Enabling dexterous physics-based manipulation via a learning framework for shared autonomy

Abstract

There is a growing need in the navy for technology that can assist with the everyday maintenance of maritime infrastructure. Several" of these tasks involve complex interactions with tools and with the world, including turning valves, calibrating gauges, cleaning a""nd operating switchboards, among others. Performing these tasks is incredibly challenging for roboticmanipulators because it involv"es reasoning beyond pick-and-place actions that are very successful in factory floors. They demand physical reasoning and commonsens"e manipulation that involves understanding the consequences of pushing, pulling, and sliding objects in the world, via nonprehensile"" actions. As a result, current robotics and HRI systems often fail atperforming these tasks effectively even in research labs, much" less in real shipboard environments. Our first thrust is to enable physics-based reasoning for autonomous manipulation. This capability is critical to free robots out of the factory floors and into real-worldenvironments. We will address key challenges in this d"omain by building tractable models for realistic physics, using these models for manipulation planning addressing robustness under u""ncertainty in actuator and pose estimates as well as physics model parameters, and incorporating feedback, especially via propriocep"#NAME?

Document Details

Document Type
DoD Grant Award
Publication Date
Jun 09, 2017
Source ID
N000141712617

Entities

People

  • Siddhartha Srinivasa

Organizations

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

Tags

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Robotics and Automation.
  • Team-Based Human-Centered Cognitive Task Decision Making and Information Performance.

Technology Areas

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