Lane-Level Road Mapping Using GPS, IMU, and Camera Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies for generating precise lane map data in autonomous vehicles face challenges such as indistinct lane markings, weather obstructions, and the need for highly accurate pre-mapped terrain features, which are difficult to achieve with existing methods like adaptive lane detection, precision GPS, and 3D simultaneous localization and mapping.
Innovation Solution
An in-vehicle system combining a GPS receiver, inertial sensor, and camera to acquire and improve positional data through curve fitting, allowing for precise lane-level road map generation, with the ability to update maps with transient road features by collecting data from multiple vehicles and transmitting it to a central repository.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If adaptive lane detection is used, then the system can operate without precision maps, but lane detection fails when markings are indistinct or obscured by weather/obstructions
Solution Approach 1:
The patent introduces an inertial measurement unit (IMU) as an intermediary device that bridges the gap between GPS data and visual lane detection. The IMU provides complementary information about vehicle motion and orientation that helps disambiguate visual data when lane markings are indistinct or obscured, thereby maintaining detection reliability while operating without precision maps
Solution Approach 2:
The system merges multiple data sources including GPS positional information, inertial sensor data from the IMU, and computer vision output into a unified lane detection framework. This fusion of heterogeneous data sources compensates for the weaknesses of individual methods, enabling reliable operation without precision maps even when visual markings are poor
2Measurement precision
If precision GPS (DGPS) augmented by inertial sensing is used, then vehicle position can be kept in lane by referencing precision GPS coordinate maps, but the required accuracy is too high for aerial and satellite based road mapping
Solution Approach 1:
The patent applies local quality by using inertial sensors to provide high-precision local position estimates specifically for lane keeping, while relying on lower-precision GPS for general positioning. This localized application of high-precision sensing avoids the need for comprehensive high-precision mapping of entire road networks
Solution Approach 2:
The system segments the positioning function into multiple components: GPS provides coarse positional information, inertial sensors provide fine-grained local position updates, and computer vision provides lane context. This segmentation allows each component to operate at its appropriate precision level, avoiding the need for uniformly high-precision mapping
3Measurement precision
If 3D simultaneous localization and mapping is used, then vehicle position can be correlated relative to surrounding pre-mapped terrain features, but 100% of road surround must be mapped to 2 cm to 10 cm accuracy
Solution Approach 1:
The patent extracts only the essential terrain features needed for lane detection and navigation, rather than mapping 100% of the road surround. By selecting and extracting only relevant features (such as road edges and significant landmarks), the system achieves sufficient positioning accuracy without the prohibitive data volume and processing requirements of complete environmental mapping
4Area of stationary object
If aerial and satellite based road mapping is used, then coverage area is extensive, but the accuracy is insufficient for lane-level precision requirements
Solution Approach 1:
The patent transitions from two-dimensional aerial/satellite mapping to three-dimensional positioning by incorporating inertial sensing and computer vision. This dimensional enhancement allows the system to achieve lane-level accuracy (vertical and lateral precision) while maintaining extensive geographic coverage, effectively adding the precision dimension without sacrificing coverage area
Data Source
AI summary
An in-vehicle system for generating precise, lane-level road map data includes a GPS receiver operative to acquire positional information associated with a track along a road path. An inertial sensor provides time local measurement of acceleration and turn rate along the track, and a camera acquires image data of the road path along the track. A processor is operative to receive the local measurement from the inertial sensor and image data from the camera over time in conjunction with multiple tracks along the road path, and improve the accuracy of the GPS receiver through curve fitting. One or all of the GPS receiver, inertial sensor and camera are disposed in a smartphone. The road map data may be uploaded to a central data repository for post processing when the vehicle passes through a WiFi cloud to generate the precise road map data, which may include data collected from multiple drivers.