[apologies for cross-posting]

Artificial Intelligence to Support the Deployment of Electric Vehicles
Research topic in Frontiers in Future Transportation


Electric Vehicles (EVs) is currently the main pathway to decarbonize the transportation sector and significantly reduce CO2 emissions. This is crucial when trying to cope with the critically negative effects of the climate change. However, to efficiently deploy large volumes of EVs and make them attractive to customers, several problems need to be tackled:

1. Given the sparse charging infrastructure and the relatively long EV charging time, the efficient scheduling of their charging is crucial. This is a challenging problem as it must consider the demand and constraints of the customers, the availability of the charging stations and the constraints of the electricity distribution network. Also, to make EVs truly environmentally friendly, the charging must be done with energy from intermittent renewable sources. Here, the close collaboration of EVs with the Smart Grid is crucial.

2. The EVs have the ability to use their batteries as storage devices when being idle. In this way excess energy can be stored for later use when demand exists. This Vehicle-to-Grid (V2G) mode of operation can significantly increase the storage capacity of the network and, crucially, increase renewable energy utilization.

3. The EVs can recuperate energy under braking or when driving downhill. Thus, energy efficient routing that exploits this EVs’ ability is important to increase the range and reduce the energy demand of the vehicles. This has a positive impact on the environment and the charging infrastructure, as the EVs will need to charge less often.

4. Emerging modes of transportation, such as the Autonomous Vehicles (AV), Connected Autonomous Vehicles (CAVs) and Mobility-on-Demand (MoD), enable different possibilities for the EVs. For example, Autonomous Electric Vehicles (AEVs) can fine-tune their acceleration profile in order to reduce their energy consumption; CAVs may exploit macro-level system decisions, e.g., traffic steering, to obtain congestion avoidance or collaborative energy-efficient path planning; MoD, especially in conjunction with AVs, may exploit complicated optimization problems involving the assignment of EVs to customers.

Controlling EVs demands efficient algorithms that can solve problems that involve a large number of heterogeneous entities (e.g., EV owners), each one having its own goals, needs and incentives (e.g., amount of energy to charge), while they operate in highly dynamic environments (e.g., variable number of EVs) and having to deal with a number of uncertainties (e.g., future energy demand). Some of these challenges can be tackled by powerful Artificial Intelligence (AI) techniques. In this Research Topic, we focus on the use of Artificial Intelligence techniques to cope with the EV-related challenges. We expect research and survey papers in one of the following sectors:

- Charging scheduling- Grid-to-Vehicle
- Dis-charging scheduling- Vehicle-to-Grid
- Renewable energy utilization
- Energy efficient routing
- Customer behavior and incentives provision
- Electronic energy auctions
- Electric vehicles and smart grids
- Electric vehicles and smart metering
- Emerging topics (MoD, Autonomous vehicles, Connected Autonomous Vehicles)

A non-exhaustive list of potential AI techniques to be used is:
- Optimization techniques
- Heuristic and meta-heuristic algorithms
- Multi-agent systems
- Electronic auctions
- Mechanism design and game theory
- Machine learning and data analysis
- Internet of Things
- Knowledge representation

Information about the article types can be found here
and information for preparing your manuscript here

- Deadline for title and abstract registration: January 29, 2021
- Deadline for papers submission: March 30, 2021

Emmanouil Rigas (University of Cyprus)
Christian Vitale (University of Cyprus)
Nick Bassiliades (Aristotle University of Thessaloniki)
Samer Hani Hamdar (George Washington University)