Asynchronous Vision-Acceleration Fusion for Bridge Displacement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing displacement measurement technologies face challenges such as the need for rigid fixed support points, limited accuracy in GNSS-denied environments, high computational costs, and noise contamination, especially when measuring large structures like bridges.
Innovation Solution
A method and system that fuse asynchronous vision measurement data and acceleration data using an adaptive multi-rate Kalman filter, automatically calculating a scale factor and applying an improved feature matching algorithm to estimate displacement with high accuracy and low maintenance cost, without requiring a fixed support point or prior knowledge of the target's size.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If vision cameras are used for displacement estimation, then displacement measurement is achieved, but the measurement results are contaminated with noises due to relatively low sampling frequency
Solution Approach 1:
The patent combines vision measurement data with acceleration measurement data to create a fused displacement estimation. The vision camera provides displacement information while the accelerometer provides high-frequency acceleration data, and their fusion through Kalman filtering produces accurate high-rate displacement estimates that overcome the low sampling frequency limitation of the vision camera alone.
Solution Approach 2:
The patent introduces acceleration measurements as an intermediary to bridge the gap between low-frequency vision data and high-frequency displacement estimation. The acceleration data serves as a mediator that, when integrated and filtered with vision data, enables high sampling rate displacement measurement without requiring the vision camera to operate at high speeds.
2Measurement precision
If the scale factor is estimated by identifying the size of the target in physical units, then displacement conversion is achieved, but it requires manual measurement and increases maintenance cost
Solution Approach 1:
The system automatically calculates the scale factor by fusing vision measurement data with acceleration measurement data, eliminating the need for manual measurement of target size. The scale factor is derived self-service through the data fusion process using the relationship between pixel displacement from vision and physical displacement from acceleration integration, reducing both installation complexity and maintenance requirements.
Solution Approach 2:
The patent changes the method of determining the scale factor from direct physical measurement to indirect calculation through data fusion. By using the relationship between vision-based pixel displacement and acceleration-based physical displacement, the scale factor is computed as a parameter through mathematical processing rather than manual measurement, simplifying installation and reducing maintenance.
3Productivity
If asynchronous vision measurement data and acceleration data are fused, then high sampling rate displacement measurement is achieved, but complex data synchronization is required
Solution Approach 1:
The patent employs an adaptive multi-rate Kalman filter that dynamically adjusts to the asynchronous nature of vision and acceleration data streams. The filter adapts its processing based on the different sampling rates and timing of the two data sources, enabling high sampling rate displacement estimation while managing the complexity of asynchronous data fusion through dynamic adaptation rather than rigid synchronization.
Data Source
AI summary
Disclosed are a method and a system for measuring a displacement of a structure based on fusion of asynchronous vision measurement data of natural target and acceleration data of the structure. A scale factor, which is a conversion factor between a pixel resolution of the image frame data and a distance resolution of real space, is calculated using image frame data taken at a predetermined first sample frequency of a stationary target outside the structure by a camera installed on the structure, and acceleration data measured at a predetermined second sample frequency by an accelerometer installed on the structure, over time. A vision-based displacement value of the structure is estimated using a feature point matching algorithm and the scale factor.