Chiwei Yan
University of California-Berkeley, CA·Engineering
Robotaxis, robovans and urban air taxis could reshape how people move through cities, but they also risk making traffic worse. Without effective ride-pooling, where several riders share a vehicle, many miles could be driven by vehicles with no one on board, something transportation networks have never seen.
- Fields
Engineering
Mathematics & Statistics
Computer Science
- Topics
autonomous vehicles
autonomous mobility
robotaxis
- Grants
- 1 known grant, through Apr 2027
What they work on
- Develop state-aware demand control models that use real-time vehicle and rider movement to shape demand toward ride-pooling in autonomous fleets such as robotaxis, robovans and urban air mobility.
- Build Markov decision process models, which plan sequences of decisions under uncertainty, that run directly on detailed city street networks with tens of thousands of nodes and edges instead of coarse zones.
- Design custom approximation algorithms that overcome the curse of dimensionality in city-scale decision models, so dynamic pricing and pooling policies can be computed for whole cities.
- Prove performance bounds for these approximate policies, including worst-case guarantees on any network and guarantees as networks grow from campus to metropolitan scale.
More about the research
Autonomous fleets also offer new opportunities, including flexible vehicle designs and central control of the whole fleet, which can make shared rides practical at scale. This exploratory project asks how to manage rider demand so that more people share autonomous vehicles.
Why it matters
These methods could help autonomous fleet operators set dynamic fares that make ride-pooling work at scale, and help local transportation authorities design dynamic subsidies that encourage sharing and reduce congestion. If successful, the same approach could apply to bundling freight and consolidating last-mile deliveries, which could lower logistics costs and carbon emissions.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Rajan Udwani
University of California-Berkeley, CA·Engineering·Early-career
Formulate new models of online resource allocation that capture the complex features, objectives and constraints of applications such as shared mobility, electric-vehicle battery swapping and online advertising.
Yiling Zhang
University of Minnesota, MN·Engineering
Build a hierarchical stochastic optimization framework that links short-term adaptive user behavior to long-term infrastructure design, applied to building energy management and last-mile package delivery with parcel lockers.
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.
The school
University of California-Berkeley: 46 labs on PI Pages, $1.1B a year on research, 825 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.