Vehicle Initial Localization Using GNSS-Map and Sensor Grid Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately determining their initial location due to GPS errors, leading to incorrect matching and increased computational burdens, which can hinder autonomous driving services and restarts.
Innovation Solution
A method and apparatus that generate a first grid map from map information and a second grid map from sensor data, comparing them to determine the initial vehicle location by setting a heading search range, detecting lanes, generating heading candidates, and matching the grids to find a correlation peak, with the processor controlling the driving operation based on the determined location.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If GPS is used to determine initial location, then location determination is simple, but GPS errors cause incorrect location and reduced accuracy
Solution Approach 1:
The patent introduces map information and sensor data as intermediary elements between the GPS and the final location determination. Instead of relying solely on GPS coordinates, the system uses map matching and sensor fusion to verify and correct the initial GPS location, thereby maintaining operational simplicity while improving accuracy.
Solution Approach 2:
The patent replaces the purely GPS-based mechanical positioning system with a hybrid system that incorporates map information processing and sensor data fusion. This substitution allows the system to overcome GPS limitations by using multiple information sources to determine location accurately.
2Measurement precision
If full map matching is performed without reducing search range, then location accuracy is maximized, but computational burden increases significantly
Solution Approach 1:
The patent segments the map into multiple regions or tiles and divides the location determination process into stages. Instead of performing map matching across the entire map, the system first identifies a rough location using GPS, then performs detailed matching only in the surrounding area, significantly reducing computational complexity while maintaining accuracy.
Solution Approach 2:
The patent performs preliminary location estimation using GPS coordinates before conducting detailed map matching. This preliminary action narrows down the search area, allowing subsequent processing to focus only on relevant map regions and reducing the overall computational burden.
3Productivity
If heading search range is reduced to improve processing speed, then computational efficiency increases, but location determination accuracy may decrease
Solution Approach 1:
The patent performs preliminary analysis of sensor data and map information to estimate the vehicle's heading direction before conducting the actual map matching. This preliminary estimation allows the system to set an optimized search range that is narrow enough to ensure fast processing but wide enough to maintain accuracy by covering the likely heading direction.
Data Source
AI summary
An apparatus for determining an initial location of a vehicle is introduced. The apparatus may comprise a processor, and memory storing instructions, that, when executed by the processor, may cause the apparatus to receive, from a global navigation satellite system (GNSS), an initial GNSS location, generate, based on map information associated with the initial GNSS location, a first grid map, generate, based on sensing information from a sensor, a second grid map, detect, based on a comparison of the first grid map and the second grid map, a correlation peak for an initial location of a vehicle, and determine, based on the detected correlation peak, whether to set the initial location of the vehicle.