Datalogging guidelines for a successful NanoEdge AI project

This message will disappear after all relevant tasks have been resolved.

Semantic MediaWiki

There are 1 incomplete or pending task to finish installation of Semantic MediaWiki. An administrator or user with sufficient rights can complete it. This should be done before adding new data to avoid inconsistencies.

This documents presents several use case studies where NanoEdge AI Studio has been used successfully to develop Anomaly Detection or Classification projects.

It aims at explaining the methodology and thought process behind the choice of crucial parameters during the initial datalogging process (that is, even before starting to use NanoEdge AI Studio) that can make or break a project.

For each use case, it will focus on the following aspects:

what is a meaningful representation of the physical phenomenon being observed
how to select the optimal sampling frequency for the datalogger
how to select the optimal buffer size for the data sampled
how to format the data logged properly for the Studio

1. Summary of important concepts

1.1. Definitions

1.2. Sampling frequency

1.3. Buffer size

1.4. Data format

2. Use case studies

2.1. Vibration patterns on a ukulele

2.1.1. Context and objective

Datalogging guidelines for a successful NanoEdge AI project

Contents

1. Summary of important concepts

1.1. Definitions

1.2. Sampling frequency

1.3. Buffer size

1.4. Data format

2. Use case studies

2.1. Vibration patterns on a ukulele

2.1.1. Context and objective

2.1.2. Implementation process

2.1.3. Results

2.2. Vibration patterns on an electric motor

2.2.1. Context and objective

2.2.2. Implementation process

2.2.3. Results

2.3. Current sensing on a 3-phase motor

2.3.1. Context and objective

2.3.2. Implementation process

2.3.3. Results

2.4. Gesture recognition using a Time-of-Flight sensor

2.4.1. Context and objective

2.4.2. Implementation process

2.4.3. Results

3. Resources