Coordinated and Collaborative Sampling by Two Long-Range Autonomous Underwater Vehicles
Yanwu Zhang, Brian Kieft, Brett Hobson, Quinn Shemet, Christina M. Preston, Christopher Wahl, Kathleen Pitz, Kelly J. Benoit‐Bird, James M. Birch, Francisco P. Chávez, Christopher A. Scholin
- Year
- 2024
- Citations
- 5
Abstract
Multiple autonomous underwater vehicles (AUVs) working in collaboration can achieve scientific goals more effectively than independently operated vehicles. In this article, we present a case of using two long-range AUVs (LRAUVs) for persistent environmental DNA (eDNA) sampling of a targeted feature. Each LRAUV was equipped with a third-generation environmental sample processor (3G-ESP), a robotic instrument for acquiring and processing water samples for molecular analysis. Each 3G-ESP can collect and process 60 samples. For continuous and persistent eDNA sampling of vertically migrating organisms at a targeted depth layer, we deployed two LRAUVs which alternately triggered the ESP, extending the total time of collecting samples. We developed a method of coordinated sampling by time shift and a collaborative sampling method that uses acoustic handshakes. In the time-shift method, each vehicle switched between two behaviors: sample collection at the targeted depth and spiraling over a large depth range to make contextual measurement. The second vehicle's mission started later than the first vehicle's by a time shift equal to the duration of one sampling event, such that at a given time one vehicle sampled at the targeted depth while the other vehicle spiraled up and down. In the acoustic-handshake method, the two LRAUVs exchanged sample-start and sample-end messages. On receiving vehicle #1’s sample-end message, vehicle #2 triggered a sampling event and transmitted a sample-start message to vehicle #1. Then, vehicle #1 waited for vehicle #2’s sample-end message before triggering the next sampling event, and so forth. The time-shift method is simple, whereas the acoustic-handshake method is accurate and adaptive. Both methods were demonstrated in experiments in Monterey Bay.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Fractional Differential Equations
Igor Podlubný
2025
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991