A Unified Approach for Electric Vehicles Range Maximization via Eco-Routing, Eco-Driving, and Energy Consumption Prediction

Driving range is one of the main obstacles to the wide diffusion of electric vehicles. In order to overcome it without needing to increase battery size and price of the vehicle, one promising solution consists in leveraging advanced driver assistance systems to increase and master the driving range. This paper proposes model-based strategies to predict and optimize the energy consumption of a trip. Before the trip, an energy efficient route is suggested. During the trip a precise prediction of the current driving range is provided, and an optimal speed profile is computed to advise the driver. These strategies take into account the specific vehicle parameters, as well as the topology of the road network in which the vehicle operates, and the real-time traffic conditions. A macroscopic version of the energy consumption model of the electric vehicle is presented in order to use the aggregated real-time data available on typical maps web-services. The road network is modeled as a weighted directed graph adapted to the proposed energy consumption model. The energy driving range and the optimal route are finally obtained by means of a suitable optimal path search algorithm. For eco-driving, a different approach using an artificial neural network has been chosen to enable real-time implementation. As for the human-machine interface, the output of these strategies is finally suggested to the driver via a smartphone application. Experimental results show promising gains as compared to the existing approaches in predicting vehicle energy consumption, in suggesting an efficient route, and in providing eco-driving assistance.