In this paper, we elaborate on the issue of reliable storage and efficient communication of large quantities of data in the absence of continuous connectivity. We illustrate how advanced machine learning techniques can run locally at the edge, in the context of data compression related to special-purpose vehicles. Two different data compression techniques are compared by calculating general compression metrics, e.g., compression rate and root mean-squared error, while also validating the results using an event detection algorithm. These techniques exploit real-world usage data captured in the field using the I-HUMS platform provided by our industrial partner ILIAS solutions Inc.