Robot Control with ChatGPT/Gemini & PyBullet

---

This project demonstrates how to control a simulated differential drive robot using natural language commands, powered by ChatGPT/Gemini large language model. The robot simulation is handled by the PyBullet physics engine.

Features

• Natural Language Control: Speak or type commands like "move forward," "turn left," or "stop."
• ChatGPT/Gemini Integration: Leverages the ChatGPT/Gemini API to interpret diverse human commands into robot-understandable actions.
• PyBullet Simulation: Provides a lightweight and robust physics-based simulation environment for a custom differential drive robot.
• Modular Design: Easily adaptable to more complex robot models (e.g., quadrupeds, humanoids) by modifying only the robot simulation module.

Setup & Installation

1. Clone this repository:

   git clone [https://github.com/your-username/RobotChatGPTControl.git](https://github.com/your-username/RobotChatGPTControl.git) # Replace with your repo URL
   cd RobotChatGPTControl

2. Create and activate a Python virtual environment:

   python -m venv venv
   .\venv\Scripts\activate # On Windows
   # source venv/bin/activate # On Linux/macOS

3. Install dependencies:

   pip install google-generativeai pybullet python-dotenv numpy

4. Set up your ChatGPT/Gemini API Key:

   • Obtain an API key from the Google Cloud Console.
   • Create a file named .env in the root of this project.
   • Add your API key to the .env file:

     GOOGLE_API_KEY="your_gemini_api_key_here"
     

     or

     OPENAI_API_KEY="your_chatgpt_api_key_here"
     

Running the Simulation

1. Activate your virtual environment (if not already active):

   Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass

   .\venv\Scripts\activate # On Windows

2. Run the main script:

   python main.py

3. A PyBullet simulation window will open. In your terminal, you can now type commands like "move forward a little," "turn right sharply," or "stop."

Project Structure

• main.py: The entry point, orchestrating communication between the user, ChatGPT/Gemini, and the robot simulator.
• chat_interface.py: Handles API calls to ChatGPT/Gemini and translates natural language into structured robot commands.
• robot_simulator.py: Manages the PyBullet simulation, loads the robot model, and executes received commands.
• robots/: Contains the URDF (Universal Robot Description Format) files for the simulated robots (e.g., simple_diff_drive.urdf).

---

Feel free to explore and modify the code to experiment with different robot behaviors or integrate other simulation environments!