Performance Evaluation of AI Capabilities in Autonomous and Human-Robot Interaction (HRI) Systems

Overview

To support the National Institute of Standards and Technology’s Measurement Science for
Manufacturing Robotics and Autonomous Systems (MSRAS) Program. The focus of this role is
to advance standard test methods, metrics, and evaluation tools for artificial intelligence,
machine learning, and advanced autonomy within human-robot environments, ensuring these
systems operate with transparent, verifiable, and securely integrated decision-making
frameworks.

Qualifications

• Ph.D. degree in Computer Science, Information and Computer Science, Modeling and Simulation Engineering, Aerospace Engineering, or a closely related quantitative field.
• Experience in: Deep learning models, large language models (LLMs), unsupervised clustering, autoencoders, probabilistic programming, AI planning tools, human-robot interaction, activity/intent recognition, path planning algorithms, state estimation, algorithmic auditing, explainable AI frameworks, usability studies, human-in-the-loop validation, risk-aware system design, 3D rendering, discrete event simulation, motion capture data integration, and inertial navigation system architectures.
• Proficiency in: Python, C/C++, Java, MATLAB, exposure to functional or specialized languages (e.g., Common Lisp), PyTorch, TensorFlow, ROS, OpenGL, probabilistic graphical libraries, LaTeX, 3D CAD modeling, 3D Printing workflow management, and modern developer tooling.

Research Proposal

Key responsibilities will include but are not limited to:

• Conducting state-of-the-art measurement science research in robotics, advanced autonomy, and artificial intelligence systems.
• Utilizing deep learning, large language models (LLMs), reinforcement learning, and unsupervised machine learning techniques to enhance robot capabilities, human objective prediction, and decision-making architectures.
• Developing verification methods, transparent evaluation frameworks, and performance metrics to ensure robotic AI systems and automated sorting or ranking methodologies are inspectable and risk-aware.
• Implementing advanced filtering techniques and symbolic spatial relation models to optimize tracking, navigation, and behavioral classification in highly dynamic environments.
• Programming, simulating, and validating physical and simulated autonomous systems using a variety of modern software, libraries, and probabilistic frameworks.
• Helping develop standards, benchmark scenarios, and performance metrics for human- robot interaction (HRI), autonomous systems, and cooperative robotics integrated into complex environments.
• Developing test apparatuses, digital twins, and virtual/physical testbeds using 3D rendering, CAD, and sensor fusion tools to validate repeatability and reproducibility.
• Collaborating with interdisciplinary teams to design and optimize systems while publishing peer-reviewed research results in high-impact journals and international conferences.