Source code for the paper "Fast generation of collision-free trajectories for robot swarms using GPU acceleration"
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Logos
CaseStudy_100Bicycles.py
CaseStudy_Sort200Quads.py
README.md
maze.txt
maze_collision.cpp
maze_collision.cu
maze_collision.py
robot_collision.cpp
robot_collision.cu
robot_collision.py

README.md

Swarm-Trajectories

This repository hosts the demo source code for the paper "Fast generation of collision-free trajectories for robot swarms using GPU acceleration", which is due to be published as open access in the IEEE Access Journal. An early-access version can be found: https://ieeexplore.ieee.org/document/8587164

A very high-level overview of the demos presented here can be found at: http://mikehamer.info/swarm-trajectories/ Further information to the approach will be released once the paper has been accepted.

Tested using:

  • Python 3
  • PyTorch version 0.4.1
  • CUDA version 9.2.148
  • cuDNN version 7.1.4

To run the demos, simply run (using python) the two CaseStudy programs.