A Trust Based Secure and Privacy Aware Framework for Efficient Taxi and Car Sharing System

1. Introduction

It is not a gainsaying that advent of autonomous vehicles would have notable impacts on the operation of car sharing and taxi services. Meanwhile, advent of autonomous vehicles does not imply that non-autonomous vehicles would suddenly be out of the transport system. It will take time before it fizzles out, therefore, a good car sharing or taxi system framework must be able to accommodate both vehicles. Apart from this kind of heterogeneity required in taxi system framework, sensitive information needs to be exchanged between passengers and drivers in case of non-autonomous or between passengers and vehicles in the case of autonomous vehicles. This brings security and privacy issues. Recent findings have shown that adversaries could intercept the requests of passengers in order to track the whereabouts of the vehicles as well as passengers (Liu, Au, Susilo, & Zhou, 2012) (Boukerche, Oliveira, Nakamura, & Loureiro, 2008) (Li, Dan, & Nahrstedt, 2015). Also, mischievous passenger or vehicle can collude with adversary to obtain the personal details of the other entities in the taxi system. In most of the taxi or car sharing systems, drivers do not know the destination of passengers before accepting their ride requests, thereby putting drivers at disadvantage position especially in a fixed price system. Example of taxi-service system with this kind of lapses is Uber; a popular taxi-service system which allows non-autonomous car sharing, it does not allow driver to know the destination before accepting the ride sharing request from passenger. Also, Uber’s rating model is susceptible to bad mouthing attacks. Apart from this, almost all the existing taxi-service systems are non-autonomous vehicles based, that is, they do not accommodate AVs.