GNSS/INS Vehicle Localization With Two-Filter Interference Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Modern GNSS-based and INS-based localization systems face significant positioning errors due to multi-path reception, which can lead to sudden changes in vehicle direction, particularly in autonomous driving scenarios, as conventional filters fail to account for non-Gaussian interference such as multi-path reception and spoofing.
Innovation Solution
A method involving a first filter to determine localization results using GNSS and INS data, followed by a second filter, such as a particle filter, to analyze past and current data to detect and correct interference, including multi-path reception and spoofing, by using additional memory to store and compare localization results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a conventional filter (e.g., Kalman filter) is used for localization, then the system can process GNSS and INS data efficiently, but it fails to detect non-Gaussian interference such as multi-path reception and spoofing
Solution Approach 1:
The filter system is segmented into two distinct filters: a first filter (e.g., Kalman filter) for normal GNSS/INS data processing and a second filter (particle filter) for interference detection. This segmentation allows each filter to specialize in specific tasks, improving interference detection capability while maintaining manageable complexity through modular architecture.
Solution Approach 2:
The second filter acts as an intermediary layer between the first filter and the final localization output. It processes localization results from the first filter and detects interference patterns, mediating the information flow to provide corrected localization data while maintaining system efficiency.
2Measurement precision
If multi-path reception occurs, then GNSS signals can be reflected by objects near the antenna, but this causes positioning errors that lead to sudden changes in vehicle direction
Solution Approach 1:
The system converts the harmful multi-path reception effects into detectable patterns by using the particle filter to analyze deviations in localization results. The interference that would normally cause positioning errors is transformed into a detectable signal that triggers correction mechanisms, ultimately improving positioning accuracy.
Solution Approach 2:
The second filter provides feedback to the first filter by detecting interference patterns in the localization results. When multi-path reception or spoofing is detected, the system adjusts the localization processing accordingly, creating a closed-loop feedback mechanism that maintains positioning accuracy despite harmful factors.
3Reliability
If a second filter is added to detect interference, then localization errors can be corrected, but the computational load and processing time increase
Solution Approach 1:
The second filter does not process all localization data uniformly but applies selective analysis only when interference patterns are detected or when specific conditions are met. This partial action approach reduces unnecessary computational load while maintaining high reliability when interference is present.
Data Source
AI summary
A method for detecting a presence of interference during global navigation satellite system (GNSS)-based and inertial sensor signals (INS)-based localization of a vehicle includes determining localization results using a first filter configured to read in GNSS data and INS data, and storing a plurality of the determined localization results. The plurality of the determined localization results are after one another in terms of time and are each determined using the first filter. The method further includes analyzing the stored plurality of localization results using a second filter which differs from the first filter.