An open API service indexing awesome lists of open source software.

https://github.com/alkaifaftab000/autonomous-maze-solver

Building an Autonomous Maze Solver using reinforcement learning to train agents for decision-making in dynamic grid-based environments
https://github.com/alkaifaftab000/autonomous-maze-solver

agent criticism cuda gymnasium-environment maze-solving-bot pytorch reinforcement-learning reward-functions

Last synced: 7 months ago
JSON representation

Building an Autonomous Maze Solver using reinforcement learning to train agents for decision-making in dynamic grid-based environments

Awesome Lists containing this project

README

          

# 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
![Description of the GIF](gify.gif)
---
## 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`