ViVa-SAFELAND: A Visual Validation Safe Landing Simulation Platform
ViVa-SAFELAND is an open-source simulation platform for testing and evaluating vision-based navigation strategies for unmanned aerial vehicles, with a special focus on autonomous landing in compliance with safety regulations.
This documentation contains the official implementation for the paper "ViVa-SAFELAND: An Open-Source Simulation Platform for Safe Validation of Vision-based Navigation in Aerial Vehicles". It provides a safe, simple, and fair comparison baseline to evaluate and compare different visual navigation solutions under the same conditions.
Key Features
- Real-World Scenarios: Utilize a collection of high-definition aerial videos from unstructured urban environments, including dynamic obstacles like cars and people.
- Emulated Aerial Vehicle (EAV): Navigate within video scenarios using a virtual moving camera that responds to high-level commands.
- Standardized Evaluation: Provides a safe and fair baseline for comparing different visual navigation solutions under identical, repeatable conditions.
- Development & Data Generation: Facilitates the rapid development of autonomous landing strategies and the creation of custom image datasets for training machine learning models.
- Safety-Focused: Enables rigorous testing and debugging of navigation logic in a simulated environment, eliminating risks to hardware and ensuring compliance with safety regulations.
Documentation
For detailed usage instructions, examples, and API documentation, please refer to the ViVa-SAFELAND Documentation.
Citation
If you use ViVa-SAFELAND in your research, please cite us.