Advancing Subsea Manipulation through Dextrous Sensor Placement

Abstract

In this project we propose new technologies to enable multi-vehicle cooperation for inspection and manipulation in underwater environments. We envision a team of semi-autonomous, high-degree of freedom robots, similar to current remotely operated vehicles (ROVs), capable of performing complex observation and manipulation tasks (e.g. hull cleaning, pier maintenance, and environmental sampling), in complex environments with poor visibility. Such a system has the potential to substantially reduce ROV commanders workload and training requirements, improve the efficiency and flexibility of subsea manipulation tasks, and enable the completion of tasks in environments where detailed inspection and dexterous manipulation are not currently possible due to poor operating conditions and lack of qualified ROV operators. Our goal is to develop a semi-autonomous underwater system where the commander describes high-level mission specifications and the system provides support to complete the task efficiently and effectively. Initial work toward the vision of flexible and robust subsea manipulation was funded by ONR in FY2019 to advance fundamental research in control, perception, planning, and human operator interaction for close-up operation and manipulation, including object placement and retrieval (N00024-20-F,-8705) with an arm on a fixed test stand, and follow-on work (N00014-21-1-2052) will consider the necessary sensing, control/optimization, decision support, and soft materials technologies to enable the manipulator to perform complex tasks in energetic environments while mounted on a mobile surrogate-ROV test stand. The work proposed in this document complements these existing efforts by developing multi-vehicle coordination techniques to enable operations by teams of robots in environments where existing single-ROV systems are insufficient. The proposed work will expand both inspection and manipulation capabilities to multiple robots operating in the same environment. We will demonstrate these novel techniques using formations of low-cost BlueROV2 robots. Taken together, these contributions provide a major step forward in the capabilities of multi-ROV systems in challenging perception-limited environments.

Document Details

Document Type
DoD Grant Award
Publication Date
Mar 11, 2022
Source ID
N000142212196

Entities

People

  • Aaron Marburg

Organizations

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

Tags

Fields of Study

  • Computer science

Readers

  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Robotics and Automation.
  • Unmanned Aerial System (UAS) Autonomous Capabilities and Mission Reconnaissance.

Technology Areas

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