CAREER: Autonomous Underwater Power Distribution System for Continuous Operation

National Science Foundation

Overview:
Success of numerous long-term robotic network missions in space, air, ground, and water is measured by the ability of the robots to operate for extended time in highly dynamic and potentially hazardous operating environments. The proposed work responds to the urgency for development of innovative mobile power distribution systems that lower deployment and operating costs, while simultaneously increasing mission efficiency, and supporting the network’s need to be responsive to changing physical conditions. The overall CAREER goal is to develop a power distribution system that responds to individual robot needs, as well as, overall robotic network goals to guarantee persistence of long-term operation in uncertain and unstructured environments.

The proposed work is informed by the hypothesis that network persistence hinges on the ability to establish stable energy transfer cycles necessary to accomplish coverage specifications, while simultaneously dealing with physical and environmental constraints. To test this hypothesis and as an example of such a system, this work will focus on creating a reliable autonomous recharging system for autonomous underwater vehicles (AUVs) that enables continuous real-time marine observation and data collection in the presence of continuously changing underwater environmental circumstances. The key challenges are two-fold: there are fundamental hardware challenges connected to energy transfer in the harsh underwater environment, but more importantly there are basic network science needs that are novel to a mobile power network. The specific research thrusts for this CAREER work include: 1) Task and Energy Routing Scheduling for Persistent Mission Planning. 2) Efficient Network Path Planning and Coordination to Accomplish Persistent Mission Plan. 3) Experimental Validation through Test-bed Development. 4) Design-based, Research-integrated Education Plan for Broadening Underrepresented Participation in STEM.

Intellectual Merit:
This project builds a roadmap to achieve robust continuous marine autonomy that advances unmanned marine systems ability to perform autonomous long-term missions. More specifically the proposed work will provide: 1) resource based task scheduling, 2) path planning formation for mission and charging, and 3) integration tools for testing. Expected outcomes will overcome the current challenge of significant interruptions during underwater missions due to battery limitations and recharging needs. Through this CAREER proposal, the Pl will establish the theoretical, computational, and experimental foundation for mobile power delivery and onsite recharging capability for autonomous underwater vehicles (AUVs). The developed power distribution system will be able to reconfigure itself depending on the scope of the mission, as well as, the energy consumption needs of the network, the number of operational AUVs and required operation time, recharging specifications, communication and localization means, and environmental variables.

Such a system will play a vital role in real-time controlled applications across multiple disciplines, such as: sensor networks, robotics, and transportation systems where limited power resources and unknown environmental dynamics pose major constraints. All developed tools will be suited for the capabilities of not only low-cost AUVs with limited sensing and computational resources, but also high-tech AUVs with state of the art sensor packages.

Broader Impacts:
The developed active power distribution system focuses on underwater scenarios, but will be transferrable to space, air, and ground missions as well. This type of feasible power distribution solution can be used to optimize: 1) immediate high-risk disaster recovery missions like the Fukushima nuclear plant accident; 2) search missions that require vast underwater inspection and detection like the Malaysia MH370 passenger aircraft; and 3) long-term space observation and monitoring like that of the lunar skylight or Europa space mission. The findings from this project will be disseminated through publications, software sharing, and technology commercialization. The project provides interdisciplinary training opportunities for graduate, undergraduate, and pre-college students, including those from underrepresented groups. Research activities will be integrated with education through curriculum development, outreach and improved GUPPIE design.

Investigator: Nina Mahmoudian

Toward Undersea Persistence

Office of Naval Research

The current challenge impeding advances in the U.S. Navy’s mobility is significant interruptions during undersea missions. Missions such as studying arctic physical environments; understanding the effects of sound on marine mammals; submarine detection and classification; and mine detection and neutralization in both the ocean and littoral environment require persistent operation of unmanned systems in challenging and dynamic environments. The proposed work will create an architecture that integrates three elements of energy, communication, and docking to guarantee undersea persistence where limited power resources and unknown environmental dynamics pose major constraints. The architecture will take into account: the number of operational AUVs required for different operation periods, recharging specifications, communication and localization means, and environmental variables.

The overall goal of this project is: to develop a mobile power delivery system that lowers deployment and operating costs while simultaneously increasing network efficiency and response in dynamic and often dangerous physical conditions. The aim is to create network optimization and formation strategies that will enable a mobile power deliver system to meet overall mission specifications by: 1) reconfiguring itself depending on the number of operational AUVs and; 2) responding to energy consumption needs of the network, situational condition, and environmental variables. The outcome of this work will be a theoretical, computational, and experimental roadmap for building and implementing an autonomous distributed system with mobile power delivery and onsite recharging capability. This roadmap will address fundamental hardware and network science challenges. The long-term outcome of this work will be a persistent and stealthy large area presence of AUV fleets able to perform undersea Navy missions by accurately and autonomously responding to energy needs, situational dynamics and environmental variables.

Investigator: Nina Mahmoudian

Mo Rastgaar

Mo_Main

Dr. Mohammad Rastgaar-Aagah is an assistant professor in the Mechanical Engineering-Engineering Mechanics Department at Michigan Technological University since 2011. Dr. Rastgaar received his Ph.D. degree in Mechanical Engineering from Virginia Tech in 2008. He was a post-doctoral associate in the Newman Laboratory for Biomechanics and Human Rehabilitation at MIT. Dr. Rastgaar is a recipient of 2014 NSF CAREER award.

Dr. Rastgaar is the founding director of the Human-Interactive Robotics Lab (HIRoLab) at Michigan Tech. The research at the HIRoLab is focused on the development of lower extremity assistive and rehabilitation devices for enhanced agility and improved mobility. The research goal at the HIRoLab is to further the critical understanding about the dynamics of gait, especially during different maneuvers, through experiments with human subjects and modeling.

Areas of Expertise

  • Dynamics and Controls
  • Robotics
  • System Identification

 

Blackout? Robots to the Rescue

September 25, 2014—

Big disasters almost always result in big power failures. Not only do they take down the TV and fridge, they also wreak havoc with key infrastructure like cell towers.  That can delay search and rescue operations at a time when minutes count.

Now, a team led by Nina Mahmoudian of Michigan Technological University has developed a tabletop model of a robot team that can bring power to places that need it the most.

“If we can regain power in communication towers, then we can find the people we need to rescue,” says Mahmoudian, an assistant professor of mechanical engineering–engineering mechanics. “And the human rescuers can communicate with each other.”

Unfortunately, cell towers are often located in hard-to-reach places, she says. “If we could deploy robots there, that would be the first step toward recovery.”

The team has programmed robots to restore power in small electrical networks, linking up power cords and batteries to light a little lamp or set a flag to waving with a small electrical motor. The robots operate independently, choosing the shortest path and avoiding obstacles, just as you would want them to if they were hooking up an emergency power source to a cell tower. To view the robots in action, see the video posted on Mahmoudian’s website.

“Our robots can carry batteries, or possibly a photovoltaic system or a generator,” Mahmoudian said. The team is also working with Wayne Weaver, the Dave House Associate Professor of Electrical Engineering, to incorporate a power converter, since different systems and countries have different electrical requirements (as anyone who has ever blown out a hair dryer in Spain can attest).

In addition to disaster recovery, their autonomous power distribution system could have military uses, particularly for special forces on covert missions. “We could set up power systems before the soldiers arrive on site, so they wouldn’t have to carry all this heavy stuff,” said Mahmoudian.

The team’s next project is in the works: a full-size, working model of their robot network. Their first robot is a tank-like vehicle donated by Michigan Tech’s Keweenaw Research Center. “This will let us develop path-planning algorithms that will work in the real world,” said Mahmoudian.

The robots could also recharge one another, an application that would be as attractive under the ocean as on land.

During search missions like the one conducted for Malaysia Airlines Flight 370, the underwater vehicles scanning for wreckage must come to the surface for refueling. Mahmoudian envisions a fleet of fuel mules that could dive underwater, charge up the searching robot and return to the mother ship. That way, these expensive search vehicles could spend more time looking for evidence and less time traveling back and forth from the surface.

The team presented a paper describing their work, “Autonomous Power Distribution System,” at the 19th World Congress of the International Federation of Automatic Control, held Aug. 24-29 in Cape Town, South Africa. Coauthors are Mahmoudian, Weaver, mechanical engineering graduate student Barzin Moridian, electrical engineering undergraduate Daryl Bennett and Rush Robinett, the Richard and Elizabeth Henes Professor in Mechanical Engineering.

Funding has been provided by Michigan Tech’s Center for Agile Interconnected Microgrids.

Nina Mahmoudian

Low-Cost Underwater Glider Fleet for Littoral Marine Research

Office of Naval Research

This research is focused on development of innovative practical solutions for control of individual and multiple unmanned underwater vehicles (UUVs) and address challenges such as underwater communication and localization that currently limit UUV use. More specifically, the Nonlinear and Autonomous Systems Laboratory (NAS Lab) team are developing a rigorous framework for analyzing and controlling underwater gliders (UGs) in harsh dynamic environments for the purpose of advancing efficient, collaborative behavior of UUVs.

Underwater gliders are now utilized for much more than long-term, basin-scale oceanographic sampling. In addition to environmental monitoring, UGs are increasingly depended on for littoral surveillance and other military applications. This research will facilitate the transition between academic modeling/simulation problem solving approach to real-world Navy applications. The importance of this research is evident in the Littoral BattleSpace Sensing (LBS) Program contract at the Naval Space and Naval Warfare Systems Command for 150 underwater gliders, designated the LBS-G. These gliders will be operated by the Navy in forward areas to rapidly assess and exploit environmental characteristics to improve the maneuvering of ships and submarines and advance the performance of fleet sensors.

Research results will provide the coordination tools necessary to enable the integration of these efficient and quiet vehicles as part of a heterogeneous network of autonomous vehicles capable of performing complex, tactical missions. The objective is to develop practical, energy-efficient motion control strategies for both individual and multiple UGs while performing in inhospitable, uncertain, and dynamic underwater environments.

The specific goals of this project are twofold. The first goal is to design and fabricate a fleet of low-cost highly maneuverable lightweight underwater gliders. The second goal is to evaluate the capability of the single and multiple developed UGs in littoral zones. The proposed work will develop UGs that would share the buoyancy-driven concept with the first generation of gliders called “legacy gliders.” However, the NAS Lab UGs will be smaller in size, lighter in weight, and lower in price than legacy gliders. This will result in more affordable and novel UG applications. Moreover, the NAS Lab design to development approach allows for technological innovation that overcomes known challenges and responds to unexpected needs that arise during testing. Therefore, the significance of this research is that it will enable implementation of recently developed efficient motion planning algorithms, multi-vehicle coordination algorithms, and extension of these algorithms in realistic conditions where absolute location and orientation of each vehicle is not known and the time-varying flow field is not locally determined.

 

Investigators: Nina Mahmoudian

Mark Vaughn

xdJJr3eLPtUPZrpPVx4z3KTToTZm8-h2bIOB7P4spj0Dr. Vaughn has joined Michigan Tech as a research professor after retiring from Sandia National Laboratories. His research expertise is in the area of mechanical and electromechanical design, stress analysis, dynamics, and innovative applications. He has over 10 patents, and has been the lead on a broad array of projects for the military.

Areas of Expertise

  • Electro-Mechanical Analysis and Design
  • Energy Storage
  • Hydrogen Peroxide Systems
  • Advanced Payloads
  • Robotic Vehicles
  • Biomedical Devices

Steven Y. Goldsmith

image94336-pers

Dr. Steven Y. Goldsmith holds dual appointments in the Mechanical Engineering and Engineering Mechanics Department, and the Electrical and Computer Engineering Department. He is also a Senior Fellow at the Technological Leadership Institute at the University of Minnesota. Dr. Goldsmith spent 32 years with Sandia National Laboratories and retired as Distinguished Member of the Technical Staff in 2011. While at Sandia he developed information and control systems for many different applications, including nuclear weapon testing, particle beam accelerators, seismic array monitoring, arms control and treaty verification, environmental life-cycle analysis, e-commerce and international trade, electric grid coordination, collective robotics, information warfare, and automated cyber defense. His current research efforts are focused on intelligent agent systems and technology, particularly the development of adaptive and multi-agent systems. His current projects involve the application of intelligent agents to “smart” electric grid controls and microgrids and cyber security of distributed control systems.

Areas of Expertise

  • Adaptive Software Systems and Intelligent Agents
  • Environmental Life Cycle Analysis of Microgrids
  • Multi-agent Systems for Microgrid Control
  • Cyber Security of Distributed Control Systems
  • Automated Cyber Warfare
  • Simulation of Large Scale Cyber Conflict

Nina Mahmoudian

Nina Mahmoudian_Fall2013-1Dr. Mahmoudian’s general research interests lie in the area of dynamics, stability, and control of nonlinear systems. Specifically, she is interested in dynamic modeling, motion planning, and developing cooperative control algorithms to autonomous vehicles. Design and control of autonomous vehicles based on the principles used by nature is another area of interest.  She works on developing analytical and computational tools for the cooperative control of a network of autonomous vehicles in complex environment using nonlinear control and stochastic analysis. The application will be for air, ground, and sea autonomous vehicles.

Areas of Expertise

  • Nonlinear Control and Dynamics
  • Cooperative Control of Multi Agent Systems
  • Autonomous Vehicles with Special Interest in Underwater Gliders

Prepositioned Power Research

Overview

Prepositioned Power RobotsResearch is focused on developing technology to create systems that can autonomously create a microgrid, for situations that require the ability to preposition a basic level of energy infrastructure such as areas damaged by natural or man-made disasters, and autonomously deploying forward operating bases. Modeling and control of robotics and power conversion systems provides the ability to create such prepositioned electric power networks.

Active Projects

Applications

Autonomous Robots can carry a variety of power equipment:

  • Intelligent power electronics for energy conversion
  • Power connection hardware
  • Generation sources, both traditional and renewable
  • Energy storage

 

Prepositioned Power

Prepositioned Power

Four autonomous microgrid robots, each with different power network functionality. Two have renewable energy generation and storage capability, another has a conventional diesel genset, and the third contains intelligent power electronics for conversion and hard-line interconnection, and switchgear. After assessing the power requirements and available resources they would physically organize and electrically interconnect to form a micro-grid.