Altruistic energy management strategy for connected new energy vehicles

The rapid development of intelligent connected vehicle technology has promoted the improvement of individual vehicle passage efficiency in the transportation system. However, traditional energy management strategies mostly focus on optimizing their own vehicles' performance, ignoring the negative impact of vehicle behavior on surrounding traffic flow, which, in turn, leads to a decline in overall traffic efficiency and an increase in energy consumption. To address this issue, this paper proposes an innovative altruistic energy management strategy for connected new energy vehicles based on conflict-of-interest regulation, which balances the interests of the vehicle itself and surrounding vehicles within the framework of reinforcement learning. This strategy establishes a hybrid state space based on shared feature extraction and a hybrid action space based on discrete action embedding, and designs a comprehensive reward function based on the adjustment of egoistic and altruistic weights. The experimental results show that, compared with the traditional intelligent driver model built into Simulation of Urban Mobility and the strategy that only considers the benefits of the vehicle itself, the strategy proposed in this paper reduces the energy consumption of the vehicle itself by 2.98% while reducing the average energy consumption of the surrounding vehicles by 4.26% and increasing the average speed of the surrounding vehicles by 39.73%. This implies that appropriately reducing self-serving behavior and shifting to an altruistic strategy that focuses on surrounding traffic can be an effective way to improve the efficiency and energy savings of the current transportation system.