AUV Fault Detection Using Model Based Observer Residuals

Abstract

In order for the Navy's next generation Unmanned Undersea Vehicles to be more robust to software/hardware faults, on-line failure detection and resolution is needed. Typically, fault detection methods include limits and trends analysis, model free, and model based techniques. Here, model based observers are proposed for the detection of fault induced dynamic signals in the diving, steering, and roll control systems. Such automatic fault detection systems were designed and implemented in a Simulink model of the "21UUV." in the course of conducting simulations with the model, numerous vehicle behaviors were studied and detection response was verified. In addition, the model based observer residuals may be designed to distinguish actuator faults from wave disturbances and fin faults from maneuvering responses.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 1998
Accession Number
ADA350913

Entities

People

  • James E. Melvin

Organizations

  • Naval Postgraduate School

Tags

Communities of Interest

  • Air Platforms
  • Autonomy
  • Materials and Manufacturing Processes
  • Sensors
  • Space

DTIC Thesaurus Topics

  • Autonomous Underwater Vehicles
  • Computers
  • Control Systems
  • Damage Detection
  • Detection
  • Detectors
  • Diving
  • Failure Mode And Effect Analysis
  • Mechanical Engineering
  • Navigation
  • Reliability
  • Simulations
  • Test And Evaluation
  • Underwater Vehicles
  • United States Naval Academy
  • Unmanned Underwater Vehicles
  • Vehicles

Fields of Study

  • Engineering

Readers

  • Computational Modeling and Simulation
  • Computer Vision.
  • Inertial Navigation Systems.

Technology Areas

  • Autonomy
  • Autonomy - Autonomous System Control