Precision and Reliability of Tightly Coupled PPP GNSS and Landmark Monocular Vision Positioning

Area of Science:

• Geomatics Engineering
• Sensor Fusion
• Navigation Systems

Background:

• Traditional navigation systems often rely on single sensors, which can be prone to failure.
• Integrating multiple data sources, such as landmark recognition and Global Navigation Satellite System (GNSS) measurements, offers potential for improved robustness and accuracy.
• Assessing the quality, precision, and reliability of such integrated systems is crucial for practical applications.

Purpose of the Study:

• To analyze the precision and reliability of an observation-level integrated system combining landmark positions and GNSS measurements.
• To demonstrate the system's performance through design computations and a practical experiment.
• To evaluate the fault-robustness of the integrated system.

Main Methods:

• Development of an observation-level integration model for landmark positions (from a single camera and digital map) and GNSS measurements (from a single-frequency GNSS receiver).
• Performance analysis using design computations and a small-scale practical experiment.
• Evaluation of external reliability by assessing the impact of undetected measurement faults on the horizontal position.

Main Results:

• The integrated system successfully produced position solutions even when individual sensors (camera/map or GNSS receiver) failed.
• With realistic measurement noise assumptions, the low-cost integrated system achieved a horizontal position precision better than half a meter.
• The external reliability was found to be at the few-decimeter level, indicating limited and acceptable impact from undetected faults.

Conclusions:

• The proposed integrated system demonstrates significant fault-robustness, enhancing positional accuracy and reliability.
• The combination of low-cost sensors and sensor fusion provides a viable solution for precise navigation.
• The system's ability to maintain performance despite individual sensor failures highlights its practical utility.