Sensor Data Fusion Pipeline for Low-Compute Multi-Sensor Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor data fusion systems struggle to accurately fuse heterogeneous, partially heterogeneous, or homogeneous data sources due to inefficiencies in data processing, storage, and computational requirements, leading to reduced accuracy and increased power consumption.
Innovation Solution
A system and method for sensor data fusion that utilizes a computer processor with a curation engine, link engine, fusion engine, inference engine, and validation engine to curate, link, fuse, and validate sensor data from multiple sources, creating a unique dataset with enhanced accuracy and reduced computational and storage demands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor data from multiple sources is fused to improve accuracy, then measurement precision is improved, but computational requirements and processing time increase
Solution Approach 1:
The patent segments the sensor data fusion process into distinct modular components: data reception module, data processing module, and result generation module. Each module handles specific tasks independently, allowing for optimized computation at each stage rather than processing all data uniformly, thus reducing overall computational complexity while maintaining fusion accuracy.
Solution Approach 2:
The patent applies preliminary filtering and preprocessing to sensor data before full fusion processing. By pre-processing data to remove redundant information and validate formats early in the pipeline, the system reduces the computational burden on subsequent fusion algorithms while preserving the accuracy-critical data elements.
2Reliability
If all sensor data is processed and stored to maintain data completeness, then information reliability is improved, but storage requirements and power consumption increase
Solution Approach 1:
The patent extracts and retains only the essential and relevant features from raw sensor data through processing. By identifying and extracting key data elements that contribute to fusion accuracy while discarding redundant information, the system maintains data reliability for critical parameters while significantly reducing storage requirements for the complete dataset.
Solution Approach 2:
The patent applies different processing and storage strategies to different portions of sensor data based on their importance. Critical sensor data that directly impacts fusion accuracy is retained in high fidelity, while less critical data is processed more aggressively or stored in compressed forms, optimizing the balance between completeness and storage requirements.
3Speed
If real-time data fusion is performed to improve responsiveness, then speed is improved, but power consumption increases
Solution Approach 1:
The patent implements periodic or event-triggered data fusion rather than continuous fusion of all sensor data. By fusing data at specific intervals or only when significant changes occur in sensor readings, the system maintains real-time responsiveness for critical updates while reducing power consumption during steady-state operation where frequent fusion provides minimal additional value.
4Adaptability or versatility
If heterogeneous sensor data is fused to improve versatility, then adaptability is improved, but difficulty of detecting and measuring increases
Solution Approach 1:
The patent introduces standardized data interfaces and common reference frames as intermediaries between heterogeneous sensor sources. By translating diverse sensor data formats into a unified intermediate representation, the system enables fusion of multiple sensor types without requiring complex direct correlation algorithms for each sensor pair, thus improving versatility while reducing measurement difficulty.
Data Source
AI summary
Systems and methods of sensor data fusion including sensor data capture, curation, linking, fusion, inference, and validation. The systems and methods described herein reduce computational demand and processing time by curating data and calculating conditional entropy. The system is operable to fuse data from a plurality of sensor types. A computer processor optionally stores fused sensor data that the system validates above a mathematical threshold.