Autonomous electric vehicles (AEVs) provide unique opportunities to cope with the uncertainties of distributed energy generation in distribution networks. However, the effects are limited by both inherent radial topology and the behaviors of decentralized AEVs. In this paper, we investigate the potential benefits of dynamic distribution network reconfiguration (DDNR), taking into account AEVs’ spatial-temporal availability and their charging demand. We propose a mixed integer programming model to optimally coordinate the charging/discharging of AEVs with DDNR, while satisfying AEVs’ original travel plan. Numerical studies based on a test system overlaying the IEEE 33-node test feeder and Sioux Falls transportation network show that DDNR and AEV complement each other, which improves the operation of distribution system. We also conduct sensitivity analyses on inputs including renewable fluctuation and AEVs penetration level.