Repository navigation
Releases
Releases · TamerHussen/Final-Year-Project
v1.1 - Post-Feedback Iteration (Viva Demo Build)
This version incorporates critical refinements based on player feedback and technical analysis following the initial user study. It is the intended version for the live Innovation Fest demonstration.
Major Enhancements:
- Adaptive Gameplay Loop: Replaced the 90-second timer with an item collection requirement (Collect 3 fragments with 'E'). This extension provides the ML-Agent (Enemy A) more time to observe, track, and adapt to specific player movement patterns.
- Combat & UX: Added a crosshair and recalibrated the stun-throw origin to the camera center for improved precision.
- AI Balancing: Reduced enemy movement speeds to create fairer "escape" windows and fixed a permanent stun bug for the BT Predator (Enemy B).
- Parity Fix: Resolved a proximity vignette error for Enemy B to ensure both AI types provide the same visual feedback to the player.
- Environment: Fixed floating object assets and increased Escape Zone collider size to ensure 100% activation reliability.
v1.0 - User Study & Research Baseline
This build represents the version of The Hunt used during the official user testing phase. All quantitative data, player preference charts, and AI performance metrics displayed on the project poster are derived from this specific version.
Core Mechanics (v1.0):
- Objective: Survive for 90 seconds. Once the timer hits zero, the Escape Zone activates.
- AI State: Initial PPO training model for Enemy A and standard A* Pathfinding for Enemy B.
- Note for Markers: Please use this build if you wish to verify the gameplay experience reported by the 11 study participants.
The Hunt
The Hunt
The Hunt - ML agent vs BT agent game
A game where pleyrs can choose to get hunted down by either a Machine learning ai or behaviour tree ai