Autonomous & connected mobility
Decision-making, collision-risk evaluation, mixed-traffic control, adaptive cruise control, and infrastructure readiness for automated vehicles.
01 Civil Engineer · Transportation AI Researcher
Shahriar
Austin Beigi
Shahriar Beigi is a civil engineer and transportation AI researcher advancing safer mobility through infrastructure, machine learning, simulation, and connected vehicles.
100+ observed driving scenarios
20+ structures inspected
3 peer-reviewed works
Biography
Research informed by real infrastructure.
Shahriar Austin Beigi is a civil engineer and transportation AI researcher working at the intersection of physical infrastructure, human behavior, and intelligent mobility.
As a Ph.D. candidate at the University of Virginia, he studies how autonomous and connected vehicles make decisions under uncertain traffic conditions. His research combines observed trajectory data, machine learning, simulation, virtual reality, and safety-performance analysis.
His professional experience spans the Federal Highway Administration, HNTB, AECOM, and the New Hampshire Department of Transportation, with work in highway safety, bridge and tunnel inspection, load rating, structural design, and construction coordination. He brings those field constraints into research aimed at safer infrastructure and mobility systems.
- Research
- Autonomous and connected mobility
- Practice
- Highway safety and structural engineering
- Methods
- Machine learning, simulation, VR, and field validation
One connected practice
Bridging physical infrastructure and intelligent mobility.
I am a University of Virginia Ph.D. candidate and civil engineer researching how autonomous and connected vehicles make decisions under uncertain traffic conditions.
My perspective is grounded in real infrastructure: federal highway safety, bridge and tunnel inspection, structural design, load rating, and field construction.
Professional focus
Engineering across
two frontiers.
Machine learning & simulation
Python and PyTorch models, time-series prediction, uncertainty calibration, VR studies, trajectory processing, and validation across observed and simulated data.
Civil & structural engineering
Highway safety, bridge and tunnel inspection, load rating, structural design, roadway audits, roadside barriers, and construction coordination.
Research experience · University of Virginia
Research
Experience.
Four connected research programs spanning observed traffic data, virtual reality, autonomous-vehicle sensing, connected control, and field-test planning.
01Prediction + control
Traffic Data Analysis and Mixed-Traffic Safety Research
University of Virginia · Charlottesville, Virginia
- Developed acceleration predictors and mixed-traffic control methods from 100+ observed car-following scenarios to study longitudinal decisions, rear-end crash risk, and string stability.
- Screened and cleaned 100 NGSIM scenarios through data-quality checks, trajectory smoothing, and recalculation of acceleration before model development.
- Designed separate training, validation, calibration, and test datasets to prevent temporal leakage; measured prediction error with RMSE, MAE, and R² and assessed uncertainty using interval coverage and width.
- Evaluated control strategies for three-vehicle mixed-connectivity platoons using stability and crash-risk measures to assess connected-vehicle behavior.
02Human factors
Virtual Reality Safety Evaluation of Connected Cruise Control
University of Virginia · Charlottesville, Virginia
- Conducted a 20-participant BeamNG/VR study, screened trial validity, and retained 17 participants for paired analysis of connected-cruise-control behavior.
- Built traffic data records for speed, acceleration, headway, throttle, and braking; evaluated countermeasures using time exposed time-to-collision, deceleration rate to avoid crash, and rear-end crash risk index.
- Documented data elements, safety-performance measures, methods, and findings in three technical reports for research sponsors.
03Systems review
Impact of Critical Situations on Autonomous Vehicles and Improvement Strategies
University of Virginia · Charlottesville, Virginia
- Reviewed 50+ studies to characterize camera, LiDAR, radar, and GNSS sensing limitations under fog, glare, contamination, faded lane markings, and construction zones.
- Assessed roadway-infrastructure needs for highway automation and published improvement strategies for lane markings, signs, maintenance, and work zones.
04Real-world validation
Connected-Vehicle Field-Test and Adaptive Cruise Control Planning
University of Virginia · Charlottesville, Virginia
- Prepared a Comma.ai openpilot field-test plan for synchronized vehicle, GPS, and driver-monitoring data collection, staged validation, and driver-override criteria.
- Proposed a risk-adaptive cruise-control strategy using time-to-collision, required deceleration, and stopping-distance margins; defined comparative tests for safety and control trade-offs.
Professional experience
Professional
Experience.
Civil Engineer, GS-11
Federal Highway Administration · EFLHD
Virginia- Promoted from Pathways Intern to full-time direct hire in Technical Services, supporting traffic operations and highway safety across a multiple-project portfolio.
- Conducted roadway safety audits and guardrail warrant evaluations and prepared plans, specifications, and estimates for federal projects.
- Analyzed National Park Service crash data for multiple park-road sites in ArcGIS Pro to support guardrail warrants and roadside safety-barrier decisions.
- Applied MUTCD standards for signs, pavement markings, and signals across five federal projects to improve safety for all road-user groups.
Civil Engineer II
HNTB
Massachusetts- Inspected 20+ structures for MassDOT, MBTA, and SFMTA, including the I-93 and Ted Williams tunnels and the Zakim Bridge.
- Authored inspection plans for 20+ structures and presented findings and specifications to officials across several states.
Civil Engineer
AECOM
Massachusetts- Performed inspections and load ratings for 20+ MassDOT and RIDOT structures, including the I-93 tunnels, against federal and state live-load standards.
- Designed and inspected structures for four owners, including MassDOT, Massport, and Logan Airport.
Civil Engineer
New Hampshire Department of Transportation
New Hampshire- Reviewed structural design plans for state-code compliance and tracked ground-improvement placement on the Spaulding Turnpike widening, including a new Route 4 prestressed-concrete beam bridge.
Field perspective
Infrastructure
takes shape.
Engineering judgment is stronger when analysis stays connected to sequence, access, tolerance, and real construction conditions.
Selected publications
Research in
the record.
2026 · AcceptedTransportation Research Record
Enhancing Cooperative Longitudinal Control in a Three-Vehicle Mixed-Connectivity Platoon Using an Uncertainty-Calibrated Deep Residual Learning Method
2026 · AcceptedIEEE ICVES
A Virtual Reality-Based Evaluation on Safety and Driving Performance of hC3 in Electric Vehicles
2025 · PublishedFuture Transportation
Impact of Critical Situations on Autonomous Vehicles and Strategies for Improvement
Education & technical toolkit
Research depth.
Engineering range.
Ph.D. in Civil Engineering
University of Virginia
Autonomous and connected-vehicle decision-making, collision-risk evaluation, ADAS, simulation, and model validation.M.E. in Civil Engineering
University of New Hampshire
B.S. in Civil Engineering
University of New Hampshire
Machine learning
Python · PyTorch · PINNs · LSTM · CLSTM · residual learning · uncertainty calibration
Simulation & autonomy
SUMO · CARLA · BeamNG.tech · Simio · VR · mixed-traffic control · field-test planning
Engineering software
ArcGIS Pro · MIDAS · CSiBridge · AASHTOWare BRR · AutoCAD
Data & evaluation
Trajectory processing · dataset design · RMSE · MAE · R² · safety-performance measures
Connect
Let’s move
safety forward.
For research collaboration, transportation innovation, civil engineering, or professional opportunities, connect with Shahriar on LinkedIn.
Connect on LinkedIn