Machine Learning
Training instead of programming
Many embedded systems use firmware for the relationship between inputs and outputs. In sensor applications, for example, the firmware processes raw sensor data and provides digital output signals.
Firmware developments are time-consuming. SSV takes a new approach here: a machine learning (ML) algorithm is connected and trained between the input and output. This ML model can be changed at any time by retraining.
At Embedded World, SSV is presenting the DNP/AISS1, a starter kit with sensors and pre-installed ML algorithms. This allows valuable information to be extracted from sensor data. A Docker container contains all the necessary tools. The company will also be demonstrating machine learning examples with raw sensor data for predictive maintenance applications in a webinar.
Embedded World, Hall 3, Stand 439









