Cyber security challenges in self-driving cars

Abstract

Self-driving cars, which guide themselves and perform operations by sensing their environment without human intervention, rely on sensors, actuators, complex algorithms, machine-learning systems and powerful processors to execute software. Their benefits include reducing accidents – including those caused by human errors such as drink-driving or using mobile phones, both of which are major causes of road carnage. A major drawback is that these vehicles are vulnerable to hacking because they are connected to the Internet. Manufacturers have developed an array of security mechanisms to shield these automobiles from unauthorised access, yet hackers continue to develop methods of bypassing their security layers.