Source-Sink Relation Framework for V2X Sensor Data Reliability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle-to-everything (V2X) communication systems face challenges in reliably aggregating and fusing sensor data from multiple sources, as the reliability of information is often dependent solely on the source device, neglecting the sink device's knowledge about the source, which can lead to inconsistent and unreliable environmental perception.
Innovation Solution
A framework that incorporates a Source-Sink Relation (SSR) variable into data fusion, considering both the source and sink device's manufacturer and other factors to determine the reliability of sensor data, using a table to assign trust values and applying machine learning to update these values based on past data accuracy, thereby enhancing the reliability of environmental value estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data fusion is performed using only source device reliability, then the data fusion process is simple, but the reliability of environmental perception decreases
Solution Approach 1:
The reliability assessment is segmented into two independent components: source reliability (R_s) and sink reliability (R_k). This segmentation allows each component to be evaluated separately and then combined, improving perception reliability while maintaining manageable system complexity through modular assessment.
Solution Approach 2:
The patent introduces a new dimension to reliability assessment by adding sink device reliability (R_k) to the traditional source-only approach. This transforms the reliability model from a single-dimensional source-based assessment to a two-dimensional source-sink joint assessment, thereby improving overall environmental perception reliability.
2Measurement precision
If source-sink relation terms are stored in tables for all device combinations, then the reliability assessment becomes more accurate, but the storage requirements and system complexity increase
Solution Approach 1:
The source-sink relation terms are segmented by device type categories rather than storing individual terms for every possible device combination. This segmentation reduces the quantity of stored data while maintaining precise reliability assessment by evaluating sources and sinks of the same type based on their categorized relationships.
3Adaptability or versatility
If machine learning is used to update source-sink relation terms, then the reliability assessment adapts to past data accuracy, but the computational complexity increases
Solution Approach 1:
Machine learning algorithms are employed to create a feedback mechanism where source-sink relation terms are automatically updated based on the accuracy of past data. This feedback loop enables the system to adapt and improve reliability assessment over time by learning from historical performance data, balancing adaptability with computational efficiency.
Data Source
AI summary
Various techniques for collective perception messaging are disclosed herein. In an example, a machine receives, from a source device, a signal value for provision to a sink device, the signal value corresponding to a measurement of an environmental value. The machine accesses, from a storage device, an error term for the signal value. The machine accesses, from the storage device, a source reliability term for the source device. The machine accesses, from the storage device, a source-sink relation term based on the source device and the sink device. The machine determines a distribution for the environmental value based on the error term, the source reliability term, and the source-sink relation term. The machine determines, based on the distribution for the environmental value, whether the signal value is reliable.