Case Closed

• 

  This is the play area where I tested my agent against random play, and the agent won

🧰 Technologies Used

Languages

• Python 3.11+ — Core programming language used to implement all game logic and heuristic scoring.
• Markdown + LaTeX — For structured documentation and formula representation.

Frameworks & Libraries

• NumPy (light usage) — For efficient grid handling and array-based operations.
• Collections (deque) — Used for fast BFS (Breadth-First Search) traversal.
• Copy & Deepcopy — To safely simulate move states without affecting the original board.
• Importlib — To dynamically load different heuristic agents (heuristic_1, heuristic_2) during AI competitions.

Platforms & Tools

• Local Python Environment / VS Code — Main development and testing environment.
• Git & GitHub — Version control and repository hosting.
• Unix Terminal (macOS/Linux) / Windows PowerShell — For running CLI-based simulation scripts.
• ANSI/Unicode Terminal Output — Used for color-based board visualization during live agent matches.

Algorithms & Concepts

• Depth-First Search (DFS) — Optimized iterative approach for computing the longest safe path.
• Breadth-First Search (BFS) — For opponent threat estimation and reachable area analysis.
• Heuristic Scoring Model — Weighted scoring logic combining survival, exploration, freedom, and opponent risk factors.
• Game Theory & Spatial Reasoning — To balance defensive and offensive strategies.

APIs / Integrations

• Custom Game State API — Internal structure (via state dictionaries) defining each frame’s board, agent position, and direction.
• Dynamic Import System — Enables switching between multiple heuristic agents without restarting the game.

Cloud / Hosting (optional, for future expansion)

• GitHub Actions — For automated testing and code linting.

Databases

• No persistent database was required — all computations are performed in-memory using Python structures.

Summary

This project runs entirely on Python and algorithmic logic, relying on fundamental data structures (stacks, queues, and graphs) rather than ML frameworks or external APIs.
It demonstrates how pure computational heuristics can deliver intelligent, explainable game behavior without cloud training or neural networks.

Built With

• bfs
• collections-(deque)
• copy
• dfs
• github
• heuristic
• heuristic-scoring-model-**apis/services:**-custom-game-state-logic-(internal)
• importlib
• importlib-**platforms/tools:**-vs-code
• latex-**frameworks/libraries:**-numpy
• markdown
• numpy
• python
• scoring
• terminal
• terminal-(cli-based-simulation)-**algorithms:**-dfs
• vs-code