GNSS Positioning Accuracy via HD Map Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional GNSS-based positioning for autonomous vehicles is plagued by inaccuracies due to errors in satellite positions, clocks, and signal propagation, and existing enhancement techniques like RTK and SBAS have limitations in cost, coverage, and reliability, making them inadequate for high-accuracy navigation.
Innovation Solution
An autonomous vehicle periodically determines RTK corrections based on its location using localization and raw GNSS signals, which are then applied to improve subsequent GNSS position estimates, and these corrections can also be used to initialize and refine localization algorithms with sensor data from lidar, cameras, and inertial measurements, enabling more accurate and reliable navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional GNSS-based positioning is used, then the system is simple and cost-effective, but the positioning accuracy is insufficient (3-5 meters)
Solution Approach 1:
The system performs preliminary localization using sensor data (lidar, cameras, inertial measurements) and high-definition maps to estimate vehicle position before refining with GNSS. This preliminary action provides an accurate initial guess that enables the GNSS receiver to achieve high precision positioning without requiring complex RTK infrastructure, thus resolving the contradiction between positioning accuracy and system complexity
Solution Approach 2:
The patent introduces sensor data fusion (lidar, cameras, inertial measurements) as an intermediary between conventional GNSS and the final position estimate. This intermediary layer processes raw sensor data to generate accurate localization estimates that correct GNSS errors, achieving high positioning accuracy without the complexity of multiple GNSS receivers or external correction infrastructure
2Measurement precision
If RTK enhancement is used, then positioning accuracy is improved, but the system requires multiple receivers and extensive infrastructure increasing cost and complexity
Solution Approach 1:
The patent extracts the localization function from the GNSS system itself and implements it using onboard sensors (lidar, cameras, inertial measurements) combined with high-definition maps. This extraction eliminates the need for multiple GNSS receivers and external RTK infrastructure, achieving high positioning accuracy with a single GNSS receiver while reducing system complexity
Solution Approach 2:
The vehicle serves itself by using its own onboard sensors and stored high-definition map data to generate accurate localization estimates. This self-service approach eliminates dependence on external RTK infrastructure and multiple receivers, achieving high positioning accuracy while simplifying the system configuration to a single GNSS receiver combined with sensor fusion
3Measurement precision
If RTK correction is applied, then positioning accuracy is enhanced, but coverage is limited geographically and reliability decreases when moving outside coverage areas
Solution Approach 1:
The patent implements a universal localization system that combines sensor fusion with high-definition map matching, enabling accurate positioning anywhere the vehicle has traveled before to collect map data. This multi-functional approach replaces geography-limited RTK correction with a system that adapts to any environment the vehicle has previously mapped, significantly expanding geographic coverage and reliability
Solution Approach 2:
The system performs preliminary collection of high-definition map data as the vehicle travels through an environment. This preliminary action creates a reusable localization reference that enables accurate positioning in future visits without requiring external correction infrastructure, thus expanding geographic coverage beyond RTK limitations while maintaining high positioning accuracy
4Measurement precision
If RTK enhancement is used, then positioning accuracy is improved, but a data link must be maintained adding complication and cost
Solution Approach 1:
The patent extracts the positioning correction function from external communication infrastructure and implements it using onboard sensor data processing. By taking out the dependency on data links for correction data transmission, the system achieves high positioning accuracy without communication infrastructure, eliminating the associated complexity and cost
Solution Approach 2:
The vehicle uses its own onboard sensors and computational resources to generate positioning corrections locally through sensor fusion and map matching. This self-service approach eliminates the need for external data links and correction servers, achieving high positioning accuracy while reducing system complexity and removing communication infrastructure dependencies
Data Source
AI summary
A vehicle, for example, an autonomous vehicle receives signals from a global navigation satellite system (GNSS) and determines accurate location of the vehicle using the GNSS signal. The vehicle performs localization to determine the location of the vehicle as it drives. The autonomous vehicle uses sensor data and a high definition map to determine an accurate location of the autonomous vehicle. The autonomous vehicle uses accurate location of the vehicle to determine RTK corrections that is used for improving GNSS location estimates at a future location. The RTK corrections may be transmitted to other vehicles.