Minimizing Transportation Energy in Urban Environments: An Optimization Approach for Connected EVs

The work proposed in this document explores a wide variety of energy-saving transportation concepts that exploit special characteristics of electric drives. The confluence of three emerging concepts in transportation, namely electric drivetrains, autonomous driving, and networked vehicles, enables the optimization of transportation efficiency in a way that will drastically change modern transportation, especially for passenger and commercial road vehicles. After introducing suitable energy, optimization, and efficiency models, several optimization problems of practical importance are formulated. If the only term in the cost function is transportation energy, and all other conditions are formulated as constraints (maximum acceleration, average speed, etc.), substantial energy cost reductions are possible. In particular, the concept of energy-optimal speed trajectories is highlighted. It is shown that assuming complete traffic situational awareness, i.e., knowledge about infrastructure and traffic flow information, for urban driving, energy-optimal speed profiles can achieve energy savings above 50%, with 30-50% being typical values.

In order to make energy-optimal speed trajectories useful in real-world situations, some modifications to the concept may become necessary, especially in dense traffic situations. Several approaches that aim to change the optimal trajectory with minimal efficiency losses are explored. These include inserting delays, filtering trajectories, and modifying trajectory components. Another approach is the use of energy-optimal speed trajectories for heterogeneous electric vehicle platoons in urban driving conditions. It is demonstrated that ``urban platooning'' is an attractive and viable implementation option for energy-optimal speed trajectories. Optimal speed trajectories are generated for individual vehicles and entire platoons, assuming that they can be executed without errors, as would be the case for self-driving vehicles. The introduced approach only requires the lead vehicle to run the optimization software, while the remaining vehicles are only required to have adaptive cruise control capability. The achieved energy savings are typically between 30% and 50% for stop-to-stop segments in cities. The prime motivation of urban platooning comes from the fact that urban platoons efficiently utilize the available road space, at least partially alleviating the problem of ``mixed traffic'' where some vehicles optimize transportation energy and others do not. An initial glimpse at the role of uncertainties and the incorrect situational awareness information on attainable energy savings is provided, followed by a more extensive analysis of the impact of the length of time horizon on the efficiency gains achieved by energy-optimal speed profiles.

The minimization of transportation energy in cities is important for many reasons, i.e., for environmental, power grid, and range considerations. The overall impact of this proposed work includes the reduction of energy cost, grid power demand, power plant emissions, global warming, as well as an increase in the range of electric vehicles. Furthermore, urban platooning can more efficiently utilize limited space in urban environments.