Yichang Tsai
Georgia Institute of Technology, GA·Engineering
Curves are the most dangerous places on a road network. In the United States they account for only about 5 percent of highway mileage but roughly a quarter of all fatal crashes.
- Fields
Engineering
Computer Science
Social Science
- Topics
roadway safety
horizontal curves
curve advisory speed
- Grants
- 1 known grant, through Oct 2027
What they work on
- Develop a crowdsourced multi-run methodology that assesses horizontal curve safety network-wide from low-cost mobile devices, handling the variation introduced by different vehicle types, suspension properties and driver behaviors.
- Build a seed-propagation model and an iterative convergence feedback model that, combined with curve-driving kinematics and spatial-temporal analysis, simultaneously estimate vehicle suspension characteristics and curve geometric properties.
- Create a confidence-level methodology that quantifies how reliable a computed curve advisory speed is, based on a vehicle's recorded trajectory and the roadway geometry, to support data-driven agency decisions.
- Validate both methodologies at network level to show that frequent, low-cost curve safety assessment is technically and economically feasible for transportation agencies.
More about the research
Part of the reason is that the posted advisory speed for a curve can quietly become wrong: pavement ages and develops distress, which changes how much grip a tire has, and resurfacing or maintenance can alter the curve's geometry. Checking every curve on a network for this drift is currently slow, labor-intensive and expensive, so agencies often learn a curve has become hazardous only after crashes happen. This project develops ways to assess curve safety cheaply and often enough to catch problems first, shifting agencies from reacting to crashes toward preventing them.
Why it matters
Horizontal curves carry a disproportionate share of fatal crashes, and outdated advisory speeds are a correctable cause. Making network-wide curve assessment feasible with low-cost mobile devices and intra-agency crowdsourcing could let transportation agencies, especially those with small budgets, find and fix hazardous curves before crashes occur, saving labor and money and extending safety improvements more equitably across jurisdictions.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Michael Levin
University of Minnesota, MN·Engineering
Develop exact solution methods for transportation network design under user equilibrium that combine high-point relaxations, value-function cuts and strong duality cuts to find provably optimal designs faster.
Xuegang Ban
University of Washington, WA·Engineering
Adapt and scale privacy-preserving techniques so that traffic, trajectory, and naturalistic driving data can be shared securely under both centralized and distributed data-sharing models, and evaluate them on real-world transportation datasets.
Seri Park
University of Nevada-Reno, NV·Engineering
Document school bus provision policies and develop built environment readiness (BER) measures that score route-based street design and neighborhood walkability for children who must walk to school.
The school
Georgia Institute of Technology: 50 labs on PI Pages, $1.5B a year on research, 572 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.