Multi-Agent Underwater Cooperative Navigation Using Low Dynamics Model

Abstract

Multi-Agent Underwater Cooperative Navigation Using Low Dynamics ModelThe objective of this work is to provide improved Autonomous Undersea Vehicle (AUV) navigation accuracy through a cooperative multi-agent method. The scope of this project is model development", a computer simulation, and trade study. The objective is to assess the performance of using our proposed multi-agent method. Our m""ethod~s navigational error will becompared to a standard, single vehicle inertial dead-reckoning method. The system level problem b"eing solved is to provide better underwater navigational aids for AUVs leveraging estimates from one or more vehicles in a swarm. Th"is is important becauseaccurate navigation is a critical capability for autonomous systems performing missions such assurveying, m""ine detection, or mine countermeasures (MCM). With an improved navigation solution, the Navy benefit is higher mission success throu""gh faster execution, improved resource utilization, and lower cost. The more accurate navigation solution may allow the AUVs to acco"mplish the mission faster because the AUVs may not need to perform as many extremecoarse corrections or refine an area of uncertainty (AOU). A co-localization scheme may improve resource utilization through fusion of sensor data from cooperative AUVs. The mission may now be accomplished at lower cost because less sophisticated navigation sensor packages and therefore cheaper AUV assets can be used.

Document Details

Document Type
DoD Grant Award
Publication Date
Jun 09, 2017
Source ID
N000141712618

Entities

People

  • Scott Koziol

Organizations

  • Baylor University
  • Office of Naval Research
  • United States Navy

Tags

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Inertial Navigation Systems.
  • Robotics and Automation.

Technology Areas

  • Autonomy
  • Autonomy - Autonomous System Control