Localized Travel Lane Perception for Correcting Road Map Drift
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digitized maps of roadways often contain localized inaccuracies and become outdated due to road changes, which can affect the accuracy of navigation and operation of advanced driver assistance systems (ADAS) and autonomous vehicles.
Innovation Solution
A system utilizing a spatial sensor, GPS sensor, navigation system, and controller to determine the vehicle's position relative to a digitized map by comparing perceived point clouds from spatial sensors with mapped point clouds, using a Gaussian process and Kalman filter analysis to correct errors and update the map dynamically.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a digitized map is used for navigation and vehicle positioning, then the system can provide routing and location information, but the map contains localized inaccuracies and becomes outdated due to road changes
Solution Approach 1:
The system continuously compares spatial sensor data (perceived point cloud) with digitized map data (mapped point cloud) to generate error vectors that indicate deviations between actual and mapped road features. This feedback loop enables real-time detection of map inaccuracies and outdated information, allowing the system to identify where the map diverges from ground truth without requiring complete map redrawing.
Solution Approach 2:
The system performs preliminary alignment and transformation of spatial sensor data into the map coordinate system before comparison. By pre-processing the sensed data through coordinate transformations and initial matching algorithms, the system prepares the data for accurate comparison with the digitized map, enabling efficient detection of actual deviations rather than raw data comparison.
2Measurement precision
If the vehicle uses spatial sensors to monitor road segments in real-time, then the system can detect current road conditions, but the computational complexity increases due to point cloud processing and transformation
Solution Approach 1:
The system divides the road environment into discrete road segments and processes spatial data segment-by-segment rather than as a complete scene. By segmenting the point cloud data and processing individual road segments independently, the system reduces computational complexity while maintaining measurement precision for each segment's position and geometry.
Solution Approach 2:
The system extracts only the essential geometric features (road edges, lane markings, curvature) from the complete spatial point cloud data rather than processing all raw sensor information. This extraction of key features reduces the data volume requiring transformation and comparison with the digitized map, lowering processing complexity while preserving the information needed for accurate positioning.
3Reliability
If the system updates the digitized map dynamically using Gaussian process and Kalman filter, then the map accuracy improves over time, but the computational resources and processing time increase
Solution Approach 1:
The system implements periodic map updates at strategically selected locations and time intervals rather than continuous real-time updates across the entire map. By updating the digitized map periodically at key road segments where significant changes are detected, the system maintains improved accuracy while reducing the cumulative processing time and computational resource requirements compared to continuous full-map updates.
Data Source
AI summary
A system and method for monitoring a road segment includes determining a geographic position of a vehicle in context of a digitized roadway map. A perceived point cloud and a mapped point cloud associated with the road segment are determined. An error vector is determined based upon a transformation between the mapped point cloud and the perceived point cloud. A first confidence interval is derived from a Gaussian process that is composed from past observations. A second confidence interval associated with a longitudinal dimension and a third confidence interval associated with a lateral dimension are determined based upon the mapped point cloud and the perceived point cloud. A Kalman filter analysis is executed to dynamically determine a position of the vehicle relative to the roadway map based upon the error vector, the first confidence interval, the second confidence interval, and the third confidence interval.