TY - GEN
T1 - Fuel Consumption Reduction of Multi-Lane Road Networks using Decentralized Mixed-Autonomy Control
AU - Lichtle, Nathan
AU - Vinitsky, Eugene
AU - Gunter, George
AU - Velu, Akash
AU - Bayen, Alexandre M.
N1 - Funding Information:
∗ Eugene Vinitsky is a recipient of an NSF Graduate Research Fellowship and funded by the National Science Foundation under Grant Number CNS-1837244. Computational resources for this work were provided by the Savio cluster at Berkeley. This material is also based upon work supported by the U.S. Department of Energy’s Office of Energy Efficiency and Renewable Energy (EERE) award number CID DE-EE0008872. The views expressed herein do not necessarily represent the views of the U.S. Department of Energy or the United States Government.
Publisher Copyright:
© 2021 IEEE.
PY - 2021/9/19
Y1 - 2021/9/19
N2 - In this work, we demonstrate optimization of fuel economy in a large, calibrated model of a portion of the Ventura Freeway using a low penetration rate of controlled autonomous vehicles. We create waves in this network using a string-unstable car-following model and introduce a ghost cell to allow waves to propagate out of the network. Using multi-agent reinforcement learning, we then design a controller that manages to partially dampen the waves and thus increase the average energy efficiency of the system, yielding a 25% fuel consumption reduction at a 10% penetration rate. Finally, we investigate the robustness properties of the designed controller. We find that the controller regulates the system to its equilibrium speed over a wide range of speeds and penetrations outside the training set, indicating generalization and robustness.
AB - In this work, we demonstrate optimization of fuel economy in a large, calibrated model of a portion of the Ventura Freeway using a low penetration rate of controlled autonomous vehicles. We create waves in this network using a string-unstable car-following model and introduce a ghost cell to allow waves to propagate out of the network. Using multi-agent reinforcement learning, we then design a controller that manages to partially dampen the waves and thus increase the average energy efficiency of the system, yielding a 25% fuel consumption reduction at a 10% penetration rate. Finally, we investigate the robustness properties of the designed controller. We find that the controller regulates the system to its equilibrium speed over a wide range of speeds and penetrations outside the training set, indicating generalization and robustness.
UR - http://www.scopus.com/inward/record.url?scp=85118470223&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85118470223&partnerID=8YFLogxK
U2 - 10.1109/ITSC48978.2021.9564682
DO - 10.1109/ITSC48978.2021.9564682
M3 - Conference contribution
AN - SCOPUS:85118470223
T3 - IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC
SP - 2068
EP - 2073
BT - 2021 IEEE International Intelligent Transportation Systems Conference, ITSC 2021
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2021 IEEE International Intelligent Transportation Systems Conference, ITSC 2021
Y2 - 19 September 2021 through 22 September 2021
ER -