AIFA Assignment

Group project for Assignment #1 of the AIFA course [AI61005].

Group Members

• Chirahg Gupta [20ME30016]
• Saksham Kumar Singh [20EE10065] @photon999
• Saumyadip Nandy [20EC10072] @Pagol1
• Vibhanshu Vaibhav [20EE30030] @Vibhanshu2803

Project Description

Our project implements the A* and Dijkstra algorithms with different hueristics.
The project uses concepts related to the AI problem of search to minimize the path cost.

Language and Libraries Used

• Python 3
• opencv-python
• numpy
• heapq

Usage

Clone the repository and run the command python3 Run.py to execute.

Configuration

The parameters in the config.json file can be changed to used to switch between algorithms, heuristics etc. The parameters are:

• height: Height of the map
• width: Width of the map
• start_pose: Start coordinates
• goal_pose: Goal coordinates
• generate_random_map:
  @Values:
  • 0: Skip map generation and use pre-existing map
  • 1: Generate a new random map
    @Note: Must be set to 1 in the initial run
• num_random_obstacles: Maximum number of obstacles
• rand_obstacle_size: Maximum ostacle dimension (length or breadth)
• clearance: Minimum obstacle distance from start and end nodes
• type: Specifies the algorithm and hueristic to be used. In the format [Algorithm]-[Hueristic]
  @Values:
  • Algorithms:
    • Astar @Hueristics: Manhattan, Chessboard, Euclidean and Diagonal
    • Dijkstra @Hueristics: Manhattan, Euclidean (Used to calculate path cost)
      @Example: Astar-Euclidean
• valid_moves:
  @Values:
  • 4: Square movement
  • 8: Octagonal movement
    @Note: For Dijkstra use the Euclidean hueristic for octagonal movement
• show_process: Set to 1 to see the current path, frontier and explored nodes; set to 0 otherwise.
  @Note: Enabling this option heavily increases the execution time.