Vehicle Localization Starting Position Using Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining a vehicle's initial position for lane-level digital maps face challenges in areas with limited or no GNSS signal availability, such as woods or tunnels, and require expensive differential GNSS hardware for improved accuracy.
Innovation Solution
A method using a control unit to combine measurement data from odometry, GNSS, LiDAR, radar, and camera sensors to determine a vehicle's starting position by comparing static features with a feature map, allowing for lane-level precision with inexpensive sensors and functionality in poor weather or GNSS-denied areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If differential GNSS methods are used to improve position accuracy, then measurement precision is improved, but device complexity and cost increase due to additional hardware requirements
Solution Approach 1:
The patent combines multiple sensor types (odometry sensors, standard GNSS receiver, LiDAR, radar, camera) into an integrated sensor system. By merging these different sensing modalities, the system achieves lane-level positioning accuracy without requiring expensive differential GNSS hardware, as the combination of sensors compensates for individual sensor limitations
Solution Approach 2:
The control unit performs multiple functions: it processes odometry data for position estimation, handles standard GNSS data when available, extracts features from LiDAR/radar/camera data, and compares features with map data. This multi-functional approach replaces specialized differential GNSS hardware with a versatile sensor fusion system that achieves comparable or superior accuracy
2Device complexity
If standard GNSS methods are used to determine initial position, then device complexity is kept low, but measurement precision deteriorates in areas with limited satellite signal availability
Solution Approach 1:
The system dynamically changes operational parameters based on environmental conditions. When GNSS signals are unavailable or inaccurate, the system transitions to using odometry-based position estimation combined with feature matching from LiDAR, radar, or camera sensors. This parameter adaptation allows maintaining lane-level accuracy across varying signal conditions without increasing hardware complexity
Solution Approach 2:
The patent introduces map data containing road boundary features as an intermediary reference system. When direct GNSS measurement is insufficient, the system uses odometry to estimate position and validates/corrects it by comparing extracted environmental features with stored map features. This intermediary map-based verification enables accurate positioning in GNSS-denied areas like tunnels and woods
3Measurement precision
If feature comparison is performed over large map sections to ensure accurate localization, then measurement precision is improved, but productivity decreases due to increased computational demands
Solution Approach 1:
The patent applies local quality by focusing feature comparison on specific, relevant portions of map data. Instead of comparing features across entire map sections, the system extracts features from the current sensor view and compares them with corresponding features in the map data at the estimated position. This localized approach maintains lane-level localization accuracy while dramatically reducing computational requirements
Data Source
AI summary
A method for ascertaining a starting position of a vehicle for a localization of the vehicle using a control unit. In the method, measurement data are received from an odometry sensor system and/or a GNSS sensor system of the vehicle, a first position and an uncertainty range of the first position are determined based on the measurement data received, at least one map section of a feature map containing a plurality of stored features is received, the map section having a position and extent which is superimposed on the first position and the uncertainty range, measurement data are received from a LiDAR sensor system, a radar sensor system and/or a camera sensor system and static features are extracted from the measurement data received, a first starting position of the vehicle is ascertained by comparing the static features extracted from measurement data with features stored in the map section.