Kalman Filter Navigation with Sensor Criticality Analysis

The navigation system is segmented into multiple independent positioning modes (GPS positioning, dead reckoning positioning, and map-matching positioning) that can operate independently or in combination. Each segmentation handles specific scenarios: GPS for open areas, dead reckoning for signal-denied environments, and map-matching for corridor-based navigation, thereby maintaining precision across diverse environments while preserving broad coverage.

The patent employs a composite positioning approach by integrating multiple positioning technologies (GPS, inertial sensors, magnetic sensors, and map-matching algorithms) into a unified navigation system. This composite structure combines the strengths of each component: GPS provides global coverage, inertial sensors maintain precision during signal loss, and map-matching enhances accuracy in structured environments like subway stations, achieving both broad coverage and consistent precision.

The system dynamically adjusts the active positioning mode based on real-time GPS signal quality and environmental conditions. The controller continuously evaluates signal strength and automatically switches between GPS positioning, dead reckoning positioning, and map-matching positioning, ensuring high reliability without requiring all components to operate simultaneously, thereby managing complexity through adaptive rather than static configuration.

The navigation system incorporates self-diagnosis and automatic mode selection capabilities. The controller automatically detects GPS signal availability and transitions between positioning modes without user intervention. Additionally, the system performs self-calibration of sensors and automatic error correction, reducing the need for complex external control mechanisms while maintaining high reliability.

The system performs preliminary calibration of inertial sensors and magnetic sensors during periods when GPS signal is available. Acceleration data, angular velocity data, and magnetic field data are pre-processed and stored for later use during signal-denied periods. This preliminary action ensures that when GPS is blocked, the system can immediately transition to dead reckoning positioning with maintained precision, as the sensor characteristics have already been characterized and stored.

The patent introduces map-matching positioning as an intermediary mechanism between GPS and dead reckoning positioning. When GPS signal is blocked, the system uses map-matching to constrain the dead reckoning position estimates to valid locations on the pre-stored map data. This intermediary layer prevents drift accumulation and maintains positioning precision in signal-denied environments while preserving the privacy and security benefits of GPS blockage.