Real-Time Instrument Planning and Perception for Novel Measurements of Dynamic Phenomena
This work addresses the challenge of capturing rare, transient events in remote sensing for scientific applications, though it is incremental as it builds on existing computer vision and planning methods.
The paper tackles the problem of automating the detection and measurement of dynamic phenomena like volcanic plumes by integrating event detection in satellite imagery with autonomous trajectory planning for high-resolution sensors, resulting in an order of magnitude increase in instrument utility compared to baselines while maintaining efficient runtimes.
Advancements in onboard computing mean remote sensing agents can employ state-of-the-art computer vision and machine learning at the edge. These capabilities can be leveraged to unlock new rare, transient, and pinpoint measurements of dynamic science phenomena. In this paper, we present an automated workflow that synthesizes the detection of these dynamic events in look-ahead satellite imagery with autonomous trajectory planning for a follow-up high-resolution sensor to obtain pinpoint measurements. We apply this workflow to the use case of observing volcanic plumes. We analyze classification approaches including traditional machine learning algorithms and convolutional neural networks. We present several trajectory planning algorithms that track the morphological features of a plume and integrate these algorithms with the classifiers. We show through simulation an order of magnitude increase in the utility return of the high-resolution instrument compared to baselines while maintaining efficient runtimes.