Adaptive Sensor Calibration via Cognitive Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sensor calibration methods are inadequate for ensuring accuracy and reliability, particularly in mobile devices like autonomous vehicles, where inaccurate readings can lead to safety issues due to the high cost and complexity of precise sensors.
Innovation Solution
Adaptive calibration of sensors through cognitive learning, utilizing machine learning to compare data from calibration sensors with itinerant sensors, generating calibration parameters based on preexisting sensor information, and executing these parameters to adjust sensor readings and reliability factors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional sensor calibration methods are used, then sensor accuracy can be maintained through periodic recalibration, but device complexity and cost increase due to multiple calibration sensors and frequent maintenance
Solution Approach 1:
The itinerant sensor autonomously performs calibration by visiting calibration sensors at predetermined locations, eliminating the need for manual calibration operations. The sensor independently determines its accuracy based on comparisons with calibration sensors and adjusts its readings accordingly, making the calibration process self-service rather than requiring external intervention or complex calibration systems.
Solution Approach 2:
The calibration system transitions from static periodic recalibration to dynamic adaptive calibration. The itinerant sensor continuously moves through the environment, opportunistically calibrating against calibration sensors when encountered. The system dynamically adjusts calibration parameters based on real-time comparisons and machine learning, rather than following fixed calibration schedules.
2Measurement precision
If precise calibration sensors are deployed, then sensor accuracy improves, but cost increases due to the high price of precise sensors
Solution Approach 1:
Instead of deploying expensive precise sensors throughout the system, the invention uses a single itinerant sensor that copies or replicates the calibration function by visiting calibration sensors at multiple locations. The itinerant sensor acts as a mobile copy of the calibration process, spreading the calibration capability across the environment rather than requiring multiple fixed calibration sensors.
Solution Approach 2:
The itinerant sensor serves multiple functions: it performs calibration for itself, calibrates other sensors it encounters, collects environmental data, and updates machine learning models. This multi-functionality reduces the need for dedicated calibration sensors at every location, as one itinerant sensor can serve the entire network of sensors across the environment.
3Reliability
If frequent recalibration is performed, then sensor reliability improves, but loss of time increases due to calibration interruptions
Solution Approach 1:
The calibration process operates periodically when the itinerant sensor encounters calibration sensors at predetermined locations during its traversal. Rather than continuous calibration that would interrupt operations, the system uses periodic calibration opportunities that occur naturally during the sensor's movement through the environment, minimizing disruption while maintaining reliability.
Solution Approach 2:
The itinerant sensor continues its primary data collection and monitoring functions uninterrupted while performing calibration. The calibration process is integrated into the sensor's normal operational workflow, allowing it to maintain useful actions continuously rather than pausing for dedicated calibration periods. The sensor calibrates opportunistically without stopping its mission-critical functions.
Data Source
AI summary
Embodiments of the present invention may be directed toward a method, a system, and a computer program product of adaptive calibration of sensors through cognitive learning. In an exemplary embodiment, the method, the system, and the computer program product include (1) in response to receiving a data from at least one calibration sensor and data from an itinerant sensor, comparing the data from the at least one calibration sensor and the data from the itinerant sensor, (2) in response to the comparing, determining, by one or more processors, the accuracy of the itinerant sensor, (3) generating, by the one or more processors, one or more calibration parameters based on the determining and based on a machine learning associated with preexisting sensor information, and (4) executing, by the one or more processors, the one or more calibration parameters.