UAH researcher wins three grants totaling $650K to tap learning-based controls

unmanned ground vehicle
Research advances autonomous agents to integrate artificial intelligence, robotics and human teaming elements in challenging environments.
Courtesy U.S. Army

Benefits Smart Personal Protective Equipment, collaboration in robotic inspections and hazardous environments

A researcher at The University of Alabama in Huntsville (UAH) has been awarded three separate federal grants from the National Science Foundation (NSF), the Naval Surface Warfare Center and the Department of Energy totaling nearly $650,000 to support diverse research projects that each involve his particular areas of expertise, learning-based control and its application in machinery lifecycle management.

Dr. Avimanyu Sahoo, an assistant professor in the Department of Electrical and Computer Engineering at UAH, a part of the University of Alabama System, will be supporting three-year studies for each project funded. The projects include seeking to advance personal protective equipment (PPE) through smart technologies; enhancing networked collaborative control tasks in uncertain or dangerous environments like marine exploration and disaster management; and developing AI-enabled autonomy of robotic inspection platforms to improve the sustainability of energy infrastructure.

Making safety a priority in dangerous environments

The project to fund research into Smart PPE is being funded by the NSF for $178,937, slated to begin in October 2023, and running through September 2026. The project entails establishing a collaborative Research Experiences for Undergraduates (REU) site to provide students with multidisciplinary research experiences in developing Smart Personal Protective Equipment (SmaPP). SmaPP advances certain types of personal protective wear where intelligent technology has been incorporated into the PPE to capture and monitor hazardous environmental data or positional information of the wearer to improve safety and response effectiveness in higher-risk work environments.

“The REU site will provide the undergraduates with experience in addressing four fundamental challenges in developing Smart Personal Protection Equipment,” Dr. Sahoo explains. “They are the development of new smart materials; incident heat flux measurement on the surface of SmaPP; intelligent wireless sensing technology for human vital signs and radiation monitoring and human perception of SmaPP. The REU site will involve nine undergraduate students in research activities related to the development of SmaPP for ten summer weeks each year.”

Michael Mercier | UAH

Working with SmaPP involves the use of a product called auxetic materials. Auxetic cellular structures are comprised of a number of unit cells arranged so that the overall structure expands when stretched and contracts when compressed. The REU will feature a number of sample projects to investigate SmaPP components, including strength and impact resistance test and analysis of the auxetic materials; design optimization and prototype development of auxetic structures; incident heat-flux measurement on the SmaPP surface; developing wireless sensors for monitoring human vital signs and monitoring radiation exposure and protective mask usage in a campus environment.

“This integrative project will expose undergraduate students to cutting-edge technologies,” Dr. Sahoo notes. “This will prime the next-generation workforce for a future dominated by these intelligent systems. Research into Smart PPE and adaptive clothing is rapidly expanding. By leveraging advancements in materials science, flexible self-sustaining electronics, the Internet of Things (IoT) and Artificial Intelligence (AI), there's immense potential for creating next-generation PPEs. These advanced protective solutions promise to greatly enhance the safety and efficiency of first responders, military personnel and others operating in high-risk environments.”

Helping collaboration between ships, aircraft and UAVs

The Naval Surface Warfare Center will be the funding agency for a project to investigate intelligent control of Networked autonomous, heterogeneous agents, or NAHA, that will be vital for working in uncertain environments. The initiative will run from June 2023 to May 2026, with a total grant to UAH of $229,837.

“Networked autonomous, heterogeneous agents performing collaborative tasks are ubiquitous in diverse applications, such as marine exploration and disaster management,” Dr. Sahoo says. “To complete these tasks in an uncertain environment, NAHA requires significant communication bandwidth over the wireless channels and computing resources for learning and coordinated control. However, higher communication (data-sharing), for example, in reconnaissance operations, threatens privacy and risks the security of NAHA performing coordinated tasks. Moreover, once a cyber-attack or fault is detected, it is challenging to swiftly isolate the compromised agent to restore even degraded operation before it cascades to others.”

In tasks involving spanning expansive terrains, such as reconnaissance or exploration, utilizing a collaborative approach with multiple agents—robots, unmanned aerial vehicles (UAVs), or unmanned ground vehicles (UGVs)—becomes essential, the scientist reports. Due to the unpredictable nature of their operating environments, the agents are endowed with adaptive and learning capabilities to efficiently navigate and perform. Critical to their operation is inter-agent communication, facilitating collaborative and distributed control.

“Traditionally, much of the research in this domain has centered around consensus-based control, ensuring that these agents adhere to desired positions and velocities while assimilating environmental knowledge,” the researcher says. “However, the efficacy of such systems is contingent upon uninterrupted communication. Any communication lapse or delay can jeopardize mission success. While the foundational research holds broad applicability, it is especially relevant in naval contexts where collaboration between ships, aircraft and UAVs is essential for task execution.

“The success of the proposed research will lead to transformative solutions. Looking ahead, we foresee a future where autonomous, Heterogeneous agents collaborate seamlessly under communication constraints. These agents will be equipped with advanced learning capabilities, allowing them to execute individual tasks efficiently, even in the face of challenges such as unexpected communication breakdowns, agent malfunctions and unpredictable environments.”

Tapping the powers of AI for robotic inspection vehicles

Dr. Sahoo’s third project was awarded $240,920 in federal funds from the Department of Energy to improve the autonomy of AI-enabled robotic inspection platforms to support the sustainability of the nation’s energy infrastructure. The initiative is already underway, slated to run from February 2023 to January 2026.

Robotic visual inspection (RVI) technologies are employed for inspecting pipes, tanks and other hard-to-access parts that are critical to energy sectors, hydrogen production, transport and combustion processes, as well as carbon capture, utilization and storage.

“The RVI assures the integrity of these critical energy infrastructures, limits fugitive emissions and ensures efficient operations,” the researcher says. “The technologies are critical to the safety of human life, emission reduction in the current energy value chain and enable the energy transition in a low carbon manner for the sustainability of energy sectors. The state-of-the-art RVI technologies employed for visual inspection by industries still require human intervention and expertise for operation, data collection and analysis. This jeopardizes human safety, requires extended time for the inspection and is subject to human error.”

The research aims to develop an integrated AI-driven RVI platform with autonomous dynamic path planning and safe navigation capability for closed-loop data collection and real-time defect identification. The effort will integrate deep learning-based defect identification models for dynamic and safe path and motion planning in real-time using multimodal data.

“Autonomous defect detection, navigation and data collection and processing algorithms will increase the level of autonomy of RVI platforms while contributing to the sustainability of the energy sectors,” Dr. Sahoo says. “The technology will be a leap forward in the area of automated component inspection and analysis enabled by robotics. In the foreseeable future, we envision inspection robots evolving beyond mere semiautonomous operation. These robots will be endowed with sophisticated learning capabilities, enabling them to operate with full autonomy. Not only will they be adept at executing tasks independently, they will also possess the agility to adapt and refine their functions based on real-time feedback from unfamiliar environments. This continuous learning and adaptation promises to revolutionize inspections of energy infrastructure in a low-carbon manner, ensuring both efficiency and heightened accuracy, even in the most unpredictable settings.”