Autonomous, or driverless, cars would seem to be our motoring future, but there clearly must be a very long way to go. In the interests of safety, this type of artificial intelligence will need, literally, to be foolproof. Briefly, there are five levels of autonomy. Level 1, the start of it all, is when one element of the driving process is taken over in isolation, using sensors and cameras. This stage has been with us for quite a while now but the driver is still in charge and responsible. Level 2 is pretty much the current situation. Onboard computer technology takes over several functions from the driver multiple data sources; lane keeping and self-parking are examples. Level 3 is described as ‘conditional automation’ – a mode which lets all aspects of driving be done by the car but with the driver’s hands and feet ready to intervene. It’s near, but not here yet. Levels 4 and 5 aren’t even close.
It’s All About Safety
Adherents to the driverless philosophy believe the technology has the potential to improve lives by providing easy and accessible mobility for more people, which makes sense these days, and also enhance the transporting of goods. The primary concern though is that this science has the potential to save lives.
The thinking is that with cars and infrastructure all talking to each other, lives will be saved, depending how the different levels of autonomy work. As things stand now the industry seems to be moving in two directions with automation: those who are implementing driver assist systems at levels 2 and 3 and those developing self-driving systems for level 4.
Although very different, both systems significantly improve safety in different markets. For example, level 2 and 3 are the basic levels of ‘Advanced Driver Assistance Systems’ (ADAS), which are used to aid the driver, who is in control of the vehicle at all times. With more advanced level 4 systems, the vehicle is programmed to take the decisions away from the driver, increasing the ‘hands-off’ experience for consumers. This is the area in which the public needs some convincing and also the area that could cause true autonomy to continue to be a long way off for the simple reason that it has got to be right first time.
It Cannot Be Rushed
Innovation cannot be rushed, especially when it comes to driverless technology. It is generally agreed that creating level 4 self-driving vehicles is a difficult challenge. Experimentation has been ongoing now for more than twelve years. The race to eliminate road fatalities is clearly going to be a marathon, not a sprint.
These early baby steps, rather than giant strides, needed to achieve a safe environment for self-driving vehicles comprise education and collaboration within government and regulatory groups. This will help the industry prepare for connected infrastructure, both inside and outside of the vehicle and that’s the point. It can’t be done piecemeal and all parties need to work together so that one-size fits all. Pushing the envelope is one thing; not sticking it down is another.
For the time being then we, as drivers, are happy with the level 2 autonomy found on the current era of cars. Many of the great used cars we sell feature just this sort of safety equipment and all receive a full AA inspection before they are released for sale.