Vehicular cloud computing is a technological paradigm shifting which takes advantage of cloud computingto provide vehicles with useful computing resources and services. With the rapid advancementof intelligent vehicles and Intelligent Transportation System infrastructures, some researches on theliteratures have put forth a vision of the combination of vehicular network with could computing inrecent. However, little efforts have concentrated on security features for vehicular cloud services. Inparticular, privacy is one of the critical security issues in vehicular cloud as well as vehicular communicationssince a third-party entity may be involved in cloud service management and operations. Inthis paper, we design a privacy preserving vehicle-to-infrastructure cloud access management systemin which neither global eavesdropper nor any single system management entity can trace a vehiclefor service provision. We make use of pre-loaded pseudonyms to generate anonymous service accesstokens for vehicles and RSU local revocation check to reduce the size of revocation list in the system.