Deployment and Development of a Cognitive Teleoreactive Framework for Deep Sea Autonomy
Christopher Thierauf
TL;DR
The paper tackles the rigidity and non-interpretable nature of the legacy MC system for AUV Sentry, which hampers adaptive sampling and real-time intervention. It introduces DINOS-R, a teleoreactive architecture that fuses symbolic deliberation with reactive execution, implemented in Python and designed as a monolithic, plugin-based system with four central databases to support planning, belief tracking, and self-assessment. The authors validate DINOS-R through digital-twin simulations and an at-sea deployment on Sentry, demonstrating autonomous plan generation, real-time re-tasking via operator input, and safe recovery during a mission. They also outline future work aimed at non-expert user interfaces, open-source release, broader platform applicability, and formal safety guarantees using LTL and model checking. The work holds promise for more robust, interpretable, and accessible autonomous oceanographic missions, enabling scientists to direct complex experiments without requiring expert operators.
Abstract
A new AUV mission planning and execution software has been tested on AUV Sentry. Dubbed DINOS-R, it draws inspiration from cognitive architectures and AUV control systems to replace the legacy MC architecture. Unlike these existing architectures, however, DINOS-R is built from the ground-up to unify symbolic decision making (for understandable, repeatable, provable behavior) with machine learning techniques and reactive behaviors, for field-readiness across oceanographic platforms. Implemented primarily in Python3, DINOS-R is extensible, modular, and reusable, with an emphasis on non-expert use as well as growth for future research in oceanography and robot algorithms. Mission specification is flexible, and can be specified declaratively. Behavior specification is similarly flexible, supporting simultaneous use of real-time task planning and hard-coded user specified plans. These features were demonstrated in the field on Sentry, in addition to a variety of simulated cases. These results are discussed, and future work is outlined.