anirudh-swaminathan / 3d_euclidean_planning

This is my repository to work on the codes for ECE 276B project 2 - Motion Planning

3d_euclidean_planning

ECE276B Project 2 - Spring 2020 UCSD MS ECE

Anirudh Swaminathan - [email protected]

This is my repository to work on the codes for ECE 276B project 2 - Motion Planning

The following are the list of source files relevant to the proper implementation of the project.

Main Source Files

These are the main, final files for the project.

• src/main.py -> The main.py file which contains the final implementation of my planning algorithm in the given 3D continuous space. This is a copy of main_astar.py
• src/main_astar.py -> The main file that uses the A* class file to run the implementation of the A* algorithm on the given environments
• src/main_original.py -> A copy of the original file provided to us, in case I make any mistakes
• src/Planner.py -> Provided file that houses all the classes of all the different planners. This includes A*, sample greedy planner and sample greedy planer with collision checking.
• src/main_ompl.py -> The main file that uses the OMPLPlanner.py to plan the RRT* algorithm using the OMPL library.
• src/OMPLPlanner.py -> The file that houses the RRTStarPlanner class file to implement OMPL library code for planning.

Auxillary Source Files

These are the auxillary source files that I implemented to build my final solution.

• src/main_collision.py -> This file uses collision checking on the given greedy algorithm file
• src/main_greedy.py -> This file just implements the provided greedy algorithm as a very raw baseline.

Extra Source Files

These are the source files that I implemented to learn and understand the tools that I use for this project

• src/ompl_demo.py -> OMPL RRT* algorithm demonstration from the original library

Input Folders

• src/maps -> These contains the different 3D environments with their AABB obstacles that is used for the project

Output Folders

• src/path_images -> This folder contains the image of all the environments along with the computed path from the start to the end node. Each sub-directory contains the image for that specific algorithm. Sub-directories are a_star/ greedy/ and rrt_star/
• src/path_properties -> This folder contains the properties, such as shortest path, path length and total number of considered nodes for each of the 7 environments. Each sub-directory contains the details for that specific algorithm. Sub-directories are a_star/ greedy/ and rrt_star/

Instructions for running the algorithm

• Install all the dependencies by running

pip install -r requirements.txt

• Install OMPL library with Python Bindings Detailed instrcutions are provided here

• Run the main program using the following command

python main.py

Acknowledgements

The collision checking code in the pyrr library was buggy, and thanks to Piazza posts in the class, I was able to see a C++ implementation provided here, and was able to adapt it to Python for the collision checking.