Radar Time Synchronization for GNSS-Denied Vehicle Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle localization methods rely on GNSS and inertial measurement units, which can be computationally expensive and prone to errors, especially in areas with reduced signal reception or tunnel environments.
Innovation Solution
The use of machine learning models that generate lane indices based on real-time image data from sensors, such as LiDAR and cameras, to improve lane offset detection and vehicle localization, reducing reliance on potentially inaccurate GNSS data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If GNSS and inertial measurement unit are used for localization, then location information can be obtained, but computational cost increases and accuracy deteriorates in areas with reduced signal reception
Solution Approach 1:
The patent introduces map features (lane markings, road signs, infrastructure elements) as an intermediary reference system. Instead of relying solely on GNSS signals that fail in tunnels or urban canyons, the system uses these map features as a mediator to establish the vehicle's position through image recognition and matching against stored map data, thereby resolving the contradiction between reliability and precision in signal-denied environments
Solution Approach 2:
The patent replaces the mechanical/GNSS-based localization system with a vision-based system using cameras and image processing. This substitution allows the vehicle to determine its position through optical recognition of map features rather than relying on satellite signals that are blocked in certain environments, thus improving both reliability and accuracy in tunnel and urban areas
2Reliability
If GNSS and inertial measurement unit are used for localization, then location information can be obtained, but computational resources are consumed excessively
Solution Approach 1:
The patent extracts and utilizes pre-existing map data (lane markings, road signs, infrastructure) that is already available in the environment, rather than computing localization from scratch using heavy inertial measurement unit processing. This extraction approach reduces computational energy consumption while maintaining localization availability by leveraging external reference information
Solution Approach 2:
The patent creates a digital representation (copy) of the physical map features through image capture and processing. By copying visual information from the environment and matching it against stored map data, the system achieves localization with lower computational energy requirements compared to continuous processing of inertial measurement data
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
This approach enhances the accuracy and reliability of autonomous vehicle localization by utilizing image data to determine lane positions, even in environments where GNSS signals are weak or unavailable, thereby improving navigation and maneuvering capabilities.
Implementation Method 1
a plurality of radar sensors arranged to provide a three-sixty degree field of view about the autonomous vehicle
Implementation Method 2
image data captured in real time or near real time by sensors of the autonomous vehicle
Data Source
AI summary
A method comprises instructing, by a processor, a time signal from a grand master clock to be transmitted to a second processor associated with a radar sensor of an autonomous vehicle; instructing, by the processor, the second processor associated with the radar sensor of the autonomous vehicle to sync an internal clock with the time signal; and retrieving, by the processor, radar data from the radar sensor.