Data-Enhanced Sensing: A New Approach to Information Acquisition

Increasing digital transformation of manufacturing has highlighted physics-based sensing and data-driven analytics as the drivers and cornerstones of Industry 4.0. At the same time, new manufacturing processes such as additive manufacturing and robot-assisted assembly highlight the need for new sensing capabilities to overcome constraints in signal transduction principles that underly sensor designs, sensor operation resources, and real-world deployment. Advancement in information theory and artificial intelligence (AI) over the past decade has provided the technological foundation for new ways to tackle constraints in time, bandwidth, and energy consumption for sensing, leading to data-enhanced, optimal strategy for data sampling and information recovery. Furthermore, it enables inference of information that is hidden from direct measurements. This paper highlights the latest development in this field and introduces data-enhanced sensing as a new paradigm that propels innovations in sensing, instrumentation, and measurement.