Enabling Robust Persistent Autonomy in Robots

Abstract

The long-term objectives of this project are to increase the level of autonomy of robots by giving them an introspective capability. Introspection involves modeling the robot's own behavior and using that model to reflect on how to change its behavior in response to unanticipated events. As a part of this project, we have developed predictive models of a number of robot models for testing these algorithms, including an autonomous car, an autonomous airship, a pedestrian robot navigating among people, and a manipulation scenario involving cloth folding. We made use of these models to demonstrate three primary goals: failure handling, mission planning, and adaptable specification. Recommendations for future research directions in persistent autonomy are included.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 22, 2020
Accession Number
AD1104271

Entities

People

  • Ross Knepper

Organizations

  • Cornell University

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Air Force Research Laboratories
  • Aircrafts
  • Algorithms
  • Artificial Intelligence
  • Autonomous Navigation
  • Autonomous Systems
  • Autonomy
  • Complex Systems
  • Engineering
  • Motion Planning
  • Natural Languages
  • Predictive Modeling
  • Robot Navigation
  • Robotics
  • Robots
  • Unmanned Vehicles
  • Vehicles

Readers

  • Aerodynamics/Aeronautics.
  • Artificial Intelligence
  • Robotics and Automation.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • AI & ML - DoD AI Strategy
  • Autonomy
  • Autonomy - Autonomous System Control