Sensor Fusion for Multi-Agent Tracking in GPS-Denied Environments

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing localization and tracking technologies face challenges in environments without GPS signals, particularly in providing accurate and robust positioning of devices using either radio-based or visual odometry methods alone, which are prone to errors and limited by environmental conditions.

Innovation Solution

A hybrid approach that combines radio-based and visual odometry data using algorithmic and data-driven methods for sensor fusion, leveraging the strengths of both passive and active tracking modalities to achieve robust and accurate multi-agent tracking, incorporating neural network models for feature extraction and fusion to improve accuracy and reliability across diverse environments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If radio-based tracking is used alone, then the system can operate in GPS-denied environments, but the tracking accuracy deteriorates to 40 cm

Engineering Contradiction:

Improveoperational capability in GPS-denied environmentsVSAvoidtracking accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent combines radio-based tracking and visual odometry tracking into a unified system that fuses both modalities to achieve both operational capability in GPS-denied environments and high tracking accuracy within 15 cm

Inventive Principle:

Principle #5Merging (Combining)

2Duration of action of moving object

If visual odometry tracking is used alone, then the system provides continuous tracking, but the accuracy deteriorates to 32 cm and is limited by environmental conditions

Engineering Contradiction:

Improvecontinuous tracking capabilityVSAvoidtracking accuracy

Core Design Contradiction:

Duration of action of moving objectVSMeasurement precision

Solution Approach 1:

The patent merges visual odometry with radio-based tracking to maintain continuous tracking capability while improving accuracy to within 15 cm by compensating for visual odometry's environmental limitations through radio signal supplementation

Inventive Principle:

Principle #5Merging (Combining)

3Measurement precision

If sensor fusion is implemented, then tracking accuracy improves to within 15 cm, but the device complexity increases

Engineering Contradiction:

Improvelocalization accuracyVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary fusion module that integrates radio-based and visual odometry data, serving as a mediator between the two tracking modalities and the final position estimate, thereby achieving high accuracy while managing system complexity through structured integration

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12000945B2Visual and RF sensor fusion for multi-agent tracking

Publication Date: 2024.06.04 NEC CORP

AI summary

Methods and systems for determining a device position include determining a first position estimate using radio-based range information. A second position estimate is determined using visual odometry information. The first position estimate and the second position estimate are fused based on radio environmental conditions and visual environmental conditions to determine a final position estimate. Resources are deployed based on the final position estimate.