Remaining Driving Range Estimation of Medium-Duty Electric Trucks During Delivery

Fleet electrification has attracted substantial interest in the trucking industry, particularly in last-mile delivery fleets. However, the limited driving range of electric trucks (eTrucks) remains a significant challenge in this endeavor. Accurate estimation of the remaining driving range (RDR) during delivery can not only assist route planning but also help alleviate the range anxiety of drivers. This paper presents an RDR estimation algorithm that consists of offline historical data processing and online range estimation for medium-duty electric trucks in delivery scenarios. Leveraging real driving data, a traffic phase classification Markov chain and a driver behavior model are developed to forecast future driver maneuvers and vehicle velocities. The RDR is calculated by considering the energy consumption of upcoming road segments and the projected final state of charge (SOC). Extensive evaluations under various scenarios demonstrate the algorithm's effectiveness, with an estimation error of less than 4 km and a 52.1% reduction in error compared to the widely used SOC-based method. The results demonstrate the algorithm's potential to enhance RDR estimation in delivery operations, offering opportunities for improved efficiency and management of electric trucking fleets.