Long-duration Environmentally-adaptive

Autonomous Rigorous Naval Systems (LEARNS)

P.F.J. Lermusiaux, P.J. Haley, Jr., T. Lolla, D.N. Subramani Massachusetts Institute of Technology Center for Ocean Engineering Mechanical Engineering Cambridge, Massachusetts	Project Summary Ongoing MIT-MSEAS Research MSEAS LEARNS-supported Publications Additional LEARNS Links Background Information
This research sponsored by the Science of Autonomy Program - Office of Naval Research.

---

Project Summary

In the ocean domain, opportunities for a paradigm shift in the science of autonomy involve fundamental theory, rigorous methods and efficient computations for autonomous systems that collect information, learn, collaborate and make decisions under uncertainty, all in optimal integrated fashion and over long duration, persistently adapting to and utilizing the ocean environment. The corresponding basic research is the emphasis of the present project.

Our long-term goal is to develop and apply new theory, algorithms and computational systems for the sustained coordinated operation of multiple collaborative autonomous vehicles over long time durations in realistic multiscale nonlinear ocean settings such that the integrated naval system optimally collects observations, rigorously propagates information backward and forward in time, and accurately completes persistent learning, environmental adaptation, machine metacognition and decision making under uncertainty.

Background information is available below.

Top of page

---

Ongoing MIT-MSEAS Research

Objectives: Derive, implement and evaluate rigorous and efficient Bayesian smoothing theory and schemes that respect nonlinear dynamics and capture non-Gaussian statistics, for robust persistent inference and learning, integrating information backward and forward in time over long durations in large-dimensional multiscale fluid and ocean dynamics. Derive and develop adaptive sampling theories and methods that predict the types and locations of the observations to be collected that maximize information about the ocean system studied (e.g. about its model state variables, parameters and/or formulations) Merge and refine our reduced-order DO stochastic equations with our path planning methods, to obtain new stochastic schemes for time-, coordination-, energy-, dynamics- and swarm- optimal path planning that efficiently account for ocean forecast uncertainties. Develop efficient onboard routing and high-level adaptation schemes that utilize observations collected by vehicles to autonomously adapt optimal plans (e.g. for paths, sampling strategies, collaboration or decision making process). Apply these schemes to simulated fluid and ocean dynamics, from idealized to realistic settings, and integrate these schemes for real sea exercises of opportunity involving distributed computations across components of the autonomous naval sensing systems.

Presentations and Meetings

• MSEAS presentation for the Science of Autonomy meeting; Washington, DC; 11-13 August 2014   [PDF]

 

MSEAS LEARNS-supported Publications

Top of page

---

Additional LEARNS Links

• Science of Autonomy Program at ONR
• Autonomous Systems Innovation Summit
  • Human/Unmanned Systems Collaboration
  • Perception, Understanding and Intelligent Decision Making
  • Scalable and Robust Distributed Collaboration
  • Intelligent C3 Architectures
  • Final Report
• Our own past A-MISSION Project Web-page which is the predecessor project related to the present LEARNS project

Top of page

---

Background Information

As humans, we often combine what we learn over time, and then make informed decisions and complete desired and new tasks. The learning over time, or backward and forward inference, is critical for long-term autonomy, especially in complex nonlinear settings that are ubiquitous in the ocean domain. Mathematically, all initial and acquired information, from both models and data, should be integrated in the form of posterior marginals of the variables of interest, while respecting nonlinearities. The accurate posterior probabilities then facilitate persistent learning, metacognition, informed decisions and tasks completion under uncertainty. Such rigorous nonlinear time-space integration of information and learning, and its application to sustained coordinated autonomous operations of multiple collaborative vehicles is a major focus of the present research. Challenges in our ocean domain arise due to the: complex nonlinear multiscale, multivariate ocean dynamics; large-dimension of the autonomy problems over long duration and large spatial extent; sparse, gappy and multivariate measurements; autonomous coordination and collaboration among heterogeneous vehicles into efficient swarms; and, integration of multiple disciplines into environmentally-adaptive, autonomous and rigorous naval systems.

Top of page

---

Go to the MSEAS home page