Sensor Data Correction Using Detection Rules and ML Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Sensor readings are often unreliable due to environmental factors like temperature fluctuations, humidity, and vibrations, leading to poor decision-making and erroneous outputs in predictive models.
Innovation Solution
A method involving the collection of sensor data and auxiliary data, followed by the generation and application of detection rules to modify and refine sensor data, using machine learning models and heuristics to correct errors and generate missing data points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If sensor readings are collected in uncontrolled environments, then the quantity of data available increases, but the reliability of the sensor data deteriorates due to environmental factors like temperature fluctuations, humidity, and vibrations
Solution Approach 1:
The patent introduces an intermediary processing layer between raw sensor data and decision-making systems. Detection rules and machine learning models act as mediators that filter, correct, and validate sensor readings, compensating for environmental disturbances while preserving the value of data collected in uncontrolled environments.
Solution Approach 2:
The system implements feedback mechanisms where sensor data is continuously monitored, analyzed against detection rules, and corrected based on identified anomalies. The correction process feeds back into the data stream, improving reliability while maintaining the continuous flow of data from uncontrolled environments.
2Reliability
If detection rules are applied to correct sensor readings, then the reliability of sensor data improves, but the device complexity increases due to the need for machine learning models and heuristic rules
Solution Approach 1:
The detection system is segmented into modular components: data collection modules, rule generation modules, validation modules, and correction modules. Each component performs a specific function, allowing the system to achieve high reliability through specialized processing while managing complexity through clear separation of concerns.
Solution Approach 2:
Detection rules and machine learning models are trained and prepared in advance before being deployed for real-time sensor data correction. This preliminary action allows the system to handle complex analysis tasks offline, reducing the computational complexity required during actual sensor data processing while maintaining high reliability.
3Loss of information
If machine learning models are used to generate missing sensor readings, then the completeness of data improves, but the loss of information increases due to potential inaccuracies in predicted values
Solution Approach 1:
The system applies machine learning models selectively to generate only those missing sensor readings where the confidence level is sufficient. Rather than attempting to predict all missing values, the system uses partial action by focusing on cases where prediction accuracy is most likely, thereby improving data completeness while minimizing the risk of introducing inaccurate information.
Data Source
AI summary
Disclosed are systems, methods, and devices for correcting or otherwise cleaning sensor data. Sensor readings and metadata or other information about the sensor readings can be collected, and one or more detection rules (e.g., machine learning models or other detection rules) can be automatically generated for modifying subsequent sensor data. Sensor readings can be refined or supplemented by applying applicable detection rules.