Radar Angle Bias Correction for Autonomous Sensor Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for identifying and correcting angle bias errors in sensor data for autonomous vehicles are computationally expensive and require specific reference geometries or driving maneuvers, limiting their effectiveness and accuracy.
Innovation Solution
A method for identifying and correcting angle bias errors in sensor data using sensor data collected during normal driving operations, without requiring specific reference geometries or driving maneuvers, by generating a representation of angle bias errors and applying corrections to raw angle measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods (reference geometry or Monte-Carlo sampling) are used to identify angle bias errors, then measurement precision may be improved, but computational cost and device complexity increase significantly
Solution Approach 1:
The patent extracts the angle bias error identification from complex conventional methods (reference geometry requirements, Monte-Carlo sampling) and implements it through a simplified neural network-based approach that processes normal sensor data during regular driving operations, eliminating the need for special reference geometries or computationally intensive sampling procedures
Solution Approach 2:
The system uses the autonomous vehicle's own normal sensor data collected during regular driving operations to identify and correct angle bias errors, eliminating the need for external reference geometries or additional specialized equipment, thereby reducing implementation complexity while maintaining measurement precision
2Measurement precision
If conventional methods (reference geometry or Monte-Carlo sampling) are used to identify angle bias errors, then measurement precision may be improved, but processing time and computational resources increase
Solution Approach 1:
The neural network is trained in advance to recognize angle bias patterns, enabling real-time correction during normal driving operations without requiring computationally intensive processing during actual error identification, thereby reducing processing time while maintaining measurement precision
Solution Approach 2:
The patent replaces computationally intensive mechanical/mathematical methods (Monte-Carlo sampling, continuous geometric calculations) with a neural network-based system that processes sensor data more efficiently, reducing computational resources and processing time while achieving comparable or superior measurement precision
3Ease of manufacture
If RADAR sensor is covered via radome for aesthetic design, then ease of manufacture and design flexibility are improved, but sensor data accuracy deteriorates due to beam distortions
Solution Approach 1:
The neural network is trained to account for the asymmetric distortion patterns introduced by the radome cover, learning to correct for the specific beam distortions caused by the radome's structural characteristics, thereby recovering measurement precision while preserving design flexibility
Solution Approach 2:
The system changes the parameters of angle measurements by applying neural network-based corrections that compensate for radome-induced distortions, transforming biased measurements into accurate data while maintaining the aesthetic radome cover structure
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate and efficient identification of angle data for improved navigation and decision-making in autonomous systems, reducing noise and increasing accuracy without the need for additional implementation limitations.
Implementation Method 1
RADAR sensors may be used, for example, to obtain sensor data that facilitates localization... a RADAR sensor(s) may be positioned on an ego-machine to emit radio waves (e.g., via a transmitter antenna) and receive reflected signals (e.g., via receiver antenna) to detect various types of sensor data or measurements
Data Source
AI summary
In various example, embodiments are directed to angle bias error identification and correction for autonomous and semi-autonomous systems and applications. Systems and methods are disclosed that identify angle bias error(s) associated with detected sensor data and correct for such angle bias error(s) for use in localization, navigation, and/or other uses by autonomous vehicles, semi-autonomous vehicles, robots, and/or other object or machine types. In embodiments, angle bias error identification is performed by detecting angle error in association with various points detected via a sensor during normal driving operation of an ego-machine. The detected angle errors may be used to generate a representation of angle bias error for various angles of the sensor, which may be used to apply a correction to raw angle measurements. Using techniques described herein, for example, corrected azimuth angle measurements may be generated for use by downstream modules to perform more efficient and effective navigation.