Fuzzable
Self-driving cars have been hyped up in the public perception for a few years now, and especially within the technology and automotive industries. When people speak of self-driving cars, it is about how they will revolutionize the transportation industries. There will be no truck driving jobs, no taxi drivers needed, and maybe we won’t even need to buy cars at all anymore. Maybe we’ll just rent them as needed for a lot cheaper than spending tens of thousands of dollars every decade or so.
Without anyone owning cars, cities will no longer need street parking or parking lots. Homes won’t need driveways, apartments won’t need parking garages. The ways we live and move around could be utterly transformed.
Imagine what a utopia it would be to take your hands off the wheel and to simply enjoy the ride while you pull down a screen and watch the latest cinema craze. Wouldn’t that be great? You won’t have to worry about things like pedestrians, traffic rules, and the general safety of the public.
But therein lies the real danger that others are starting to point out. What if autonomous, self-driving cars put convenience at the forefront at the expense of safety. Here’s why this might be the case.
Susceptibility To Failure
The danger of relying completely on machines is that they tend to fail. The more complex the array of sensors required to allow the car to make its determinations, the higher the risk of failure. What separates human drivers from self-driving cars is the ability to make decisions based on intuition when information is incomplete or unavailable.
Whether it’s to slow down as soon as they lose sight of the road, or to when they choose to pull over as soon as they “feel” the car behaving differently.
Because autonomous cars function purely on information from the sensors, when these sensors are unable to collect information, it would also follow that the car will be unable to make a quick decision. Car accident fatalities are high enough. This is an illustration of fatal crashes in Texas, one of which, was also the first fatal accident wherein an autonomous car was involved.
Dependence On Perfect Conditions
If you haven’t noticed, most companies that test self-driving cars only do so in perfect weather conditions. They don’t test their cars in heavily populated areas, nor are these cars tested in difficult weather conditions such as heavy rain and snow, and especially not in zero-visibility conditions. While you can argue that these are simply the first steps toward a milestone and that these can be improved upon later on, the question to ask now is why are we trying to create problems that we’re able to solve without autonomous cars anyway?
Susceptibility To Hacking
Now, let’s assume that self-driving car manufacturers are able to create cars that are able to function in adverse weather, one more danger that they need to be able to address is that of hacking. It is, after all, one of the most prevalent fears in the digital age and that fact that new exploits are being created on a regular basis should already raise questions about autonomous car security.
The benefits of autonomous cars being able to eliminate the human error factor is balanced by things that, at least for now, only human drivers can do. So far, that means that autonomous cars cannot completely eliminate the need for human input at least some of the time. The technology that completely changes how we live has not come yet, and doesn’t look like it will any time soon.
