Dynamic Programming for New Energy Vehicles based on their work modes Part I: Electric Vehicles (EVs) and Hybrid Electric Vehicles (HEVs)

Currently, there are some great challenges in the application of dynamic programming on new energy vehicles. Apart from the two common dynamic programming problems of the interpolation leakage and the dimension disaster introduced in existing literatures, two new problems are found in our study, which are the standardization problem and the Markov problem. In this series of two papers, we introduce our research how to solve these four main problems of dynamic programming in the application on new energy vehicles. The results show that the proposed method not only has the ability to solve the above four dynamic programming problems, but also has much shorter computation time and higher calculation accuracy when compared with the calculation results of Basic Dynamic Programming and Level-Set Dynamic Programming. And, all the cases can be solved successfully.

当前，动态规划在新能源汽车领域的应用仍存在诸多严峻挑战。除现有文献中已提及的插值泄漏（interpolation leakage）与维度灾难（dimension disaster）两类常见动态规划问题外，本研究还发现了两类全新问题：标准化问题与马尔可夫问题（Markov Problem）。在这两篇系列论文中，我们将介绍针对动态规划在新能源汽车应用场景下上述四类核心问题的解决方案。实验结果表明，所提出的方法不仅能够有效解决上述四类动态规划问题，相较于基础动态规划（Basic Dynamic Programming）与水平集动态规划（Level-Set Dynamic Programming）的计算结果，其计算时长更短、计算精度更高，且可成功求解所有测试案例。