Enhancement of energy utilization efficiency and speed control of autonomous electric vehicles (AEVs): A hybrid approach

Abstract

The efficiency of energy utilization in autonomous electric vehicles greatly impacts their longitudinal motion control. However, the complexity of driving scenes poses challenges to this control. This study introduces a hybrid approach that combines the improved coot optimization algorithm with adaptive reinforcement equilibrium learning to enhance both energy efficiency and speed control in autonomous electric vehicles. The primary innovation lies in optimizing and managing the powertrain efficiency operating point distribution to increase energy utilization efficiency. In the first phase, the improved coot optimization handles vehicle energy utilization efficiency by optimizing operational point transfers. The system normalizes motor torque and velocity to maximize efficiency within constrained conditions. Subsequently, in the second phase, adaptive reinforcement equilibrium learning effectively predicts vehicle speed control on irregular pathways. The proposed technique is implemented on the PYTHON platform to evaluate performance. The analysis also investigates two specific operating conditions: New European Driving Cycle (NEDC) and World Light-Duty Vehicle Test Cycle (WLTC). The findings demonstrate that the proposed strategy effectively optimizes vehicle powertrain efficiency operating point distribution, resulting in improved energy consumption outcomes. The energy utilization efficiency of the proposed approach is 90%, 93%, 95%, 96%, and 98.4%, respectively, at time 100 s, 200 s, 300 s, 400 s, and 500 s.