Assured Onboard Autonomy Architecture for Autonomous Underwater Vehicles

Balasuriya, A.

AUVSI XPONENTIAL 2021 (May 2021)

In this paper, we present an adaptive autonomy architecture which considers the uncertainties introduced by autonomous underwater vehicle (AUV) onboard sensors, as well as the challenges introduced by dynamic underwater environments. By addressing these considerations, our architecture provides AUVs with additional capabilities that enable a wider range of underwater applications. First, our architecture uses onboard observations to adapt sensor processing algorithms to the current environment. Second, our architecture consists of a dynamic mission planning algorithm to identify the best mode of operation to successfully complete mission goals given the current situation (i.e., environment and AUV health and status). Our Dynamic Mission Planner can react to known as well as unknown state changes. Each operating mode consists of multiple low-level AUV behaviors. Each behavior reacts to the state of the AUV and the environment by generating a cost function to complete the mission goals. Third, the onboard Performance Manager uses process models and meta-strategies to monitor the performance of every component in the autonomy architecture.

For More Information

To learn more or request a copy of a paper (if available), contact Arjuna Balasuriya.

(Please include your name, address, organization, and the paper reference. Requests without this information will not be honored.)