Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/bniladridas/simubot
This project combines machine learning and robotics simulation to create a system capable of object detection and interaction within a simulated environment. Using TensorFlow for the AI model and PyBullet for the simulation, the project demonstrates how a robot can be trained to recognize objects and respond accordingly.
https://github.com/bniladridas/simubot
machine-learning python-programming robotics-simulation software-engineering
Last synced: 13 days ago
JSON representation
This project combines machine learning and robotics simulation to create a system capable of object detection and interaction within a simulated environment. Using TensorFlow for the AI model and PyBullet for the simulation, the project demonstrates how a robot can be trained to recognize objects and respond accordingly.
- Host: GitHub
- URL: https://github.com/bniladridas/simubot
- Owner: bniladridas
- License: mit
- Created: 2025-01-07T17:16:04.000Z (24 days ago)
- Default Branch: main
- Last Pushed: 2025-01-07T17:17:59.000Z (24 days ago)
- Last Synced: 2025-01-07T18:27:31.154Z (24 days ago)
- Topics: machine-learning, python-programming, robotics-simulation, software-engineering
- Language: Python
- Homepage:
- Size: 13.8 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# ML Project
This project uses machine learning for object detection and simulates a robot using PyBullet.
## Installation
Follow these steps to set up the project:
```bash
# Clone the repository
git clone https://github.com/bniladridas/SimuBot.git
# Navigate to the project directory
cd SimuBot
# Install dependencies
pip install -r requirements.txt
```
## Usage
Instructions on how to use the project:
```bash
# Run the simulation
python src/main.py
```
## Project Structure
- `src/`: Contains the source code for the simulation and AI components.
- `simulation.py`: Sets up and runs the simulation.
- `ai.py`: Contains the AI model and feature extraction code.
- `utils.py`: Utility functions for image preprocessing.
- `main.py`: Main entry point for running the simulation and AI tasks.
- `assets/`: Contains URDF files for the robot and plane.
- `models/`: Contains the pre-trained AI models.
- `img/`: Contains sample images for testing.
## License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
## ML Workflow
### Control Flow Diagram
This diagram illustrates the control flow of the ML component for object detection:
```mermaid
flowchart TD
A[Start] --> B[Load Image]
B --> C{Preprocess Image?}
C -->|Yes| D[Resize Image]
D --> E[Normalize Image]
E --> F[Add Batch Dimension]
F --> G[Feature Extraction]
G --> H[Model Prediction]
H --> I[Output Features]
I --> J[End]
C -->|No| G
```
### Model and Algorithm Diagram
Here's an overview of the model architecture and algorithms used:
```mermaid
classDiagram
class Model {
+MobileNetV2 model
+load_model(model_path: str)
+predict(input_data: Tensor)
}
class ImageProcessing {
+preprocess_image(image: np.array)
+extract_features(image_path: str)
}
class Simulation {
+setup_simulation()
+run_simulation(physicsClient)
}
Model -- ImageProcessing : Uses
ImageProcessing -- Simulation : Provides features for
```
### Explanation
- **Model**: The project leverages a pre-trained MobileNetV2 for object detection, loaded using TensorFlow's Keras API.
- **Image Processing**: Images are preprocessed (resized, normalized) before being fed into the model for feature extraction.
- **Simulation**: The extracted features could guide actions within the PyBullet simulation environment, simulating real-world scenarios where object detection informs robot behavior.