Pathfinder_AI

Pathfinder_AI is a part of Assignment 2A for the COS30019 - Introduction to Artificial Intelligence. This project implements search algorithms such as DFS, BFS, GBFS, ASTAR and 2 custom methods.

---

Project Structure

Pathfinder_AI/
│__ Assignment2A.pdf     #project report 
├── Assets/              # Visuals, structure diagrams
├── methods/             # DFS, BFS, GBFS, ASTAR, CUS1, CUS2
├── pseudocode/          # Pseudocode references
├── tests/               # Input graph test files
├── utils/               # Graph + Heuristic utility functions
│
├── GUI.py               # Graphical search visualizer
├── search.py            # Command-line search executor
├── requirements.txt     # Required dependencies
├── create_test.py       # Path testing helper
├── TreeBasedSearch.ipynb# Jupyter testing notebook
├── README.md            # Project information

---

How It Works

Search methods are implemented as a class with a search(self, graph, origin, destination) method.
The input graph is taken from the input text file stored in tests/test*.txt by using utils/graph.py.

We use Euclidean distance in utils/heuristics.py for GBFS and A* (suited to 2D graphs).
Manhattan distance can be added in the future for other environments.

---

How to Run (Command Line)

Use the command line to run the script with your input graph file and method:

python3 search.py <path_to_input_text_file> <search_method>

Example:

python3 search.py tests/test2.txt DFS

---

How to Run (Graphical GUI)

1. Install requirements:

pip install -r requirements.txt

Or manually:

pip install pygame

2. Run the GUI version:

python GUI.py tests/test2.txt DFS

3. Press Enter when prompted to start the visual traversal.

---

Output Format

The execution output required by the assignment looks like this:

<input_text_filename> <search_method>  
<destination_goal> <number_of_nodes_in_path>  
<full_path>

Example:

tests/test2.txt GBFS  
11 8  
1 3 5 6 7 9 10 11

---

Test Files

• test1.txt – Basic pathfinding test
• test3.txt – Larger graph with multiple branches
• test4.txt – Multi-destination test with forked paths
• test5.txt – Edge case with multiple potential goal paths
• DFS_test.txt – Specifically designed for DFS validation
• BFS_test.txt – BFS-focused graph structure
• GBFS_test.txt – Heuristic-based graph to test GBFS
• ASTARTest.txt – Test case targeting A* algorithm
• CUS1_test.txt – Designed for Uniform Cost Search (CUS1)
• CUS2test.txt – Tailored for the custom CUS2 algorithm
• PathFinder-test.txt – Overall integrated scenario testing

---

Notes

• Follows assignment rule: stops at first reachable goal
• Expands nodes in ascending order using sorted()
• All methods return both the path and cost
• Modular codebase built for easy extension

---

Group Members

• Ananda Pathiranage Ruveen Thathsilu Jayasinghe - 104317649
• Denver Cope - 104738758
• Rahat Alam - 103810105
• Fayiz Kallupalathingal - 104658733

---

Course Info

• Course: COS30019 - Introduction to Artificial Intelligence
• Semester: 2025 HS1

---

Repository

GitHub Repo – Pathfinder_AI