Inertial Sensor Calibration Using Observer Device Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Inertial sensors in vehicles face calibration errors due to geographical limitations such as high-rise buildings and underground passages where GPS signals are lost or weakened, leading to degraded positioning continuity and reliability in navigation systems and adaptive headlight functionality.
Innovation Solution
An inertial sensor calibration method that involves mounting an observer device and inertial sensor on a vehicle, acquiring motion data, creating a dynamic variation model with parameters for offset and scale factors, and calibrating it using energy optimization and discretization techniques to learn various road conditions, allowing for continuous and reliable positioning even without GPS or satellite signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS navigation system and electronic compass are used for calibration, then positioning accuracy is improved, but reliability deteriorates in signal-attenuated zones such as underground passages and high-rise building areas
Solution Approach 1:
The patent introduces an intermediary calibration method that uses observer device measurements (actual vehicle motion data) as a mediator between the inertial sensor and the external environment. Instead of directly relying on GPS or electronic compass signals that may be unavailable, the system uses the observer device to provide reference data for calibration in signal-attenuated zones, enabling continuous and reliable positioning without requiring external satellite or magnetic field signals.
2Measurement precision
If inertial sensor calibration is performed using external reference signals (GPS, electronic compass, pressure sensor), then calibration accuracy is improved, but adaptability deteriorates in geographical zones where reference signals are lost or attenuated
Solution Approach 1:
The patent implements self-service calibration where the inertial sensor system performs its own calibration using observer device measurements. The system independently acquires actual vehicle motion data from the observer device, processes this data through the same computational steps as external reference signals, and uses the results to calibrate the inertial sensor without requiring external calibration infrastructure. This enables the system to adapt to any geographical zone including underground passages and high-rise building areas where external reference signals are unavailable.
3Measurement precision
If navigation system relies on satellite signals for positioning, then positioning accuracy is improved, but reliability deteriorates when passing through underground passages or areas with signal loss
Solution Approach 1:
The patent applies preliminary action by performing calibration using observer device measurements before the vehicle enters signal-attenuated zones. The system continuously acquires and processes actual vehicle motion data from the observer device, maintaining calibrated inertial sensor parameters in advance. This preliminary calibration ensures that when the vehicle transitions to underground passages or signal-loss areas, the inertial sensor already has accurate calibration data, enabling continuous positioning without interruption.
Data Source
AI summary
An inertial sensor calibration method has steps of mounting an observer device and an inertial sensor of a vehicle carrying on an inertial move, acquiring actual vehicle motion data from the observer device and inertial signal data of the inertial sensor, calculating an integral corresponding to the vehicular dynamic variation model with respect to the inertial signal data to obtain predicted vehicle sensor data and calculating variations of the actual vehicle motion data, acquiring differences between the two calculated data, applying an energy optimization and a discretization to the differences so as to obtain parametric error variances, and feeding back the parametric error variances to the vehicular dynamic variation model to calibrate the parameters associated with offset and scale factor and acquire a calibrated vehicular dynamic variation model. Under the premise of no GPS, electronic compass or pressure sensor, the present invention can secure positioning continuity and reliability.