Loosely-Coupled GNSS and INS Integration Filter
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Low-cost inertial sensors used in GNSS receivers often fail to provide accurate navigation data, especially in pedestrian scenarios, due to inaccuracies in step length estimation and heading biases, which are difficult to estimate and compensate, leading to errors in position and velocity estimates.
Innovation Solution
A loosely-coupled integration filter, based on an Extended Kalman Filter (EKF), combines GNSS navigation information with INS navigation information to generate blended position and velocity estimates, estimating and compensating for speed and heading biases, and adapting to changing GNSS signal conditions using location-dependent and time-varying reliability metrics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If low-cost inertial sensors are used in GNSS receivers, then cost is reduced, but navigation accuracy deteriorates due to inaccuracies in step length estimation and heading biases
Solution Approach 1:
The patent introduces an integration filter as an intermediary component that processes and combines data from both GNSS receiver and low-cost inertial sensors. This filter acts as a mediator that compensates for the inaccuracies of low-cost sensors by fusing them with GNSS data, thereby maintaining navigation accuracy while using cost-effective inertial components.
Solution Approach 2:
The patent creates a composite navigation system that combines GNSS data with low-cost inertial sensor data through a unified filtering framework. By treating the navigation solution as a composite of multiple data sources with different characteristics, the system leverages the strengths of each source while compensating for their individual weaknesses, achieving accurate navigation without requiring expensive inertial sensors.
2Device complexity
If low-cost inertial sensors are used, then device complexity is reduced, but reliability deteriorates in pedestrian scenarios due to difficult-to-estimate biases
Solution Approach 1:
The integration filter implements feedback mechanisms that continuously monitor and compensate for speed and heading biases in real-time. By using feedback from GNSS measurements to correct inertial sensor errors, the system maintains reliable navigation performance in pedestrian scenarios without requiring complex hardware modifications.
Solution Approach 2:
The patent dynamically adjusts estimation parameters and compensation factors based on operating conditions and signal quality. By changing parameters such as filter gains, compensation coefficients, and weighting factors adaptively, the system maintains reliability across varying pedestrian scenarios while keeping the device complexity low.
3Measurement precision
If integration filter combines GNSS and INS navigation information, then navigation accuracy is improved, but device complexity increases
Solution Approach 1:
The integration filter is designed as a universal processing component that handles multiple functions: fusing GNSS and inertial data, compensating for biases, adapting to different scenarios, and providing unified navigation output. By consolidating these functions into a single multi-functional module, the patent achieves high navigation accuracy without proportionally increasing overall device complexity.
Data Source
AI summary
Techniques for loosely coupling a Global Navigation Satellite System (“GNSS”) and an Inertial Navigation System (“INS”) integration are disclosed herein. A system includes a GNSS receiver, an INS, and an integration filter coupled to the GNSS receiver and the INS. The GNSS receiver is configured to provide GNSS navigation information comprising GNSS receiver position and/or velocity estimates. The INS is configured to provide INS navigation information based on an inertial sensor output. The integration filter is configured to provide blended position information comprising a blended position estimate and/or a blended velocity estimate by combining the GNSS navigation information and the INS navigation information, and to estimate and compensate at least one of a speed bias and a heading bias of the INS navigation information.