INS Quadratic Correction via Radar MLE Motion Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current INS systems face measurement biases that impact the quality of SAR images, especially in wide area imaging and short wavelengths, and require corrections relative to the scene center, which existing autofocus techniques, such as maximum entropy and phase-gradient autofocus, fail to adequately address, particularly in Ku band wavelengths and large scene sizes.
Innovation Solution
A computer-implemented method using maximum likelihood estimation (MLE) to calculate three-dimensional residual motion errors of a moving platform by receiving radar signals, forming radar images, and recursively correcting the location and velocity of scatterers relative to the scene center, thereby updating the platform's radial acceleration and velocity, which is then used to improve INS data accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional autofocus techniques (maximum entropy, phase-gradient) are used to correct INS errors, then the processing speed is improved, but the measurement precision of motion parameters deteriorates due to spatially varying quadratic phase in wide area imaging
Solution Approach 1:
The patent changes the estimation approach from conventional autofocus methods to maximum likelihood estimation, transforming the parameter estimation problem into a statistical optimization framework that jointly estimates range, radial velocity, and radial acceleration while accounting for spatially varying quadratic phase effects
Solution Approach 2:
The patent extends the estimation from 2D range-velocity space to 3D range-velocity-acceleration space, adding the acceleration dimension to resolve the limitations of conventional autofocus techniques in wide area imaging with spatially varying phase
2Device complexity
If INS data is used directly for SAR imaging, then the device complexity is reduced, but the manufacturing precision of SAR images deteriorates due to measurement biases in INS systems
Solution Approach 1:
The patent enables the radar system to self-correct INS measurement biases by using the radar return signals themselves to estimate and compensate for range, velocity, and acceleration errors through maximum likelihood estimation, making the system self-correcting without external references
Solution Approach 2:
The patent implements a feedback mechanism where the estimated motion parameters from radar signals are used to update and correct the INS data, creating a closed-loop system that continuously improves navigation accuracy for SAR imaging
3Measurement precision
If MLE technique is used to estimate target motion parameters, then the measurement precision is improved, but the device complexity increases due to finding the maximum of a nonlinear 3D likelihood function
Solution Approach 1:
The patent segments the 3D likelihood function optimization into a series of 2D optimizations at different range-velocity projections, breaking down the complex 3D problem into manageable 2D steps that can be solved more efficiently while maintaining accuracy
Data Source
AI summary
System and method for calculating three dimensional residual motion errors of a moving platform with respect to a point of interest by receiving a radar signal from the point of interest (302); forming a radar image including a plurality of scatterers (304); using an MLE method to obtain range, radial velocity and acceleration of the moving platform for a first peak scatterer in the radar image (306); correcting a location of the first peak scatterer with respect to a scene center of the point of interest (312); updating the obtained radial acceleration responsive to the corrected location (314); and updating the obtained radial velocity of the moving platform responsive to the updated radial acceleration (316).