A Unifying AI Architectural Framework for Developing Complex Autonomous Robotic Teammates

Abstract

The goal of this project is to provide an architectural framework that is at once versatile, extensible, and scalable, and enables h,uman-like mixed-initiative human-robot interactions at human- like levels of interactivity and effectiveness. Specifically, the fram,ework will have several notable features: (1) it will be hardware- neutral (to the maximum extent possible) and thus work with almos,t any autonomous robotic platforms (as long as the platform provides API interfaces or ways to communicate with sensors and controll,ers); (2) it will provide a systematic interface for perception, navigation, and manipulation and allow for the seamless integration, of existing perception, navigation, and manipulation algorithms even at run-time;(3) it will provide a systematic integration of RO,S-based modules that can be included automatically and utilized by the all higher-level cognitive functions of the architecture (e.g,., planning, reasoning, natural language, etc.); (4) it will provide systematic introspective and natural language access to all cap,abilities enabled by any of its component algorithms, including any added algorithms; and (5) it will provide all necessary task pla,nning, mental-state inference and natural language interaction mechanisms to enable advanced natural language human-robot interactio,ns in mixed-initiative teams, including tasking and teaching interactions.Approved for Public Release.

Document Details

Document Type
DoD Grant Award
Publication Date
May 16, 2022
Source ID
N000142212206

Entities

People

  • Matthias J Scheutz

Organizations

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

Tags

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Database Systems and Applications
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • Autonomy
  • Autonomy - Human-Robot Interaction