Non-linear decision making for robust navigation in role based autonomy

Robert Thome, Christopher Fedor, John Sustersic, Ucheoma Ukah

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

A common approach to the robotic path planning problem is to discretize the search space and plan an optimal path using graph search methods. As the optimality criteria becomes more complex it becomes increasingly difficult to optimize the search with respect to each metric being considered. A hybrid path planning system is introduced that supplements the graph search approach with a cognitive architecture. The selected cognitive architecture, Soar, will provide high level reasoning, guiding the implementation of the graph search and reasoning on the results generated. Given scenario-specific context, the hybrid approach will allow for multiple, complex and possibly conflicting criteria to be considered during path generation.

Original languageEnglish (US)
Title of host publicationOCEANS 2015 - MTS/IEEE Washington
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9780933957435
DOIs
StatePublished - Feb 8 2016
EventMTS/IEEE Washington, OCEANS 2015 - Washington, United States
Duration: Oct 19 2015Oct 22 2015

Publication series

NameOCEANS 2015 - MTS/IEEE Washington

Other

OtherMTS/IEEE Washington, OCEANS 2015
CountryUnited States
CityWashington
Period10/19/1510/22/15

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Oceanography
  • Ocean Engineering
  • Instrumentation
  • Acoustics and Ultrasonics

Fingerprint Dive into the research topics of 'Non-linear decision making for robust navigation in role based autonomy'. Together they form a unique fingerprint.

Cite this