Leveraging arbitrary mobile sensor trajectories with shallow recurrent decoder networks for full-state reconstruction
Published in arXiv, 2023
Sensing is a fundamental tasks for the monitoring, forecasting, and control of complex systems. In many applications, a limited number of sensors are available and must move with the dynamics. Most model-free sensing paradigms aim to map current sparse sensor measurements to the high-dimensional state space, ignoring the time-history. Our aim was to quantify feasibility and reliability of using mobile sensor trajectories with shallow recurrent decoder networks to accurately reconstruct high-dimensional state space. We found the time-history of mobile sensors can successfully be used to encode global information of the measured high-dimensional state space.
Submitted to IEEE ACCESS