Navigation System Integrity Modeling via Error Mixture Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional navigation systems struggle to model the integrity of filtered global navigation satellite systems, as existing methods do not account for feedback from previous position estimates, making it difficult to determine the reliability of navigation solutions under arbitrary conditions.
Innovation Solution
A method is introduced that defines a set of fault conditions, calculates component navigation system error distributions, and forms a mixture distribution to determine the overall navigation error distribution, allowing for improved integrity evaluation and modeling of filtered navigation systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional fault tree approach is used for integrity modeling, then system-level integrity allocation can be divided between subsystems, but the approach is arbitrary and lacks quantitative determination of actual system integrity
Solution Approach 1:
The patent applies feedback by using the filter covariance from the Kalman filter as input to the integrity modeling process. The integrity monitor uses the filtered navigation solution and its associated covariance to detect faults, creating a feedback loop where the filtering process informs the integrity assessment. This resolves the contradiction by providing a quantitative, non-arbitrary method that leverages existing filter outputs.
Solution Approach 2:
The patent introduces an intermediary integrity monitor that bridges the filtering process and the navigation solution. This intermediary component processes the filtered solution and covariance to generate integrity information, avoiding the need for arbitrary fault tree allocations while providing quantitative integrity determination.
2Measurement precision
If filter covariance is used to indicate navigation accuracy, then nominal position accuracy can be quantified, but the filter covariance underestimates true positioning errors when faults are present
Solution Approach 1:
The patent segments the error analysis into two distinct components: the filter covariance representing nominal accuracy under normal conditions, and the integrity monitor analyzing fault conditions separately. This segmentation allows each component to serve its specific purpose without the limitations of the other, resolving the contradiction between precision measurement and reliability under faults.
Solution Approach 2:
The integrity monitor acts as an intermediary that compensates for the filter covariance's underestimation of errors during faults. It processes the same filtered solution but applies different analysis to detect fault conditions, providing the missing reliability information without interfering with the primary filtering function.
3Reliability
If integrity monitors are included at various stages in the processing chain, then reliability of position estimates is improved, but it is difficult to determine quantitatively the actual system integrity
Solution Approach 1:
The patent merges the integrity monitoring function with the existing filtering process by using the same filtered navigation solution and covariance as inputs. This consolidation eliminates the need for separate, information-loss-prone integrity determination steps, providing quantitative integrity information directly from the combined processing chain.
Data Source
AI summary
A method for modelling integrity of a filtered global navigation satellite system, by calculating component navigation system error distributions for a set of fault conditions and a fault free condition, and determining overall navigation error distribution by forming a mixture distribution from these component navigation system error distributions. The mixture distribution may be determined by weighted summation of navigation system error (NSE) distributions with weightings determined according to prior probabilities for the fault conditions. Once the overall NSE mixture distribution is determined in this way, it can be used to derive one or more statistical quantities relevant to the integrity of the navigation system such as the probability of exceeding given alert limits in a desired coordinate geometry.