants.py" to point at the complete path at which you want to store training data, model weights, etc. (e.g. "/home/USERNAME/catkin_ws/src/gtrf/data")

License

Non-Commercial Government License (see LICENSE)

About

Grid-based Target Recovery Framework

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published