Search

This paper presents an end-to-end deep reinforcement learning (DRL) approach aimed at efficiently determining energy-optimal routes for a group of electric logistic vehicles, with the objective of minimizing operating costs. First, an Energy-Minimization Electric Vehicle Routing Problem (EM-EVRP) is formulated with an energy consumption model for electric vehicles, rather than Distance Minimization EVRP commonly favored in the literature. The energy consumption model incorporates several factors such as vehicle dynamics, road information, and charging losses. Then, the problem is reformulated based on the Markov decision process and solved using the transformer-based DRL method. The policy network is designed following the Transformer structure, including an encoder, a feature embedding module, and a decoder, where the feature embedding module is added to provide contextual information. Finally, extensive experiments demonstrate the superior of the proposed DRL method over existing learning-based methods and conventional methods, in solving both EM-EVRP and DM-EVRP. Notably, the formulated EM-EVRP achieves greater cost reduction than the traditional DM-EVRP. • Build an energy consumption model of electric vehicle for Electric Vehicle Routing Problem. • Formulate a novel energy-oriented Electric Vehicle Routing Problem. • Propose a deep reinforcement learning with transformer-based neural network to solve the Electric Vehicle Routing Problem. • Feature embedding module is added between the encoder and decoder of Transformer to provide contextual information. [ABSTRACT FROM AUTHOR]

• An eco-driving control strategy is proposed for vehicles approaching a signalized intersection. • The discharge of the vehicle queue is modeled and predicted. • The proposed strategy is based on a computationally efficient iterative dynamic programming algorithm. • Monte-Carlo simulation using real-world traffic data is performed for in-depth energy efficiency analysis. • Energy-saving potential enabled by the proposed strategy is investigated for both internal combustion engine and electric vehicles. This paper takes into consideration of vehicle queues at the intersection and proposes an energy-efficient driving strategy to improve vehicle energy efficiency and overall traffic throughput in an urban traffic environment. The proposed strategy is applicable for both electric vehicle and internal combustion engine vehicle, and the control framework is formed by three sections, a vehicle queue discharge predictor, a spatial-domain optimal control strategy for energy consumption minimization, and a speed tracker with consideration of collision avoidance constraints. The former is based on the intelligent driver model, which predicts an accurate vehicle queue discharge time. Then the iterative dynamic programming is utilized to find the optimal solutions with fast computational speed. Finally, the optimal speed profile is followed by a Proportion-Integration controller while keeping a safe inter-vehicular distance. A Monte-Carlo simulation is designed to evaluate the energy efficiency of the proposed strategy in the stochastic traffic environment. Compared to the regular eco-approach and departure and constant speed strategies that lack awareness of the queue, significant energy saving can be achieved of the proposed strategy. In addition, three typical cases are selected to study the energy efficiency when the proposed strategy is applied to internal combustion engine and electric vehicles, respectively. The results show the energy efficiency of electric vehicles is less sensitive to the queuing effect at the intersection because of regenerative braking and the overall higher efficiency of the electric motor in contrast to the internal combustion engine, especially in stop-and-go scenarios. [ABSTRACT FROM AUTHOR]