Path Planning

Thesis Path Planning Simulations

Thanks to Atsushi Sakai (@Atsushi_twi) for publishing PythonRobotics

Proposed solution

• run*.sh : Different scripts to run all worlds in batch
• world*.csv : Different worlds which you can run
• main.py : The main program to run (it uses the proposed hybrid algorithm).
• main_astar.py : An alternative resolution using Astar (used for comparisson purposes).
• navApfNavigation.py : Potential Field based path planner
• navFollowPath.py : FollowPath navigation
• navBrushfire.py : Brushfire map calculation
• navDeliverative.py : Deliverative planner & decision making
• navLidar*.py : LIDAR related processing
• navData.py : Class that handles data storing

Tools

• WorldBuilder.ods : Spreadsheet where you create the world and export as CSV
• show_world.py : Tool to create a usable world from CSV output
• show_limits.py : Show a JSON encoded limit
• show_map.py : Show a JSON encoded map
• show_path.py: Show a JSON encoded path

Usage example

Download & install (in Linux):

git clone https://github.com/ibonelli/pathplanning.git
apt install python3 python3-pip
pip install numpy matplotlib

Single map run:

python3 main.py world15.csv
less -S navigation.log
python3 main_astar.py world15.csv
pushd tools ; python3 show_path.py ../world15_path.json ; popd
pushd tools ; python3 show_map.py ../world15_map.json ; popd

All maps:

run.sh

All maps with different escenarios for each world:

run_worlds.sh