Modeling vehicle trajectories in cities is an important task that can help transportation planners make smart decisions around traffic policies and transportation infrastructure. Often times there is not enough data available for the planners to use, or the data available contains sensitive personal information which, if used, would violate data privacy policies. Without data, uninformed decisions can lead to traffic congestion and inadequate public transportation infrastructure. With the annual costs of traffic congestion exceeding one trillion U.S. dollars worldwide, this is obviously a problem worth addressing. We propose a method of generating new samples of vehicle trajectory data for a given city, leveraging existing GAN models with novel data representations.
Worcester Polytechnic Institute
Major Qualifying Project
All authors have granted to WPI a nonexclusive royalty-free license to distribute copies of the work, subject to other agreements. Copyright is held by the author or authors, with all rights reserved, unless otherwise noted.