Remote Sensor Calibration via Peer Population Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Air quality sensors (AQSs) in vehicles face challenges with sensitivity variations over time, leading to inconsistent pollution detection and map accuracy, requiring effective calibration methods to maintain reliability and accuracy.
Innovation Solution
A remote sensor calibration system that compares sensor output to a peer sensor population, using connected car features or wireless devices for continuous calibration, identifies outliers, and adjusts performance metrics to ensure consistency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional sensor calibration methods are used, then sensor sensitivity can be maintained initially, but sensor performance degrades over time due to drift and variability
Solution Approach 1:
The system implements continuous feedback by comparing each sensor's readings against the crowd-sourced population data in real-time. The server receives sensor data from multiple sources, identifies outliers through statistical analysis, and sends calibration commands back to individual sensors to maintain their performance within acceptable ranges.
Solution Approach 2:
Sensors automatically calibrate themselves by comparing their readings against the crowd-sourced reference data without requiring manual intervention. The system performs self-diagnosis and self-correction by identifying when a sensor deviates from the population norm and automatically adjusting its offset or gain parameters.
2Measurement precision
If manual calibration procedures are implemented, then sensor accuracy can be improved, but system complexity and human intervention requirements increase
Solution Approach 1:
The calibration system operates autonomously without requiring manual intervention. Sensors automatically transmit their data to the server, which performs statistical analysis and sends back calibration commands when needed, eliminating the need for human operators to physically access and calibrate each sensor.
Solution Approach 2:
A cloud-based server acts as an intermediary between sensors and the calibration process. The server collects data from multiple sensors, performs centralized statistical analysis to identify outliers, and distributes calibration commands back to individual sensors, simplifying the overall system architecture.
3Reliability
If frequent calibration is performed, then sensor performance is maintained, but energy consumption and operational time increase
Solution Approach 1:
The system uses continuous feedback from crowd-sourced data to determine when calibration is actually needed. By monitoring the distribution of sensor readings across the population, the system only triggers calibration for sensors that deviate from the norm, rather than performing scheduled calibration on all sensors.
Solution Approach 2:
Instead of calibrating all sensors frequently, the system applies calibration only to the subset of sensors that are identified as outliers through statistical analysis of the crowd-sourced data, reducing unnecessary calibration operations.
Data Source
AI summary
A remote sensor calibration system remotely calibrates sensors of various types in an automated fashion. The remote sensor calibration system receives sensor output, associated with one or more locations and one or more times, from a sensor and its peer sensor population. The sensor output from the sensor and its peer sensor population is compared to generate a correction factor. The correction factor is transmitted to the sensor and used to calibrate the sensor. Communication will typically occur via one or more wireless links allowing the sensors to be remotely calibrate while maintaining their mobility.