This position is part of the National Institute of Standards and Technology (NIST) Professional Research Experience Program (PREP). NIST recognizes that its research staff may want to collaborate with researchers at academic institutions on specific projects of mutual interest and, therefore, requires those institutions to be recipients of a PREP award. The PREP program involves staff from a wide range of backgrounds conducting scientific research across various fields. Individuals in this position will perform technical work supporting the collaboration's scientific research.
Research Title: Measurement Science for AI Decision-Making in Automated Driving Systems
The work will entail: The Measurement Science for Automated Vehicles project at NIST is seeking a candidate to support measurement science research for AI decision-making in automated vehicles. This position contributes to NIST's development of a tiered measurement framework for evaluating AI decision-making performance. The candidate will focus on two core areas: 1) building and curating a scenario database for decision-making evaluation and 2) developing standardized state representations for the evaluation framework.
U.S. Citizen PreferredKey responsibilities will include but are not limited to: - Scenario Database Development
- Build and curate a database of behavioral planning test scenarios sourced from Safety Pool™ and other relevant datasets
- Develop scenario classification and tagging systems to support systematic evaluation of behavioral competencies (e.g., lane changes, merges, yielding, intersection navigation)
- Implement tools for scenario selection, parameterization, and configuration for use in simulation-based testing
- Create methods for generating scenario variants to ensure comprehensive coverage of edge cases and challenging traffic situations
- Vehicle State Representation
- Design and implement a standardized format for representing vehicle state information exchanged between the simulation environment and the automated driving (AD) stack under test
- Define world state schemas that capture relevant traffic context, road geometry, and dynamic agent information needed for behavioral planning evaluation
- Develop message schemas and interface specifications for the Evaluation Gateway, including cryptographic hashing methods for data integrity verification
- Ensure compatibility of state representations with industry standards and common AD stack architectures (both end-to-end and modular)
Qualifications - MS or (BS + 2 years of experience) in Computer Science, Robotics, AI/Machine Learning, or related engineering fields
- Strong programming experience in Python and C++, with familiarity with AI/ML frameworks (TensorFlow, PyTorch, etc.)
- Experience with autonomous vehicle simulation environments (CARLA, SUMO, or similar)
- Knowledge of autonomous vehicle systems architecture and behavioral planning concepts
- Experience with ROS 2 on Linux systems
- Experience with version control software and workflow (Git/GitHub/GitLab)
- Understanding of data modeling principles and validation methodologies
- Familiarity with database design and management for storing and querying structured scenario data
