We will assess the feasibility of using deep reinforcement learning to develop a tool that can optimize driving and charging strategies for electric vehicles that are driven for commercial mobility services. If successful, such a tool would reduce energy costs, lower emissions, reduce power grid impacts, and increase the usability of limited-range electric vehicles for mobility services.
Donadee, J., 2019. "Increasing Adoption of Electric Vehicles in Mobility Services with Deep Reinforcement Learning." Thirty-Third Conference on Neural Information Processing Systems NeurIPS 2019, Vancouver, Canada, December 2019. LLNL-CONF-789379.
Lawrence Livermore National Laboratory • 7000 East Avenue • Livermore, CA 94550
Operated by Lawrence Livermore National Security, LLC, for the Department of Energy's National Nuclear Security Administration.