Nonlinear Navigation Module Error-State Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current navigation systems using INS/GNSS integration, especially with low-cost MEMS-based sensors, suffer from significant errors due to nonlinear models and misalignment issues, leading to divergence and inaccurate positioning in GNSS-degraded or denied environments.
Innovation Solution
A navigation module employing a nonlinear filtering technique, such as the Mixture Particle Filter, integrates inertial sensors with GNSS information using nonlinear error-state or total-state models to reduce errors and account for sensor misalignment, allowing for accurate navigation even in environments with limited satellite coverage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional Kalman Filter-based methods are used for INS/GNSS integration, then the system is simpler to implement, but positioning accuracy deteriorates significantly in GNSS-degraded or denied environments due to nonlinear model errors and sensor misalignment
Solution Approach 1:
The patent transforms the state estimation problem by changing the parameters being estimated from absolute states to error states (differences between true and estimated values). This error-state formulation allows the nonlinear filtering approach to focus on correcting deviations rather than computing complete state vectors, improving accuracy while managing complexity through targeted parameter transformation
Solution Approach 2:
The patent introduces an intermediary error model that mediates between the complex nonlinear sensor models and the filtering algorithm. By modeling and compensating for misalignment errors and nonlinear effects as separate intermediate components, the system achieves higher positioning accuracy without requiring the filter to directly handle all nonlinear complexities
2Reliability
If low-cost MEMS-based sensors are used, then the device cost is reduced, but positioning accuracy deteriorates due to significant sensor errors and drift in nonlinear integration
Solution Approach 1:
The patent implements feedback through the error-state formulation where sensor errors and misalignment deviations are continuously estimated and fed back to correct the navigation solution. The nonlinear filtering algorithm uses feedback from GNSS measurements when available to adjust error estimates, maintaining reliability of low-cost MEMS sensors while compensating for their inherent inaccuracies through continuous error correction
Solution Approach 2:
The patent converts the harmful effect of sensor misalignment and nonlinear errors into beneficial information by explicitly modeling these errors as state variables. The misalignment between sensor coordinate frames, which traditionally causes accuracy degradation, is transformed into a measurable and correctable parameter through the error-state model, turning a source of error into a source of correction information
3Measurement precision
If device misalignment with the moving platform is not accounted for, then the system complexity is reduced, but positioning accuracy deteriorates due to coordinate frame misalignment errors
Solution Approach 1:
The patent merges the misalignment parameter estimation with the standard state estimation process by incorporating orientation errors and misalignment angles into the unified error-state vector. This combining of misalignment compensation with regular navigation state estimation allows the system to achieve high positioning accuracy without requiring separate complex coordinate transformation systems, as both functions are handled within the integrated nonlinear filtering framework
Data Source
AI summary
A navigation module and method for providing an INS/GNSS navigation solution for a device that can either be tethered or move freely within a moving platform is provided, comprising a receiver for receiving absolute navigational information from an external source (e.g., such as a satellite), an assembly of self-contained sensors capable of obtaining readings (e.g. such as relative or non-reference based navigational information) about the device, and further comprising at least one processor, coupled to receive the output information from the receiver and sensor assembly, and operative to integrate the output information to produce an enhanced navigation solution. The at least one processor may operate to provide a navigation solution by benefiting from nonlinear models and filters that do not suffer from approximation or linearization and which enhance the navigation solution of the device.