TECHLAV 3rd Annual Meeting

Followed by successfully organizing TECHLAV’s second Annual Meeting in 2017, the third Annual Meeting was held from May 31, 2018 to June 1, 2018 in Greensboro, NC. The meeting brought together researchers from academia, military, and industry, as well as, students from North Carolina A&T State University, the University of Texas at San Antonio and Southwestern Indian Polytechnic Institute. The visit provided updates on various tasks, modeling, control, testing, and evaluation of autonomous vehicles, as well as demos and poster presentations on the most recent TECHLAV research outcomes. In addition, there was a technical panel and two keynote speakers who gave talks deeply embedded in the current world of autonomy. For any information or any questions, please contact Mrs. Shar Seyedin, the TECHLAV Program Manager at (336) 285-3260, or sseyedin@ncat.edu. The agenda for this meeting is available here.

Keynote Speakers

Lee M. Seversky, Ph.D.

Autonomy, Command & Control Portfolio Lead
Air Force Research Laboratory Information Directorate

Abstract: "Air Force Machine Learning Opportunities & Challenges”

The global pace of research and development in the areas of machine learning and artificial intelligence combined with seemingly limitless application holds large potential for not only the commercial industry but also for the Air Force and larger defense communities to revolutionize future operations. A key AF challenge is the ability to learn with limited data, to adapt to new mission needs, and to combine the expertise and experiences of highly-skilled Air Force professionals with the speed and scale of machines. This talk will focus on the research challenges associated with interactive learning, where the goal is to develop interactive techniques for collecting, training, and deploying learned models that are tailored to specific problem and mission needs. The talk will highlight ongoing research at the Air Force Research Laboratory and present and discuss open research challenges.

Bio:

Dr. Lee Seversky is the Autonomy, Command & Control, and Decision Support Core Technical Competency (AC2 CTC) Portfolio Lead for the Air Force Research Laboratory. In this role, Dr. Seversky leads the strategy development and technical investments across government, industry, and academia for the Air Force Research Laboratory. Dr. Seversky oversees a portfolio spanning fundamental research and advanced prototyping with focus on developing core AF competencies in the areas of Complex Adaptive C2 Systems, Complex Effects Analysis, and Machine Intelligence. Prior to serving as CTC lead, Dr. Seversky served as principal researcher for the Complex Networks and Information (CNI) research group at the Information Directorate, Air Force Research Laboratory, where he led the research and development of new data-driven techniques for the efficient processing and analysis of complex information spanning a diverse set of traditional and nontraditional sources. In this role, Dr. Seversky led multiple efforts spanning fundamental and applied research where he fostered extensive collaborations with leading academic researchers to identify, develop, and fast track cutting edge emerging research with USAF applications. His research interests include statistical techniques for concise modeling and tracking of complex dynamic data, machine learning (including data-efficient & interactive learning), information inference under sparse and incomplete measurements, and active sampling techniques.



Jonathan P. How, Ph.D.

Director of the Ford-MIT Alliance
Richard Cockburn Maclaurin Professor,
Massachusetts Institute of Technology,
Editor-in-Chief of IEEE Control Systems Magazine

Abstract: "Planning Under Uncertainty: Theory and Practice"

This talk will describe the recent progress in the planning, learning, and control of autonomous systems operating in dynamic environments, with an emphasis on addressing the planning challenges faced on various timescales. For example, autonomous robotic agents need to plan/execute safe paths and avoid imminent collisions given noisy sensory information (short timescale), interact with other dynamic agents whose intents are typically not known (medium timescale), and perform complex cooperative tasks given imperfect models and knowledge of the environment (long timescale). These planning tasks are often constrained to be done using onboard computation and perception, which typically adds a significant complexity to the system. The talk will highlight several recently developed solutions to these challenges that have been implemented to demonstrate high-speed acrobatic flight of a quadrotor in the unknown, cluttered environments; autonomous navigation of a ground vehicle in complex indoor environments alongside pedestrians; and real-time cooperative multiagent planning with an onboard deep learning-based perception system.

Bio:

Dr. Jonathan P. How is the Richard C. Maclaurin Professor of Aeronautics and Astronautics, and has been honored for his contributions to guidance and control of air and space vehicles. He currently serves as the head of the information sector within the Department of Aeronautics and Astronautics, is the Director of the Ford-MIT Alliance; he was a member of the USAF Scientific Advisory Board (SAB) from 2014-2017. His research focuses on planning and learning under uncertainty, and he was the control lead for the MIT DARPA Urban Challenge team. Other research interests include the design and implementation of distributed robust planning algorithms to coordinate multiple autonomous vehicles in dynamic uncertain environments; robust and adaptive control to enable autonomous agile flight and aerobatics; and reinforcement learning for real-time mechanical and aerospace applications.



Panel Discussions

Panel Theme: “Operationalizing Autonomy”

Moderator: Dr. Eddie Tunstel, Associate Director, Robotics, United Technologies Research Center (moderator)


Background: Unmanned vehicles have a large impact on DoD operation. Despite the great deal of progress that has been made thus far, still unmanned vehicles’ technologies are at initial stages of innovation for autonomy. This panel will discuss the challenges and future outlook for operationalizing autonomy toward DoD interests and missions.


Questions for Panelists:
  • What is operational autonomy and what does that mean to you?
  • How can we quantify or rank the degree or level of operational autonomy for DoD autonomy technologies?
  • What are the critical technical and technological challenges for operationalizing autonomy technologies?
  • What are the non-technical challenges/barriers for operationalizing autonomy?
  • How does the test and evaluation (T&E) community play a role in facilitating operationalizing autonomy technologies?
  • How can academia play a role in operationalizing autonomy technologies?