Enhanced Hierarchical Eco-Driving Control for Electric Vehicles via Global Velocity Optimization With Road Slope Adaptation

The advancement of intelligent transportation systems has catalyzed noteworthy progress in autonomous driving technologies. Confronted with the pressing issue of high energy consumption in road transportation, this study presents an eco-driving strategy tailored specifically for electric vehicles. First, an energy-oriented powertrain model is formulated, laying the groundwork for addressing the eco-driving challenge. Subsequently, the eco-driving problem, segmented by varying road slopes, is reformulated as an optimal control problem and solved by dynamic programming. Through an in-depth analysis of the optimal dataset, the intricate relationship between the optimal speed profile and road slope is extracted by proximal policy optimization reinforcement learning, thereby achieving online and rapid eco-driving velocity planning. Furthermore, a hierarchical eco-driving strategy is developed, comprising velocity planning at the upper layer and model predictive control at the lower layer, alongside velocity re-planning to accommodate time constraints. Simulation and experiment results corroborate the efficacy of the eco-driving scheme that the proposed method can effectively generate suboptimal eco-velocity profiles under varying slope conditions while maintaining fast computational performance. Comparative analysis against velocity planning methods that neglect slope variations underscores the substantial reduction in energy consumption achieved by the proposed approach.