Back in 1985 when Back to the Future II was first released, we all thought hoverboards and flying cars would be our future. Flash forward over 30 years, much of the tech that the film predicted has come true. But one thing it didn’t get right was the flying cars; instead, the closest thing we have are driverless cars. As exciting as that still sounds, the idea of driverless has been around ever since motorised cars were invented back in the 1920s and realised from at least 2009 when Google launched its self-driving car project, Waymo. Ten years later, autonomous cars are not exactly clogging up the streets. So, how close are we to driverless becoming a reality?
As recently as February, Elon Musk expressed that he was ‘certain’ Tesla cars would be able to drive themselves by the end of this year, albeit with a driver at the wheel, and would be completely autonomous by the end of 2020. Considering Musk’s predictions have fallen short before, this doesn’t necessarily mean you’ll find driverless cars on the road by Christmas. However, it does indicate the industry’s optimism. Waymo, widely considered to be the industry leader for self-driving vehicles, began testing their driverless taxis in Phoenix back in December and have just announced that they are expanding their fleet. From these reports, it seems as if a driverless future is well on its way to becoming a reality. However, Waymo has only tested their cars in a small, suburban area of Phoenix and most still have a driver at the wheel. Moreover, it only has permission to test fully driverless cars (i.e. without a safety driver) on public roads in California and Arizona – far from its vision of providing robo-taxi services across America. There are many hurdles the autonomous car industry must first overcome before they can become commonplace: competition from the traditional car industry, gaining consumer trust and complying with new regulations are just a start.
Traditional carmakers are desperately trying to catch up with tech companies who have had a head start in developing driverless tech. Thus, several joint ventures between usually fierce rivals have been announced or discussed in recent months: BMW and Daimler announced in February that they were investing €1bn in a new joint venture to develop mobility services and autonomous driving and Volkswagen and Ford similarly agreed in January to ‘investigate’ ways of working on autonomous vehicles together. There have been similar partnerships between carmakers and tech firms, such as Toyota working with Softbank and Nvidia, a chipmaker, teaming up with Volkswagen. The advantages of collaborating are clear: it reduces costs, lowers risk and allows carmakers to share in the expertise already gathered by these tech firms, all of which speeds up the time it takes to bring products to market. Neither is the relationship one sided: tech firms need the engineering and manufacturing expertise that only the automobile industry can provide.
There is a strong sense among carmakers that they need to innovate or risk becoming obsolete. Electric cars, ride-sharing and autonomous driving will change how we use cars and it is likely that in the future car ownership will fall significantly, considering the high upkeep costs and the increasing availability of other quick and reliable transport services. The auto industry has recognised that potential value lies in services and not ownership; if the current trend continues, it is likely that soon most people will not buy their own cars but rent or borrow them. As a result, carmakers have started to offer their own mobility services such as Daimler’s Car2Go and BMW’s DriveNow for car sharing. The car industry has already started to show signs of difficulty due to falling demand in China, the world’s largest vehicle market, and the shift away from diesel and petrol cars towards electric: Jaguar Land Rover is cutting 4500 jobs due to a sales slowdown in China and Volkswagen is still dealing with the backlash from its ‘dieselgate’ scandal. There is no longer a choice but to innovate – not only are carmakers facing pressure from driverless disrupters, but both the EU and China have introduced new regulations that will result in steep fines for manufacturers if they don’t meet emissions targets.
Another large obstacle that driverless cars must surmount is gaining consumer trust. After all, what would be the point of making cars that no one will get into? This endeavour has not been helped by reports of accidents involving self-driving vehicles, some of which have been fatal: in March 2018, an Uber autonomous Volvo killed a pedestrian in Arizona. It must be noted that these accidents have occurred in cars with only Level 3 autonomy i.e. with a safety driver in the car. There are five levels of driving automation, ranging from Level 1 (most functions are controlled by the driver but a specific function, like steering or accelerating, can be done automatically by the car) to Level 5 (a fully autonomous system that is equal in ability to a human driver in every driving scenario). Most driverless cars at the moment operate at Level 3, so the car is in control but a human safety driver is present to intervene when required. If fatal mistakes are already happening with a driver at the wheel, at what point can we trust automated systems to completely take over?
Level 3 is also the first point at which full responsibility and legal liability shifts from driver to car, and it is this technicality that has been hardest for car manufacturers to stomach. At the Consumer Electronics Show in January, carmakers boasted driverless tech in driver assistance and safety applications but not in autonomous driving systems – goals which the FT deemed to be ‘less ambitious’ and ‘less controversial’ because they stopped short of handing over full responsibility to the car. The FT article argues that technology was not the limiting factor but rather the regulatory framework, as carmakers were unclear as to what is legal and what isn’t. Regulatory bodies have also recognised this, and in November the Law Commission in the UK launched a three-year consultation to examine likely areas of change to the legal and regulatory framework for automated vehicles. This includes issues of liability, which has been a particular sticking point for many: who does the fault lie with following a driverless car accident – the car manufacturer, the software provider or the driver? Despite legal liability technically passing from the driver to the car at Level 3, Uber has not been held ‘criminally liable’ for the March 2018 accident but the car’s back-up driver potentially faces criminal charges. Until these questions can be answered comprehensively, autonomous cars will not be able to drive on public roads even if they are otherwise fully functional.
One way of getting around regulatory limitations, as we have already seen, is to deploy driverless tech in non-driverless cars. Carmakers have already begun to implement driver assistance services to vehicles, ranging from automatic braking and blind spot detection functions to Tesla’s Autopilot, which can park itself and automatically change lanes. Driverless tech is also being applied to other types of vehicles, such as delivery robots and driverless trucks, whose routes are more predictable and carry less risky cargo and so could be rolled out faster. By introducing driverless technology slowly and in ways that people can interact with on a regular basis, consumer trust can be gradually gained. As the car industry and regulatory bodies start to catch up with self-driving innovation, it won’t be long before driverless cars will be ready for the roads. The only question is, will we be ready for them?
Vivien Zhu is a recent graduate from the University of Oxford, which is where her interest in consulting first developed. She has experience working with consulting firms and in student consultancies and hopes to pass on her knowledge to those who need it.
Image: Pexels
🔴 Interested in consulting?