Autonomous Maze Solver 🚀

Project Overview 🧩

Leveraging the power of reinforcement learning, this project develops an Autonomous Maze Solver purely based on reinforcement learning principles.

---

Workflows ⚙️

1. Creating agent.py:

   • Controls the actor and critic functionalities.
   • Trains the system from previous mistakes using reward functions.

2. Creating model.py:

   • The actor performs actions while the critic evaluates by giving punishments or rewards.

3. Creating gymnasium.py:

   • Develops grid structures for the maze environment.

Note: No traditional machine learning models were created.

---

Learning Outcomes 🎓

1. Developed a deeper understanding of reinforcement learning frameworks.
2. Acquired insights into implementing actor-critic models and reward-driven training systems.

---

UI

Plugins and Libraries Used 🛠️

• --extra-index-url https://download.pytorch.org/whl/cu118
• torch
• torchvision
• torchaudio
• gymnasium>=0.29.0
• matplotlib>=3.10.0
• gymnasium-robotics>=1.3.0
• pybullet>=3.2.0
• tensorboard>=2.15.0