GNSS Location Validation with Sensor Fusion for HD Map Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately determining their location due to the inaccuracies of Global Navigation Satellite System (GNSS) data, which can be obscured by environmental factors, and conventional maps lack the required precision for safe navigation.
Innovation Solution
A system that combines GNSS location data with additional sensor data such as acceleration, image, and LIDAR data to generate a probabilistic model and confidence scores, validating the location data for improved navigation and HD map updates using machine learning models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GNSS location data is used for autonomous vehicle navigation, then the system can determine vehicle position globally, but the location accuracy deteriorates to over 100 meters due to large error conditions
Solution Approach 1:
The patent combines GNSS location data with multiple sensor data sources (IMU, LIDAR, cameras) to create a fused location estimation system. This merging of multiple data sources compensates for the poor accuracy of individual GNSS measurements, achieving the required 10 cm or less accuracy for safe autonomous navigation.
Solution Approach 2:
The system uses probabilistic models and confidence scores to continuously evaluate and validate location data quality. By feeding back the confidence score from the probabilistic model, the system can identify when GNSS data is unreliable and switch to alternative localization methods or request additional sensor data to improve accuracy.
2Measurement precision
If conventional maps are used for autonomous vehicle navigation, then the system can provide route guidance, but the map precision is insufficient for safe navigation requiring 10 cm or less accuracy
Solution Approach 1:
The patent segments the localization problem into multiple components: GNSS provides coarse global position, IMU provides short-term motion tracking, LIDAR and cameras provide environmental feature matching. Each sensor handles a specific aspect of localization, and their results are fused to achieve high precision without requiring a single complex system.
Solution Approach 2:
The system transitions from relying solely on 2D map coordinates to incorporating 3D spatial information from LIDAR point clouds and camera images. By adding these dimensional layers of environmental data, the system achieves precise localization even when conventional 2D maps lack sufficient detail.
3Measurement precision
If multiple sensor data sources are integrated to improve location accuracy, then the localization precision improves to HD map level, but the system complexity increases
Solution Approach 1:
The probabilistic model and confidence score validation system serve multiple functions: they evaluate GNSS data quality, validate sensor fusion results, and determine when to trust location data for navigation. This universal validation mechanism simplifies the overall system architecture by providing a single framework that handles multiple data sources and decision-making scenarios.
Data Source
AI summary
A vehicle computing system validates location data received from a Global Navigation Satellite System receiver with other sensor data. In one embodiment, the system calculates velocities with the location data and the other sensor data. The system generates a probabilistic model for velocity with a velocity calculated with location data and variance associated with the location data. The system determines a confidence score by applying the probabilistic model to one or more of the velocities calculated with other sensor data. In another embodiment, the system implements a machine learning model that considers features extracted from the sensor data. The system generates a feature vector for the location data and determines a confidence score for the location data by applying the machine learning model to the feature vector. Based on the confidence score, the system can validate the location data. The validated location data is useful for navigation and map updates.