Key Features

• Multi-Threaded A* Pathfinding:

  • Implemented a custom heap to optimize A* performance, allowing faster open list operations .

  • Enabled efficient navigation and decision-making for AI agents.

  • Used a job system to offload A* computations to worker threads, allowing multiple agents to compute paths simultaneously.

• AI Mechanics: 

  • Developed state-based behaviors like patrolling and attacking.

  • Patrol State:

    • Agent randomly selects a goal within a defined patrol range.

    • Agent requests a new path if it reaches the current goal. 

  • Chase State:

    • Tracks the player's last known position.

    • If the player is out of sight, the agent switches back to patrol mode after a delay.

• Job System Workflow:

  • Agent requests a pathfinding job if a new path is needed.

  • The job system queues the request an assigns it to a worker thread.

  • Worker thread executes  A* and stores the result.

  • Agent retrieves the completed path from the job queue and starts moving along it.

• Debugging & Visualization:

  • AI path: Displays the agent's current planned route.

  • Goal Highlighting: Marks the AI's current destination.

• Map Generation:

  • Created mazes using loaded map designs and each time the player loas into the game, it will randomly pick a new maze.