Inertial Navigation EKF Error Flags for Sensor Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Inertial navigation systems face issues with unreliable or unavailable sensor data leading to persistent errors and integration drift, necessitating system resets that disrupt operations, especially in situations like tunnels where satellite-based navigation is unavailable.
Innovation Solution
Utilizing error flags from inertial sensors to adjust observation noise covariance values and employ replacement data during temporary sensor failures, allowing the Extended Kalman Filter (EKF) to maintain stability and quickly recover without full system resets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the Extended Kalman Filter (EKF) is used for navigation parameter estimation combining GPS and inertial data, then position accuracy is improved through fusion of multiple sensors, but the system becomes vulnerable to persistent errors when inertial sensor data becomes unreliable or unavailable
Solution Approach 1:
The system performs preliminary actions by detecting sensor quality degradation through error flags before complete sensor failure occurs. When error flags are detected, the system proactively increases observation noise covariance and switches to using only GPS data for position estimation, preventing the propagation of unreliable inertial sensor data into the EKF. This preliminary response mechanism maintains system reliability by avoiding integration of degraded sensor data.
Solution Approach 2:
The system changes parameters dynamically by adjusting the observation noise covariance matrix Q based on sensor quality indicators. When error flags are detected in inertial sensor data, the corresponding diagonal elements of Q are increased, which reduces the weight given to the degraded sensor data in the EKF fusion algorithm. This parameter adaptation allows the system to maintain optimal estimation performance under varying sensor reliability conditions.
2Measurement precision
If the system resets entirely when sensor errors occur to eliminate persistent errors, then navigation accuracy is restored, but operation is disrupted and productivity is reduced
Solution Approach 1:
The system segments the error handling response by sensor and error type. Instead of resetting the entire navigation system, the patent applies corrective actions only to the specific sensor or data stream experiencing errors. The EKF continues to process data from healthy sensors while adjusting observation noise covariance for degraded sensors. This segmented approach maintains operational continuity while restoring navigation accuracy by isolating and treating only the affected components.
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring error flags from inertial sensors and dynamically adjusting the EKF observation noise covariance in response. This closed-loop feedback allows the system to automatically adapt to sensor degradation without human intervention or system reset, maintaining both navigation accuracy and operational continuity through real-time parameter adjustment based on sensor health status.
3Reliability
If observation noise covariance is increased in response to sensor errors, then the EKF becomes more robust to erroneous sensor data, but the filter may become less responsive to genuine changes in navigation parameters
Solution Approach 1:
The system applies local quality by adjusting observation noise covariance selectively for specific sensors or sensor axes that exhibit errors, rather than uniformly increasing noise for all sensors. When error flags are detected in particular inertial sensors, only the corresponding diagonal elements of the observation noise covariance matrix Q are increased. This localized adjustment maintains robustness against erroneous data from affected sensors while preserving the EKF's responsiveness to genuine navigation parameter changes from healthy sensors.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
The present invention relates to a method and a system for inertial navigation applying Extended Kalman Filter (EKF). Upon detecting degraded quality of inertial sensor data provided by an inertial sensor device and concerning at least one inertial axis, a noise estimate used by the EKF for the inertial sensor data is temporarily increased from a first noise estimate value to a second noise estimate value. The noise estimate concerns the at least one inertial axis or the inertial sensor associated with the inertial sensor data, and the second noise estimate value is greater than the first noise estimate value. Upon detecting normalization of quality of the at least one type of inertial sensor data, the temporarily increased noise estimate value used by the EKF calculation is reset back to the first noise estimate value after a predetermined delay period after detecting said normalization of quality.