Height-Constrained Extended Kalman Filter for GNSS Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional satellite positioning methods require external auxiliary information and devices, such as sensors, to improve height accuracy, increasing costs and complexity.
Innovation Solution
A height-constraint-based extended Kalman filter method that estimates and corrects state values using a prediction formula, height constraint conditions, and Kalman filter gain matrix to optimize GNSS navigation without external sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If external auxiliary devices such as sensors and barometers are used to improve height positioning accuracy, then positioning accuracy in the height direction is improved, but device complexity and cost increase
Solution Approach 1:
The system uses itself to provide the constraint information needed for height positioning. By utilizing the geometric relationship between satellite signals and the receiver, and applying height constraint conditions through the EKF algorithm, the system achieves improved height accuracy without requiring external auxiliary devices like barometers or sensors.
Solution Approach 2:
The invention extracts and utilizes only the necessary height constraint information from the satellite positioning system itself, rather than relying on external devices. By separating the height constraint function from external sensors and implementing it through algorithmic constraints within the EKF framework, the system achieves the same effect without additional hardware.
2Measurement precision
If external auxiliary information such as heading and real-time speed is used to improve height positioning accuracy, then positioning accuracy in the height direction is improved, but information requirements and processing complexity increase
Solution Approach 1:
The system generates and uses its own motion constraint information derived from the satellite positioning data and EKF algorithm, rather than requiring external input of heading and speed information. The height constraint condition is formulated based on the geometric relationship between consecutive positioning epochs, eliminating the need for external auxiliary information.
3Ease of operation
If conventional Kalman filter algorithm is used without height constraints, then computational simplicity is maintained, but height positioning accuracy deteriorates due to dramatic changes in height direction
Solution Approach 1:
The invention dynamically adjusts the filtering process by introducing height constraint conditions that adapt to the geometric relationship between satellite signals and receiver motion. The constraint condition is formulated based on the angle between the satellite-receiver line and the height direction, allowing the system to maintain computational simplicity while improving height accuracy through dynamic constraint application.
Solution Approach 2:
The invention changes the parameter space by introducing height constraint conditions into the EKF algorithm. By formulating and applying the height constraint condition as an additional mathematical constraint, the system modifies the estimation process to produce more accurate height values without fundamentally changing the EKF framework, thus maintaining computational tractability.
Data Source
AI summary
A positioning method using height-constraint-based extended Kalman filter, suitable for a GNSS navigation and positioning system, comprises: obtaining an estimated state value of a current epoch by using an extended Kalman filter algorithm and according to an estimated state value of a previous epoch; constraining a positioning height of the current epoch by establishing a height constraint condition, so as to obtain an optimum estimated value of the current epoch and a corresponding mean square error, wherein the optimum estimated value satisfies the height constraint condition; further correcting the estimated state value by using a pseudorange obtained from the mean square error and a measured Doppler shift residual to obtain a final estimated state value of the current epoch, thus more accurately obtaining positioning information of a target to be positioned in the current epoch and enhancing the accuracy of GNSS navigation and positioning.