https://github.com/simplisoni/3d-reinforcementlearning-simulator
An interactive, web-based simulator for visualizing reinforcement learning policies in 3D environments
https://github.com/simplisoni/3d-reinforcementlearning-simulator
3d reinforcement-learning simulation three-js
Last synced: about 2 months ago
JSON representation
An interactive, web-based simulator for visualizing reinforcement learning policies in 3D environments
- Host: GitHub
- URL: https://github.com/simplisoni/3d-reinforcementlearning-simulator
- Owner: SimpliSoni
- License: gpl-3.0
- Created: 2025-09-25T16:09:02.000Z (9 months ago)
- Default Branch: main
- Last Pushed: 2025-09-25T16:54:09.000Z (9 months ago)
- Last Synced: 2025-09-25T18:12:25.475Z (9 months ago)
- Topics: 3d, reinforcement-learning, simulation, three-js
- Language: TypeScript
- Homepage: https://3-d-reinforcement-learning-simulato.vercel.app
- Size: 99.6 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# 3D-ReinforcementLearning-Simulator
An interactive, web-based simulator for visualizing reinforcement learning policies in 3D environments
-----
* **Render 3D Environments**: Load and display different simulation environments, like a Grid World.
* **Upload and Test Policies**: Users can upload pre-trained policy files (e.g., a Q-table in JSON format) to guide the agent.
* **Control the Simulation**: Play, pause, step through, and reset the agent's interaction with the environment.
* **Visualize Key Metrics**: Toggle visual aids like Q-values to better understand the agent's decision-making process at each step.
This tool is invaluable for students, researchers, and developers who need to bridge the gap between RL theory and practical application.
-----
## Built With
* React
* Three.js
-----
## Getting Started
To get a local copy up and running, follow these simple steps.
### Prerequisites
You'll need Node.js and npm installed on your machine.
```bash
npm install npm@latest -g
```
### Installation
1. Clone the repository:
```bash
git clone https://github.com/SimpliSoni/3D-ReinforcementLearning-Simulator.git
```
2. Navigate to the project directory:
```bash
cd 3D-ReinforcementLearning-Simulator
```
3. Install NPM packages:
```bash
npm install
```
4. Start the development server:
```bash
npm start
```
Your browser should automatically open to `http://localhost:3000`.
-----
## Usage
Once the application is running, you can interact with the simulation using the control panel:
1. **Select an Environment**: Use the dropdown menu to choose a simulation world (e.g., "Grid World").
2. **Upload a Policy**: Click "**Choose File**" to upload a compatible policy file (e.g., a JSON file containing a Q-table).
3. **Run the Simulation**: Use the "**Play**," "**Pause**," and "**Step**" buttons to observe the agent's behavior.
4. **Analyze**: Toggle the "**Show Q-Values**" checkbox to visualize the agent's learned values for different actions in its current state.
This allows for a direct and intuitive analysis of how the trained policy performs in a 3D space.