tinyML device monitors packages for damage in transit


Although the widespread use of online shopping has been a great convenience, it has also led to a sharp increase in the number of returned items. This can be attributed to a number of factors, but shipping damage is a big contributor to this problem. Shebin’s solution José Jacob consists of building a small tracer that accompanies the package throughout its journey and sends alerts when mishandling is detected.

Jacob started with create a new Edge Impulse project and collecting approximately 30 minutes of motion samples from a Sense Arduino Nano 33 BLEof the onboard three-axis accelerometer. Each sample was classified into one of five categories ranging from no movement to a hard drop or a vigorous jolt. The features were then generated and used to train a Keras model, which yielded 91.3% accuracy in testing.

To communicate with the outside world, Jacob added a GSM module that allows the Nano 33 BLE Sense to send alerts over a 3G network to a standby Firebase endpoint. When the database is updated, the new data is propagated to a user web page that displays the current status of the package as well as any important events.

More details can be found here in writing Jacob’s project.


About Author

Comments are closed.