A unified dynamic information guided particle frameworkfor mission design and execution

Abstract

The goal of this research is to create a unified Markov chain Monte Carlo (MCMC)platform that enables efficient processing of complex data and communication ofinformation among its constituent algorithms for designing and conducting missionoperations such as modeling, sensing, forecasting, assimilation, allocation andcontrol. This will result in an inter-compatible algorithm architecture consistentwith the DDDAS philosophy of unification of the computational and sensing modulesin the mission. New methodologies, all based on the MCMC particle paradigm forperforming each of the above mentioned computational mission tasks will bedeveloped. For forecasting, a new MCMC ensemble approach combined withKarhunen-Loève expansions of the process noise and the method of characteristicsfor solving PDEs will be developed. Data assimilation will be performed via MCMCBayesian inference. Trajectory planning and system-level control will be posed aschance-constrained optimal control problems, which will be solved using directcollocation-based nonlinear programming supported by a semi-analytical/split-Bernstein approximation of the probabilistic constraints involving state and controlvariables.

Document Details

Document Type
DoD Grant Award
Publication Date
Sep 11, 2017
Source ID
FA95501710307

Entities

People

  • Mrinal Kumar

Organizations

  • Air Force Office of Scientific Research
  • Ohio State University
  • United States Air Force

Tags

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Distributed Systems and Data Platform Development

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • AI & ML - Bayesian Inference
  • AI & ML - Machine Learning Algorithms