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.

Volcán de Fuego stands 12,346 feet overlooking Guatemala's former capital, Antigua, and because it erupts multiple times every hour, researcher have yet to closely study what is happening deep beneath it – until now. A team of UK volcanologists are employing a fleet of drones fitted with special sensors to make the treacherous journey to the peak and'take measurements that would otherwise never have been possible'. The data will help researchers investigate how the massive mountain changes overtime, allowing them to better protect the 60,000 people that are at risk of it erupting. Volcán de Fuego stands 12,346 feet tall overlooking Guatemala's former capital, Antigua, and because it erupts multiple times every hour, researcher have yet to closely study what is happening deep beneath the volcano – until now Volcán de Fuego is one of Central American's most active volcanoes and is one of three large stratovolcanoes overlooking Guatemala's former capital, Antigua. It was formed when the Meseta volcano collapsed about 8,500 years ago, which caused an avalanched that traveled 31 miles (50 km) onto the Pacific coastal plain.

In the thick, dry grass of the Paso de Cortés mountain pass in Mexico, MIT's Earth Signals and Systems (ESS) group struggled to launch a small aircraft high up into the volcanic plume rising steadily from Popocatépetl. It was a windy day, and nothing was going right. Then the volcano began to erupt. "It sounded like a big engine going off, a loud boom," says Sai Ravela, principal research scientist in the MIT Department of Earth, Atmospheres, and Planetary Science. Their escort from CENAPRED, Mexico's federal disaster prevention agency, told them to drop everything and leave.