Fixed Interpolation Error Data Simplification for Telematics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Telematics systems face challenges in accurately simplifying and transmitting data from assets, leading to inconsistent error profiles in value estimation calculations, which affects the reliability of asset status information, especially when data is collected over time and changes rapidly.
Innovation Solution
A fixed estimation error data simplification process that limits interpolation and extrapolation errors by using a 'vertical' deviation metric, excluding 'horizontal' deviations caused by data collection time, ensuring a consistent error bound across simplified data sets, and triggering data logging based on vertical deviations from trends.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If data simplification algorithms are applied to reduce data transmission and storage requirements, then data quantity is reduced, but interpolation error consistency deteriorates
Solution Approach 1:
The patent changes the parameter used for data simplification from time-based sampling to vertical deviation-based selection. By using a fixed threshold for vertical deviation from the trend line, the system ensures that interpolation error remains bounded and consistent regardless of how much data is simplified. This parameter change resolves the contradiction by making error consistency independent of data quantity reduction.
Solution Approach 2:
The patent applies preliminary filtering of data points based on their vertical deviation from the trend line before transmission. By pre-identifying and removing only those points that exceed the fixed vertical threshold, the system reduces data quantity while guaranteeing that remaining points maintain interpolation error within acceptable bounds, thus resolving the contradiction between data reduction and error consistency.
2Measurement precision
If data collection frequency is increased to capture rapidly changing asset status, then measurement precision improves, but data transmission load increases
Solution Approach 1:
The patent applies local quality by differentiating between data points based on their vertical deviation significance. Instead of uniform sampling, the system selectively transmits only those local data points that exceed the vertical threshold, capturing rapid changes where they occur while ignoring periods of stability. This resolves the contradiction by concentrating transmission resources on locally significant events rather than uniform high-frequency sampling.
3Ease of operation
If traditional time-based sampling is used for data collection, then data logging is simple, but error profile consistency deteriorates due to horizontal deviations
Solution Approach 1:
The patent inverts the traditional approach by not sampling based on time intervals, but rather selecting data points based on their vertical deviation from the trend line. This inversion transforms the selection criterion from horizontal (time-based) to vertical (value-based), ensuring that only points contributing to interpolation error beyond the fixed threshold are considered for transmission, thus achieving error profile consistency.
Data Source
Figure 1
Figure 2A~2C
Figure 3A~3B
AI summary
Methods and systems for simplifying data collected from assets are provided. An example method involves receiving a set of simplified data at a server. The simplified set of data is generated by application of a dataset simplification algorithm on raw data obtained from a data source at an asset upon satisfaction of a data logging trigger, wherein the dataset simplification algorithm causes interpolation error within the simplified set of data to be limited by an upper bound that is fixed across the simplified set of data. The method further involves receiving a request for a status of the asset and interpolating a status of the asset based on the simplified set of data in response to the request.