Human-in-the-Loop Reinforcement Learning for Real-World Robotics Applications

About the Research

The Army Research Laboratory (ARL) has a research opportunity available in the development of human-in-the-loop reinforcement learning (RL) systems. Specifically, ARL is seeking an individual to advance the development of human-in-the-loop deep reinforcement learning techniques for solving complex, real-world robotics applications such as obstacle avoidance, path navigation, and grasping tasks. A successful candidate will have expertise in one or more of the following areas: Robotics, statistical classification and machine learning methods, deep reinforcement learning, optimal control, experimental design, and computer programming. Emphasis will be placed on translational research and technology development that leverages current internal ARL research on human-in-the-loop RL.

Candidate responsibilities include supporting the short-term goal of developing a working proof-of-concept system demonstrating the viability of human-in-the-loop RL control in robotic environments. This involves system development, conducting experiments, publishing papers, and integrating ideas and methods with a multidisciplinary research team.

ARL Advisor: Nicholas Waytowich
Email: nicholas.r.waytowich.civ@mail.mil

About HRED

The Human Research and Engineering Directorate (HRED) is ARL's principal center for research and development focused on optimizing soldier performance and man-machine interactions. HRED examines human performance in perceptual, cognitive, and psychomotor domains to enhance understanding of human capabilities and limitations and assess the impact of emerging technologies on soldier performance. Research areas include intelligent decision aids and interfaces, human control of automated systems, control display and workstation design, and MANPRINT design, analysis, and integration methods. Additionally, HRED develops unique and innovative methods, tools, models, and simulations for measuring and characterizing soldier performance.

About ARL-RAP

The Army Research Laboratory Research Associateship Program (ARL-RAP) aims to significantly increase the involvement of creative and highly trained scientists and engineers from academia and industry in scientific and technical areas relevant to the Army. Participants contribute to the Army's program for meeting future operational needs through scientific research and technological developments in diverse fields such as applied mathematics, atmospheric characterization, simulation and human modeling, digital/optical signal processing, nanotechnology, material science and technology, multifunctional technology, combustion processes, propulsion and flight physics, communication and networking, and computational and information sciences.

A complete application includes:

• Curriculum Vitae or Resume
• Three Reference Forms (email with link to reference form available in Zintellect)
• Transcripts (must verify receipt of degree, student/unofficial copy acceptable)

If selected by an advisor, participants will be required to write a research proposal for submission to the ARL-RAP review panel. The proposal should:

• Relate to a specific opportunity at ARL
• Have a clear objective and defined outcome
• Pursue a specific direction and include an expected period for completion
• Include a background and motivation for the research
• Reference published efforts to improve the proposal

Questions about this opportunity?