DRL-Based Coordinated Optimization Control for Intelligent Connected Hybrid Electric Mining Trucks Under Coupled Dynamic Load-Continuous Slope Conditions

Coordinated speed planning and energy management for intelligent connected hybrid electric mining trucks in car-following scenarios is critical to ensuring transportation safety and achieving low-carbon, energy-efficient operations. However, random disturbances from the preceding vehicle cause unnecessary host vehicle speed fluctuations. Furthermore, dynamic loads interacting with undulating terrain increase operational uncertainty, resulting in highly nonlinear, time-varying energy consumption and safety boundaries. Thus, it remains a pressing challenge to achieve coordinated optimization of car-following speed and energy management, particularly under the coupled effects of dynamic loads and undulating terrain, while simultaneously ensuring driving safety. To address this, this paper proposes a coordinated optimization control method for energy efficiency and driving safety based on the deep deterministic policy gradient (DDPG) algorithm. By incorporating the preceding vehicle’s motion trend into dynamic optimization, the method effectively suppresses unnecessary host vehicle speed fluctuations, thereby further improving its energy efficiency. Simulations on real-world mining conditions demonstrate that the proposed method improves economic performance by 7.1% compared to the MPC-DQN hierarchical optimization approach while ensuring car-following safety. These results highlight its feasibility for balancing transportation safety and energy consumption control under complex conditions, supporting energy-saving and emission-reduction goals in green mine development.