Multi-Sensor Structure Fusion via Relational Vector Scoring

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Fusing data from multiple sensors to determine the layout of structure systems, such as buildings, is challenging due to the difficulty in aligning and combining sensor data sets from different sources.

Innovation Solution

A system that generates structure pairs from sensor data sets, calculates relational vector scores and association scores to determine the likelihood that structures represent the same physical structure, and fuses these structures using a slope test and relational vector analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If multiple sensors are used to determine the layout of structure systems, then the completeness and accuracy of structure detection is improved, but the difficulty of fusing sensor data increases

Engineering Contradiction:

Improvestructure detection accuracyVSAvoidsensor data fusion difficulty

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary processing system that receives sensor data from multiple sensors, generates structure pairs, calculates association scores using relational vectors and slope tests, and determines which structures represent the same physical structure. This intermediary layer mediates between the raw sensor data and the final fused layout, resolving the complexity of direct multi-sensor fusion while maintaining high detection accuracy

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent transforms sensor data into a standardized parameter space by calculating relational vectors (comprising distance, angle, and orientation parameters) and slope test scores. By converting diverse sensor measurements into comparable parametric forms, the system enables effective data fusion across different sensor types while maintaining measurement precision

Inventive Principle:

Principle #35Parameter changes

2Measurement precision

If structure pairs are evaluated using multiple criteria (slope test and relational vector analysis), then the accuracy of determining same physical structure is improved, but the computational complexity increases

Engineering Contradiction:

Improvestructure identification accuracyVSAvoidcomputational complexity

Core Design Contradiction:

Measurement precisionVSPower

Solution Approach 1:

The patent applies partial action by implementing a two-stage evaluation process: first a slope test for quick filtering, then relational vector analysis for detailed comparison. Not all structure pairs undergo both tests - the slope test serves as a preliminary filter to reduce the number of pairs requiring full relational vector analysis, thereby maintaining high identification accuracy while reducing overall computational complexity

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS8271224B2Fusing structures from multi-sensor data

Publication Date: 2012.09.18 RAYTHEON CO

AI summary

In particular embodiments, fusing structures includes receiving sensor data sets generated by sensors in response to sensing a structure system. Each sensor data set describes structures of the structure system. Structure pairs are generated, where a structure pair comprises a first structure from a first sensor data set and a second structure from a second sensor data set. A relational vector set is defined for each structure pair, and a relational vector score is calculated for each structure pair according to the relational vector set of the structure pair. An association score is calculated for each structure pair according to the relational vector score of the structure pair. The association score of the structure pair indicates a likelihood that the structure pair is fusable.