BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:ENAC Seminar Series by Prof. R. Pedarsani
DTSTART:20200206T094500
DTEND:20200206T104500
DTSTAMP:20260928T184152Z
UID:cb88356eec0e5eae02b09fa75b1de28614e58d7d25760a14418974c1
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Ramtin Pedarsani\n09:45 – 10:45 – Prof. Ramtin Pedar
 sani\nAssistant Professor\, University of California\, Santa Barbara\, USA
 \n\nTraffic Networks with Mixed Autonomy: Analysis\, Learning\, and Contro
 l\n\nTraffic congestion has large economic and social costs. The introduct
 ion of autonomous vehicles can potentially reduce this congestion by incre
 asing road capacity via vehicle platooning and by creating an avenue for i
 nfluencing people’s choice of routes. In this talk\, we consider traffic
  networks with mixed autonomy where a fraction of vehicles are human-drive
 n and the rest are autonomous. We formalize a model of vehicle flow in mix
 ed autonomy based on the fundamental diagram of traffic. We consider a net
 work of parallel roads\, characterize user equilibria\, and provide a poly
 nomial-time algorithm that computes optimal equilibria. Incorporating auto
 nomous ride-hailing services in our model\, we next develop an active pref
 erence-based learning algorithm to learn how people value time and money i
 n choosing their preferred transportation option. This enables learning a 
 model for people’s routing choices in a data-efficient manner. We then f
 ormulate a planning optimization that chooses service prices to maximize a
  social objective. We demonstrate the benefit of the proposed scheme by co
 mparing the results to theoretical benchmarks. Finally\, we study a dynami
 c routing game\, in which the route choices of autonomous cars can be cont
 rolled and the human drivers react selfishly and dynamically to autonomous
  cars’ actions. As the problem dimension is prohibitively large\, we use
  deep reinforcement learning to learn a policy for controlling the autonom
 ous vehicles. This policy influences human drivers to route themselves in 
 such a way that minimizes congestion on the network.
LOCATION:GC B1 10 https://plan.epfl.ch/?room==GC%20B1%2010
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR