Driver's anxiety about the remaining driving range of electric vehicles (EVs) has been quite improved by mounting a high-capacity battery pack. However, when EVs need to be charged, the drivers still feel uncomfortable if inaccurate range prediction is provided because the inaccuracy makes it difficult to decide when and where to charge EV. In this paper, to mitigate the EV range anxiety, a new machine learning (ML) method to enhance range prediction accuracy is proposed in a practical way. For continuously obtaining the recent traffic conditions ahead, input features indicating the near-future vehicle dynamics are connected to a long short-term memory (LSTM) network, which can consecutively utilize a relation of neighboring data, and then the output features of the LSTM network with another input features consisting of energy-related vehicle system states become another input layer for deep learning network (DNN). The proposed LSTM-DNN mixture model is trained by exploiting the driving data of about 160,000 km and the following test performance shows that the model retains the range prediction accuracy of 2 ~ 3 km in a time window of 40 min. The test results indicate that the LSTM-DNN range prediction model is able to make a far-sighted range prediction while considering varying map and traffic information to a destination.

This paper presents a real-time vehicle speed planning system called the real-time energy-optimal deceleration planning system (RT-EDPS). Connectivity and autonomous driving technologies that provide map and navigation data, traffic light information, and front detection sensor data are exploited to perceive and forecast upcoming deceleration events. A parameterized polynomial-based deceleration model is employed as the deceleration strategy. Real driving test data that characterize the physical limits of regenerative braking are used to model the objective function and constraints of the parameterized deceleration commands. The proposed RT-EDPS involves two speed-planning strategies with different scales of the planning horizon. Within a deceleration event horizon ascertained by the signal phase and timing data of traffic signals and 3D geographic map data, a dynamic programming-based energy-efficient deceleration strategy is used to generate long-sighted speed profiles. While the scheduled vehicle speed is being tracked, if a preceding vehicle is detected within a predefined threshold range, then a model predictive control-based energy-efficient deceleration strategy that incorporates the traffic status ahead of the ego vehicle and exploits V2V communication with the preceding vehicles is activated to re-plan the speed trajectory of the vehicle to guarantee a collision-free distance for the preceding vehicle. In virtual driving experiments, we record energy-recuperation efficiency gains of over 40% as compared to human drivers.