Embedded artificial intelligence augmented sensors
US12366589B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Sep 15, 2022 |
| Grant date | Jul 22, 2025 |
| Priority date | — |
| Expiry date | Oct 27, 2043 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06N20/00
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
A method, sensor, and non-transitory computer-readable storage medium are provided for estimating actual amplitudes of a waveform. A machine learning model may be trained for an embedded system of a first three-axes sensor having a limited range to estimate the actual amplitudes of a waveform that saturates the first three-axes sensor in a direction of one of the three axes. The embedded system acquires a second waveform during use of a tool including the first three-axes sensor. The second waveform that occurs after a second waveform producing event is isolated. The embedded system extracts a multi-dimensional feature from the isolated second waveform and estimates, using the machine learning model, the actual amplitudes of the second waveform based on the extracted multi-dimensional feature.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.