Towards Accessible Shared Autonomous Electric Mobility With Dynamic Deadlines

Guang Wang, Zhou Qin, Shuai Wang, Huijun Sun, Zheng Dong, Desheng Zhang

Abstract: Shared autonomous electric mobility has attracted significant interest in recent years due to its potential to save energy consumption, enhance mobility accessibility, reduce air pollution, mitigate traffic congestion, etc. Although providing convenient, low-cost, and environmentally-friendly mobility, there are still some roadblocks to achieve efficient shared autonomous electric mobility, e.g., how to enable the accessibility of shared autonomous electric vehicles in time. To overcome these roadblocks, in this article, we design <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Safari</monospace> , an efficient <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">S</u> hared <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">A</u> utonomous electric vehicle <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">F</u> leet m <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">A</u> nagement system with joint <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">R</u> epositioning and charg <underline xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">I</u> ng based on dynamic deadlines to improve both user experience and operating profits. Our <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Safari</monospace> considers not only the highly dynamic user demand for <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">vehicle repositioning</i> (i.e., where to relocate) but also many practical factors like the time-varying charging pricing for <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">charging scheduling</i> (i.e., where to charge). To perform the two tasks efficiently, in <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Safari</monospace> , we design a dynamic deadline-based deep reinforcement learning algorithm, which generates dynamic deadlines via usage prediction combined with an error compensation mechanism to adaptively learn the optimal decisions for satisfying highly dynamic and unbalanced user demand in real time. More importantly, we implement and evaluate the <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Safari</monospace> system with 10-month real-world shared electric vehicle data, and the extensive experimental results show that our <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Safari</monospace> achieves 100% of accessibility and effectively reduces 26.2% of charging costs and reduces 31.8% of vehicle movements for energy saving with a small runtime overhead at the same time. Furthermore, the results also show <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Safari</monospace> has a great potential to achieve efficient and accessible shared autonomous electric mobility during its long-term expansion and evolution process.