The invention relates to an HEV energy management method based on deep reinforcement learning in a vehicle following environment, and belongs to the field of intelligent hybrid electric vehicle control. The method comprises the steps of determining the shortest safety distance and the longest following distance between two vehicles based on a real-time vehicle speed and other information of following vehicles; establishing a tandem DDPG algorithm model, and learning a vehicle following speed planning strategy and an energy management control strategy; defining a variable space in the DDPG algorithm; when a total reward value obtained in each iteration process tends to be in a stable convergence state and a learning effect is ideal, ending an offline training process and storing a persistence model; and formulating evaluation indexes, and taking a dynamic planning algorithm as a comparison standard of an energy management strategy. Aiming at the intelligent hybrid electric vehicle withan automatic vehicle following function, the speed planning of the vehicle following environment and the energy management of a transmission system are completed by utilizing a deep reinforcement learning algorithm, and comprehensive control of the whole vehicle is realized.
本发明涉及一种跟车环境下基于深度强化学习的HEV能量管理方法,属于智能混合动力汽车控制领域。该方法包括:基于跟随车的实时车速等信息,确定两车之间的最短安全距离以及最长跟车距离;建立串联式DDPG算法模型,学习跟车速度规划策略以及能量管理控制策略;定义DDPG算法中的变量空间;当每次迭代过程所获得的总奖励值趋于稳定收敛状态并且学习效果较为理想后,结束离线训练过程并且保存持久化模型;制定评价指标,并且利用动态规划算法作为能量管理策略的对比标准。本发明针对具有自动跟车功能的智能混合动力汽车,利用深度强化学习算法,完成跟车环境的速度规划以及传动系统的能量管理,实现整车的综合控制。
HEV energy management method based on deep reinforcement learning in vehicle following environment
一种跟车环境下基于深度强化学习的HEV能量管理方法
2020-10-30
Patent
Electronic Resource
Chinese
IPC: | B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion |
Vehicle following speed control method based on deep reinforcement learning
European Patent Office | 2023
|European Patent Office | 2024
|Hydrogen fuel unmanned aerial vehicle energy management method based on deep reinforcement learning
European Patent Office | 2021
|Car following energy management method based on safety reinforcement learning
European Patent Office | 2023
|