But, the elephant in the room regarding all of the latest advances in automotive electronics in particular is the computational complexity used to monitor, maintain or facilitate virtually every function on a modern car. All but the most basic of modern cars now possess in the region of 30 to 50 processors. Even more remarkable is the fact that a typical premium-level car is driven by approximately 100 million lines of programming – or over 16 times more than the US Air force’s F35 Lightning…
Much has been made of connectivity technology that will one day be available in-car, with claims that a revolution in user experience parallel to that of the smartphone is just around the corner. In the second half of 2013, Blackberry – a major supplier of software for vehicle on-board systems –announced its QNX automotive software system would support “over-theair” (OTA) updates.
This means wireless software updates are now possible for an industry where it is estimated 70% of recalls in 2013 were for electronic/software-related issues. It can only be a matter of time before all adopt the technology. The potential for reduction in the costs of firmware upgrades, as well as the eradication of inconvenience for the owner/driver is immense.
This very scenario played out with Tesla earlier this year after a fire risk was identified with its Model S traction battery. A “recall” involved a OTA system update which reprogrammed how the vehicle recharged itself and introducing adjustable onboard ride-height settings. The ease and speed with which this recall was conducted illustrated how pain-free a recall could potentially be, and served to strengthen the Tesla brand in terms of its customer satisfaction.
There are, however, some fundamental risks. The National Highway Traffic Safety Administration (NHTSA) is due to release the results of a study into vehicle-to-vehicle (V2V) connectivity after it trialled the technology with 3,000 vehicles in Ann Arbor, Michigan in 2012. Amongst the outcomes of the study are draft rulings which mandate V2V devices in vehicles “in a future year” with a view to creating a world where automobiles send position and speed data to one another 10 times per second.
The first risk to address is that of data protection. The NHTSA found that any V2V system would contain some element of tracking capability. One can envisage that liberal-minded groups could object, but this is surmountable when it is considered that every mobile phone has the ability to transmit location data. The data is protected from abuse by law enforcement agencies, and one would hope that the same level of protection is afforded for V2V systems.
Secondly – and more importantly – is the issue of code vulnerability. Vehicles are effectively motorised computers, and it is concerning that relatively little in the way of firewall protection is in place. It has been shown by computer scientists at institutes such as the University of Washington that it is alarmingly easy to “hack” into a vehicle’s control systems wirelessly or through the OBD2 port.
This presents a terrifying prospect which manufacturers must take seriously. One of the companies responding to the threat is Cross Border Technologies, which develops firewall software designed to hive off the sections of an automobile’s “brain” which is devoted to throttle, brake and steering inputs, which is a start. However, the industry needs only to look at the proliferation of software vulnerabilities to see how quickly cyber-related threats could become a reality. There is clear anecdotal and hard evidence that fraudsters and other hackers are increasingly focusing on the vulnerabilities of connected devices.
OEMs can learn from the software industry’s perennial fight against viruses and malicious coding. Cyber-criminals will not stop if thwarted by a first generation ECU firewall and will continue to look for weaknesses. It is ironic that a new generation of technology poised to revolutionise safety and reliability could inadvertently expose us to a new generation of serious risks.