Autonomous Ridesharing using Machine Learning
Interdisciplinary Areas: | Data/Information/Computation, Smart City, Infrastructure, Transportation |
---|
Project Description
Faster delivery of nearly everything is an important problem. Shoppers - whether it comes to a new smartphone, groceries or a sofa - increasingly want their orders to arrive at their doorsteps as soon as they click a button. With shipping costs rising and freight volumes outpacing the supply of available trucks, companies are thinking of radical new initiatives to get their products into customers hands more easily, helping to transform shopping as we know it. Thus, innovations to enable intelligent freight transportation and to drive future growth with new business models are of central importance. This project aims to develop a novel deep reinforcement-learning-based framework for autonomous freight transportation management.
Start Date
May 2020
Postdoc Qualifications
The postdoctoral scholar should have PhD in CS/ECE or related areas.
Co-advisors
Vaneet Aggarwal
School of IE and ECE (courtesy)
https://engineering.purdue.edu/CLANLabs
Satish Ukkusuri
School of CE
http://www.satishukkusuri.com/
References
Abubakr Al-Abbasi, Arnob Ghosh, and Vaneet Aggarwal, "DeepPool: Distributed Model-free Algorithm for Ride-sharing using Deep Reinforcement Learning," Accepted to IEEE Transactions on Intelligent Transportation Systems, Jul 2019, available at https://arxiv.org/abs/1903.03882
Yimeng Wang, Yongbo Li, Vaneet Aggarwal, and Tian Lan, "DeepChunk: Deep Q-Learning for Chunk-based Caching in Data Processing Networks," in Proc. Allerton, Oct 2019.
Bao, J., Liu, P., & Ukkusuri, S. V. (2019). A spatiotemporal deep learning approach for citywide short-term crash risk prediction with multi-source data. Accident Analysis & Prevention, 122, 239-254.
Qian, X., Zhang, W., Ukkusuri, S.V. and Yang, C. (2017). Optimal Assignment and Incentive Design in the taxi group ride problem. Transportation Research Part B.