The rising wave of attacks on communicating embedded systems has exposed their users to risks of information theft, monetary damage, and personal injury. Unprotected remote access to connected features on conventional vehicles already can provide attackers operational control. With the impending introduction of autonomous vehicles, their increased connectivity of autonomous vehicles will offer even more avenues for attack. In this paper, we present a survey of previous attacks on connected vehicles, and the prospective security risks threatening autonomous vehicles. To address these vulnerabilities, previous projects on connected vehicle security, such as EVITA, proposed countermeasures including key distribution and Hardware Security Modules. The expense and overhead of these solutions can be high, leading us to place them only where necessary. We propose developing systems using our toolkit TTool, which locates insecure communications, and then determines performance impact of these countermeasures through modelling and formal verification.