v2.1.0

This release includes the early start detection of the TUG test.

What's Changed

• Procedural breach detection: early start by @matey97 in #12

Full Changelog: v2.0.0...v2.1.0

v2.0.0

This release contains the following features:

• Sensing devices: the paired smartwatch and the smartphone can be the source of the sensing data.
• Computing device: the smartphone is the device in charge of data processing.
• Activity Inference Models: the application supports CNN and MLP models for activity inference for each type of data source (i.e., smartphone, smartwatch).
• The application has two modes:
  • COLLECTION: the data gathered by the sensing device is stored in a json file.
  • TUG: the gathered data is used processed and used in the activity inference model to obtain the current activity the user is performing and to compute the results of the test.

Update README

• README improved with environment setup
• Screenshots added

Initial release

First release of TugTestSmartphone.

The application, paired with a smartphone running TugTestSmartwatch, allows the user the self-assessment of the TUG test.

The application receives sensor data from the smartwatch, processes it, and uses it in a ML model for activity inference during the execution of the test. Once the test has finished, the app computes and stores the results.