综合考量出行时段流量和优惠政策的动态电子计费收费系统以缓解城市交通拥堵.pdf

In Proceedings of the 32nd Conference on Arti铿乧ial Intelligence (AAAI 2018), New Orleans, Lousiana, USA. February 2018 DyETC: Dynamic Electronic Toll Collection for Traf铿乧 Congestion Alleviation Haipeng Chen, Bo An, Guni Sharon, Josiah P. Hanna, Peter Stone, Chunyan Miao, Yeng Chai Soh Nanyang Technological University University of Texas at Austin chen0939,boan,ascymiao,ecsoh@.sg gunisharon@ jphanna,pstone @ Abstract traf铿乧 demands are 铿亁ed and are known a priori, and thus are static in essence. In practice, traf铿乧 demands 铿倁ctuate To alleviate traf铿乧 congestion in urban areas, electronic and cannot be predicted accurately, especially when the toll collection (ETC) systems are deployed all over the traf铿乧 is abnormal (e.g., in case of traf铿乧 accidents, or city world. Despite the merits, tolls are usually pre-determined events). As a result, these static tolling schemes usually and 铿亁ed from day to day, which fail to consider traf铿乧 dynamics and thus have limited regulation effect when traf铿乧 have limited regulation effect. One recent work of Sharon conditions are abnormal. In this paper, we propose a novel et al. (2017) proposes a dynamic tolling scheme called - dynamic ETC (DyETC) scheme which adjusts tolls to t