Advanced Navigation Assistance System (ANAS)
Advanced Navigation Assistance System (ANAS)
Welcome to the official documentation for ANAS - an AI-powered navigation assistance system designed to help visually impaired users navigate safely through real-world environments.
🚀 Quick Start
Get started with ANAS in minutes:
📚 Documentation
User Guides
Developer Resources
Research & Testing
🎯 Key Features
- Real-time Obstacle Detection: YOLOv8-powered object recognition
- Surface Hazard Identification: 67-class texture classification
- Multi-modal Feedback: Audio speech + haptic vibration
- Cross-platform Support: Windows, Linux, macOS, Raspberry Pi
- Lighting Robustness: Tested across 8 illumination conditions
📊 System Architecture
┌─────────────────────────────────────────────────────┐
│ User Feedback Layer │
│ (Audio via TTS + Haptic Vibration) │
│ [hardware_interface.py] │
├─────────────────────────────────────────────────────┤
│ Decision & Fusion Layer │
│ ├─ Proximity Tracking [proximity_tracker.py] │
│ ├─ Surface Hazard Mapping │
│ ├─ Severity Assessment │
│ └─ Alert Filtering │
├─────────────────────────────────────────────────────┤
│ Detection Layer │
│ ├─ Obstacle Detection [YOLO v8n] │
│ ├─ Surface Classification [ResNet50 - 67 classes] │
│ ├─ Direction Classification │
│ └─ Distance Estimation │
├─────────────────────────────────────────────────────┤
│ Input Layer │
│ ├─ Camera Capture [OpenCV] │
│ ├─ Frame Preprocessing │
│ ├─ Lighting Adaptation │
│ └─ Performance Optimization │
└─────────────────────────────────────────────────────┘
📈 Performance Metrics
| Metric | Target | Current Status |
|---|---|---|
| Obstacle Detection Rate | ≥95% | ✅ Achieved |
| False Positive Rate | ≤3% | ✅ Achieved |
| Detection Latency | <200ms | ✅ Achieved |
| Surface Classification | 67 classes | 🔄 Training (57.65% acc) |
🔬 Research Impact
ANAS represents a significant advancement in assistive navigation technology:
- Accessibility: Enhances independence for visually impaired users
- AI Integration: Combines computer vision with haptic feedback
- Real-world Testing: Comprehensive field testing protocols
- Open Source: Freely available for research and development
📞 Contact & Support
- Repository: GitHub
- Issues: Report Bugs
- Discussions: Community Forum
This system is designed to assist visually impaired users but should not replace human judgment or professional mobility training.