Radar Sensor Alignment Angle Estimation via Linearized Kalman Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining the alignment angles of road vehicle radar sensors are inefficient, particularly when longitudinal velocity, lateral velocity, and yaw-rate data are unknown, limiting the ability to correct sensor misalignment without mechanical adjustments.
Innovation Solution
A method using signals from at least two radar sensors to derive a linearized signal processing model, applying a Kalman filter algorithm to estimate alignment angles, and producing signals for radar offset compensation, allowing for efficient and swift estimation of alignment angles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional methods using constrained least squares problem are used to estimate alignment angles, then measurement precision can be improved, but device complexity and computational requirements increase significantly
Solution Approach 1:
The patent transforms the non-linear constrained least squares estimation problem into a linear least squares problem by changing the mathematical parameters and formulation. This linearization simplifies the computational complexity while maintaining the ability to estimate alignment angles accurately, resolving the contradiction between measurement precision and device complexity
Solution Approach 2:
The patent replaces the complex mechanical/mathematical optimization system (constrained least squares with multiple constraints) with a simpler linear algebraic system. This substitution reduces computational burden and implementation complexity while preserving the essential function of alignment angle estimation
2Device complexity
If separate alignment procedures are used for each radar sensor, then device complexity is reduced, but measurement precision and overall system accuracy deteriorate
Solution Approach 1:
The patent merges the alignment estimation for multiple radar sensors into a single unified linear least squares framework. By combining measurements from multiple sensors and performing joint estimation, the system achieves higher accuracy than separate procedures while maintaining relatively simple computational structure
Solution Approach 2:
The patent creates a universal alignment estimation framework that simultaneously handles multiple radar sensors with different configurations. This multi-functional approach estimates alignment angles for all sensors through a single computational process, improving overall system accuracy without proportionally increasing complexity
3Measurement precision
If recursive geometric calculations are used to converge on actual sensor pose values, then measurement precision improves, but productivity and processing speed decrease
Solution Approach 1:
The patent performs preliminary linearization of the geometric relationships before actual estimation, transforming the problem into a form that can be solved directly through linear least squares. This preliminary transformation eliminates the need for iterative recursive calculations, achieving both high precision and fast processing speed
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
A method of determination of the alignment angles of two or more road vehicle (1) borne radar sensors (4) for a road vehicle radar auto-alignment controller (3) starting from initially available rough estimates of alignment angles. From at least two radar sensors (4) are obtained signals related to range, azimuth and range rate to detections. The detections are screened (5) to determine detections from stationary targets. From the determined detections from stationary targets is derived a linearized signal processing model involving alignment angles, longitudinal and lateral velocity and yaw-rate of the road vehicle (1). A filter algorithm is applied to estimate the alignment angles. Based on the estimated alignment angles are produced signals suitable for causing a road vehicle (1) radar auto-alignment controller (3) to perform radar offset compensation.