Vehicle Lane Localization via Sensor Fusion and HD Map Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous driving systems face inaccuracies in vehicle lane level localization due to insufficient and noisy data, particularly in camera-only based systems, leading to inaccurate automated lane keeping and lane changing.
Innovation Solution
A lane level localization system that combines a plurality of perception sensors, including a GNSS receiver, RTK system, IMU, and cameras, with a high-definition map system to detect and align lane lines, using weighted matching and filtering to improve localization accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If camera-only based systems are used for autonomous driving, then device complexity is reduced, but measurement precision of vehicle lane level localization deteriorates
Solution Approach 1:
The patent combines multiple perception sensors (cameras, LIDAR, radar, GNSS, IMU) with HD map data to create a fused localization system. This merging of multiple data sources compensates for the weaknesses of individual sensors and achieves high-precision lane level localization that exceeds what camera-only systems can provide.
Solution Approach 2:
The system creates a composite localization solution by integrating data from heterogeneous sensor types (optical cameras, electromagnetic LIDAR/radar, satellite GNSS, mechanical IMU) with digital HD map information. This composite approach leverages the strengths of each sensor type to achieve robust and precise localization.
2Measurement precision
If multiple perception sensors and HD map systems are combined, then measurement precision of lane level localization is improved, but device complexity increases
Solution Approach 1:
The controller is designed as a multi-functional unit that performs diverse tasks: processing data from multiple sensor types, aligning lane lines from different sources, fusing perception data with HD map data, and generating localized vehicle position information. This universal controller consolidates complexity into a single coordinating component.
Solution Approach 2:
The system introduces an intermediary alignment process that matches lane lines from perception sensors with lane lines from HD map data. This intermediary step reconciles the different coordinate systems and data formats from multiple sources, simplifying the integration process and enabling precise localization despite the complexity of multiple inputs.
3Ease of operation
If conventional camera-only systems are used, then ease of operation is maintained, but reliability of automated lane keeping and lane changing deteriorates
Solution Approach 1:
The system continuously compares lane line detections from perception sensors with expected lane lines from HD map data, using this feedback to correct localization errors in real-time. This feedback mechanism enhances reliability by constantly validating and adjusting the vehicle's perceived position against the known map structure.
Solution Approach 2:
The system pre-aligns and matches lane lines from multiple sources before using them for automated driving decisions. By preparing and validating the aligned lane line data in advance, the system cushions against potential localization failures during critical lane keeping and lane changing operations.
Data Source
AI summary
Lane level localization techniques for a vehicle utilize a plurality of perception sensor systems each configured to perceive a position of the vehicle relative to its environment, a map system configured to maintain map data that includes lane lines, and a controller configured to detect a position of the vehicle and a first set of lane lines using the plurality of perception sensors, detect a second set of lane lines using the position of the vehicle and the map data, obtain an aligned set of lane lines based on the first and second sets of lane lines, and use the aligned set of lane lines for an autonomous driving feature of the vehicle.