Vehicle-Roadside Data Fusion for Extended Autonomous Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The sensing range of vehicle sensing apparatuses in self-driving vehicles is narrow, making it difficult to meet the requirements for effective road environment sensing.
Innovation Solution
A data fusion method that combines roadside sensing data with vehicle sensing data using a fusion formula to extend the sensing range by integrating confidence factors based on apparatus parameters, sensing distances, and angles, and utilizing neural networks for matching and correlation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If only vehicle sensing apparatus is used for road environment sensing, then the system structure is simple, but the sensing range is narrow and cannot meet self-driving requirements
Solution Approach 1:
The patent combines vehicle-mounted sensing apparatus with roadside sensing apparatus to form a collaborative sensing system. The roadside sensing apparatus is deployed along the road to provide extended coverage, while the vehicle-mounted apparatus provides mobile sensing capability. Their sensing ranges are overlapped and fused to achieve comprehensive road environment perception that neither system could achieve alone.
2Area of stationary object
If roadside sensing apparatus is added to extend sensing range, then the sensing coverage is improved, but the data processing complexity increases
Solution Approach 1:
The patent segments the sensing system into distinct functional modules: vehicle-mounted sensing apparatus, roadside sensing apparatus, and a fusion center. Each module independently processes data within its sensing range, and the fusion center integrates these segmented data streams using data fusion algorithms. This segmentation allows parallel processing and reduces the computational burden on any single component.
Solution Approach 2:
The patent introduces a fusion center as an intermediary component that receives data from both vehicle-mounted and roadside sensing apparatus. This mediator performs data association, fusion, and management operations, separating the complex data processing tasks from the sensing apparatus themselves and enabling scalable system expansion.
3Measurement precision
If data fusion is performed to integrate multiple sensing sources, then the sensing accuracy and confidence are improved, but the computational requirements increase
Solution Approach 1:
The patent performs preliminary data processing and feature extraction at the sensing apparatus level before data fusion. The vehicle-mounted and roadside sensing apparatus pre-process their respective data streams, extracting relevant features and filtering out redundant information. This preliminary action reduces the dimensionality and complexity of data requiring fusion, lowering computational requirements while maintaining fusion accuracy.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
This application discloses a data fusion method and a related device. The method includes: obtaining vehicle sensing data, where the vehicle sensing data is obtained by a vehicle sensing apparatus by sensing a road environment in a sensing range by using a vehicle sensor; obtaining roadside sensing data, where the roadside sensing data is obtained by a roadside sensing apparatus by sensing a road environment in a sensing range by using a roadside sensor; and fusing the vehicle sensing data and the roadside sensing data by using a fusion formula, to obtain a first fusion result. According to the foregoing solution, overlapping can be implemented between the sensing range of the roadside sensing apparatus and the sensing range of the vehicle sensing apparatus can be implemented, so that the sensing range is effectively extended.