Vehicle Positioning via Object Size Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle positioning systems, especially for autonomous vehicles, face challenges in achieving accurate and reliable longitudinal and lane positioning due to limitations in cost-effective GNSS sensors, requiring the integration of multiple sensors and complex data processing.
Innovation Solution
An object size mapping system that uses a reference camera on-board a vehicle to capture images of stationary physical objects, determining their size values, and associating these with digital map data to improve vehicle positioning accuracy by linking reference size values with specific positions, enabling more precise estimation of vehicle location.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If cost effective GNSS sensors are used for vehicle positioning, then the system cost is reduced, but the positioning accuracy and reliability deteriorate
Solution Approach 1:
The patent combines multiple data sources including GNSS positioning, map matching algorithms, and visual recognition of road features (lane markings, signs, buildings) to create a integrated positioning system. This fusion of multiple low-cost components achieves positioning accuracy comparable to expensive dedicated systems while maintaining cost-effectiveness.
Solution Approach 2:
The patent introduces map data as an intermediary layer between the GNSS sensor and the final position determination. By matching GNSS coordinates with pre-stored map information and refining the position based on visual features from the map, the system enhances positioning accuracy without requiring expensive high-precision hardware.
2Measurement precision
If multiple sensors are integrated to improve positioning accuracy, then the positioning precision is improved, but the device complexity increases
Solution Approach 1:
The patent makes the camera system multi-functional by using it for both visual recognition of road features and for positioning refinement. The same image processing pipeline serves multiple purposes: identifying lane markings for lane positioning, recognizing signs for location verification, and detecting buildings for map matching, thereby reducing the need for separate dedicated sensors.
Solution Approach 2:
The system uses freely available map data and publicly accessible road feature information as self-service resources. Instead of requiring expensive dedicated sensors, the system leverages open-source map databases and standard computer vision techniques that are already widely available, reducing both hardware complexity and system cost.
3Reliability
If map matching with sensor detected objects is used to improve positioning, then the positioning reliability is improved, but the difficulty of detecting and measuring increases
Solution Approach 1:
The patent pre-processes and stores map data including positions of road signs, buildings, and lane markings before runtime. During actual positioning, the system only needs to detect and match these pre-identified features rather than performing complex analysis from scratch, significantly reducing the real-time detection and measurement difficulty while maintaining high positioning reliability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure relates to a method performed by an object size mapping system (1) for enabling improved positioning of a vehicle (12). The object size mapping system comprises at least a first reference camera (3) adapted to be arranged on-board a reference vehicle (2). The object size mapping system determines (1001) a current reference position (21) of the reference vehicle. Furthermore, the object size mapping system captures (1002) by means of the at least first reference camera, at the current reference position of the reference vehicle, a current reference image (31) of a stationary physical reference object (4) situated in the surroundings of the reference vehicle. The object size mapping system then determines (1003) a current reference size value (311) of at least a portion of the stationary physical reference object, in the current reference image. Moreover, the object size mapping system stores (1005) the current reference size value to be associated with the current reference position of the reference vehicle and a mapped digital reference object (511) corresponding to the stationary physical reference object. The disclosure also relates to an object size mapping system in accordance with the foregoing, a vehicle positioning system (11) and method performed therein, and a vehicle (2, 12) comprising at least a portion of the object size mapping system and/or vehicle positioning system.