Robot Motion Planning

Overview

This project is a part of coursework Planning and Learning done in Spring 2023.

This code implements A* search based planning and RRT sample based planning for a 3D environment. We are given 7 environments described as:

• All the blocks and the boundaries are given as axis-aligned bounding rectangles.
• Each rectangle is described by a 9-dimensional vector, specifying its lower left corner $x_{min}$ , $y_{min}$ , $z_{min}$ and, its upper right corner $x_{max}$ , $y_{max}$ , $z_{max}$ , and its RGB color (for visualization).
• A start point $x_s$ and goal point $x_d$ is given. The objective is to find a path between the start node and to goal node without colliding with any obstacles.

The objective is to find a path between the start node and to goal node without colliding with any obstacles.

Results

Visualization of A* results:

Flappy Bird	Maze	Monza

Room	Tower	Window

Visualization of RRT results:

Flappy Bird	Maze	Monza

Room	Tower	Window

Code Implementation

1. main.py

This file is the main entry point to the algorithm. Uncomment the testcase to see the visualisation.

2. my_astar.py

Class that implements A* algorithm and is being called in main.

3. my_rrt.py

Class that implements RRT algorithm and is being called in main.

4. RRT_source

This folder contains certain classic RRT algorithm implementations.

5. maps

This folder contains 7 test environments described via a rectangular outer boundary and a list of rectangular obstacles. The start and goal points for each environment are specified in main.py.

6. Results_Astar

This folder contains images of the output from Astar.

7. Results_RRT

This folder contains images of the output from RRT.