Scholarlink

This paper primarily focuses on the research of braking energy recovery control strategies for multi-axle electric heavy-duty truck vehicles. Multi-axle truck vehicles have high requirements for energy supply and braking stability control. However, the direct application of traditional braking energy recovery methods to multi-axle vehicles encounters significant limitations. To address these issues, this paper proposes a braking energy recovery control strategy for multi-axle electric heavy-duty truck vehicles based on the twin delayed deep deterministic policy gradient algorithm (TD3), which is a Reinforcement Learning (RL) algorithm grounded in deterministic policy gradients. This strategy learns and adapts through interaction with the environment to achieve optimal control, thereby effectively enhancing the energy recovery rate. The effectiveness of the proposed method is verified through simulations using Trucksim and Simulink.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Reinforcement Learning-Based Braking Energy Recovery Control Strategy for Multi-axle Electric Heavy-Duty Truck Vehicles

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Pengxiang Yan,
  • Liyu Yang,
  • Jiarui Zhang,
  • Chuanxi Chen,
  • Qingsong Gao,
  • Yi Liu

摘要

This paper primarily focuses on the research of braking energy recovery control strategies for multi-axle electric heavy-duty truck vehicles. Multi-axle truck vehicles have high requirements for energy supply and braking stability control. However, the direct application of traditional braking energy recovery methods to multi-axle vehicles encounters significant limitations. To address these issues, this paper proposes a braking energy recovery control strategy for multi-axle electric heavy-duty truck vehicles based on the twin delayed deep deterministic policy gradient algorithm (TD3), which is a Reinforcement Learning (RL) algorithm grounded in deterministic policy gradients. This strategy learns and adapts through interaction with the environment to achieve optimal control, thereby effectively enhancing the energy recovery rate. The effectiveness of the proposed method is verified through simulations using Trucksim and Simulink.