Usage Guide

Basic Operation

Starting the System

cd Stage2_Advance_navigation_System
python combined_pipeline_v2.py

Controls

  • Camera Feed: Real-time video with object detection overlays
  • Audio Alerts: Voice announcements of detected obstacles
  • Haptic Feedback: Vibration patterns for different alert levels
  • Exit: Press q to quit the application

Alert System

Alert Types

Severity Vibration Audio Example Use Case
Critical 300ms “Danger: person directly ahead!” Immediate threat
Warning 200ms “Warning: obstacle ahead” Close obstacle
Caution 150ms “Caution: uneven surface” Surface hazard
Info 100ms “Person detected to the left” General information

Surface Hazard Alerts

  • Safe: “Normal surface detected”
  • Caution: “Caution: soft surface detected”
  • Unsafe: “Warning: slippery surface detected (aluminium_foil)”

Advanced Usage

Command Line Options

# Run with specific camera
python combined_pipeline_v2.py --camera 1

# Disable audio alerts
python combined_pipeline_v2.py --no-audio

# Enable debug logging
python combined_pipeline_v2.py --debug

Configuration Files

Edit config.json for custom settings:

{
  "camera": {
    "device_id": 0,
    "resolution": [640, 480]
  },
  "detection": {
    "confidence_threshold": 0.5,
    "max_objects": 10
  },
  "alerts": {
    "audio_enabled": true,
    "haptic_enabled": true,
    "cooldown_seconds": 2.0
  }
}

Testing Modes

Component Testing

# Test proximity tracking only
python test_proximity_tracker.py

# Test lighting robustness
python test_lighting_conditions.py

# Test integrated pipeline
python test_combined_pipeline.py

Hardware Testing

# Test audio output
python hardware_interface.py --test-audio

# Test haptic feedback
python hardware_interface.py --test-haptic

# Test combined alerts
python hardware_interface.py --test-all

Real-world Usage Scenarios

Indoor Navigation

  1. Start system in well-lit indoor environment
  2. Walk at normal pace
  3. System provides alerts for furniture, doors, and people
  4. Surface alerts help identify floor changes

Outdoor Navigation

  1. Ensure good lighting conditions
  2. System adapts to changing light levels
  3. Enhanced alerts for vehicles and street obstacles
  4. Surface detection helps identify crosswalks vs roads

Public Transport

  1. Use in stations and on platforms
  2. Alerts for moving people and vehicles
  3. Surface detection identifies platform edges
  4. Proximity tracking prevents collisions

Safety Guidelines

Important Safety Notes

  • Always maintain situational awareness
  • Do not rely solely on the system
  • Test in controlled environments first
  • Keep emergency contacts accessible
  • Report any system failures immediately

Emergency Procedures

  1. System Failure: Stop movement immediately
  2. False Alerts: Continue with caution, verify visually
  3. Hardware Issues: Switch to backup navigation method
  4. Medical Emergency: Call emergency services

Performance Monitoring

Log Files

  • navigation_log.txt: Real-time event logging
  • diagnostics.csv: Performance metrics
  • lighting_test_report.txt: Lighting condition analysis

Monitoring Commands

# View recent alerts
tail -f navigation_log.txt

# Check system performance
python -c "import psutil; print(f'CPU: {psutil.cpu_percent()}%, RAM: {psutil.virtual_memory().percent}%')"

# Test detection latency
python test_combined_pipeline.py --benchmark

Customization

Alert Customization

Modify hardware_interface.py to change:

  • Alert messages and languages
  • Vibration patterns and durations
  • Audio voice settings

Model Customization

  • Train custom object detection models
  • Add new surface texture classes
  • Adjust detection thresholds

Hardware Integration

  • Add new sensor types
  • Integrate with smart canes
  • Connect to wearable devices

Troubleshooting

Common Issues

No Camera Feed

Error: Camera not accessible

Solution: Check camera permissions and connections

Audio Not Working

Error: Audio device not found

Solution: Install audio drivers and test speakers

Slow Performance

Warning: Detection latency > 500ms

Solution: Close other applications, use GPU if available

False Positives

Alert: Person detected (confidence: 0.3)

Solution: Adjust confidence threshold in configuration

Support

Getting Help

  1. Check the troubleshooting guide
  2. Review field testing protocols
  3. Open an issue on GitHub
  4. Join the community discussions

System Information

When reporting issues, include:

  • Operating system and version
  • Python version
  • Hardware specifications
  • Error messages and logs
  • Steps to reproduce the issue