Artificial Intelligence Assignment – CSMI17

This repository contains the solutions and reports for the CSMI17 Artificial Intelligence course assignment.
It covers two classic AI problems implemented and analyzed using different techniques.

---

Problem 1: Robot Path-Finding using A* Search

Objective

A robot must find a path from a start cell to a goal cell in a grid that contains random obstacles.

Approach

• Implemented A* search algorithm.
• Used three different heuristics:
  • Manhattan Distance
  • Euclidean Distance
  • Diagonal Distance (Chebyshev)
• Randomly generated grids with random start and goal cells for testing.

Outcome

Performance of heuristics was compared using graphs to identify which heuristic achieved faster convergence and optimal pathfinding efficiency.

Problem 2: Timetable Generation as a Constraint Satisfaction Problem (CSP)

Objective

To automatically generate a valid timetable by satisfying constraints such as no time-slot or room conflicts.

Approach

Two CSP solving techniques were implemented:

1. Backtracking with MRV & LCV Heuristics
2. Backtracking with Forward Checking

Outcome

Forward checking demonstrated better pruning and reduced backtracking steps, while heuristic-based backtracking performed competitively on smaller datasets.

---

Setup

• Language: Java (for A* and CSP implementations)