The use of self-driving cars is devoid of the practices of human drivers, such as pressing the accelerator pedal too hard or turning the steering wheel excessively. So, self-driving cars seem to be the solution to all of our problems on the road.
Algorithms can't get drunk, sleepy, or distracted. These are the leading causes of road deaths, being largely responsible for the 1.3 million people killed in road accidents each year.
They are also not prone to driver rage, eating while driving, or tampering with driving entertainment systems. They can move faster and more safely across roads, which reduces congestion.
Computer systems are better than their human counterparts at choosing routes that provide more fuel efficiency, as well as smoother acceleration and stopping. Combined, these environmentally friendly practices save fuel, which ultimately reduces exhaust emissions.
This all looks great. After all, cars, trucks, and buses currently contribute about 30 percent of the pollution-related global warming in the United States. Cars are also a major source of air pollution in cities around the world.
It is easy to see why some see the precision and predictability that results from handing over vehicle control to algorithms as a solution not only to safety issues, but to environmental problems faced by road transport as well. But realizing this reality means overcoming many challenges. Here are three of them.
Energy
It may have been counterintuitive when a 2018 University of Michigan study found that a shift toward self-driving vehicles could increase (rather than decrease) a vehicle's energy requirement.
Skip the podcast and read onMorahakatyTeenage taboos, hosted by Karima Kawah and edited by Mais Baqi.
The episodes
The end of the podcast
The growing thirst for power comes, in part, from all the extra equipment and hardware that self-driving cars need. To navigate our chaotic and complex world, these cars are packed with advanced cameras, lasers and other sensors.
Although this technology helps keep occupants safe, it also means more weight, which increases the amount of energy needed to move the vehicle. And adding the devices to the outside of the car (rather than molding them inside the body) means they suck the air in less efficiently.
The computational ability is a related issue. Modern vehicles monitor everything from oil temperature to engine timing and brake operation. This involves the constant movement of huge amounts of data every hour. The amount of data is expected to be much greater in self-driving cars.
What is the reason? All of these additional sensors produce much larger amounts of data, which self-driving algorithms must comb through, compose, and act upon.
Probes should do this quickly and correctly. This requires superior computing power. But no one can guess how powerful this is. Manufacturers have said little about how much extra power self-driving vehicles need to do all these computational tasks. Nor has she categorically shared information about how they plan to meet this need.
One solution is to equip cars with bigger batteries. But this adds weight, which reduces efficiency. Switching from a petrol engine to an electric one is another solution. The latter is much more efficient at channeling stored energy into moving the wheels. However, some of the raw materials used in electric cars come from the Democratic Republic of the Congo, a country that has long suffered from allegations of child labour. Although minerals can be extracted from the sea, this poses political challenges, as countries have long disputed over seabed mining rights.
Cost
Like many other modern technologies, affordability may be one of the biggest obstacles standing in the way of self-driving cars. These vehicles may be good for the planet, but winning over consumers won't be easy if the price isn't right.
Research shows that when it comes to selling cars, cost is the deciding factor. This is one of the main reasons why electric cars - despite being more environmentally friendly - remain less popular compared to their petrol-powered counterparts.
It's still not really clear what the full cost of a self-driving car will cost, but one thing is for sure: it won't be cheap.
Current estimates put the cost of self-driving technology, excluding the car itself, at between $70,000 and $150,000. This is already three to four times the average price of a new car in the United States.
It is recognized that prices will go down. German engineering firm Robert Bosch recently announced plans to develop "affordable" self-driving technology for a mass market. How this will be achieved remains unclear.
Autonomous sensors alone currently cost more than $10,000. This figure must fall, according to auto industry experts, to less than $200 for self-driving technology to be commercially viable.
So self-driving cars are not destined, at least soon, to be owned by individuals. Instead, we can expect fleets of self-driving taxis called robocabs. The RoboCab idea has another environmental benefit: reducing the number of cars required to serve the population, which ultimately reduces emissions.
It's an idea some companies spend billions to implement. For example, giant Uber has spent more than $1 billion developing self-driving taxis, and has tested the technology in San Francisco and Pittsburgh.
Similar tests have been conducted by Waymo, an Alphabet-backed self-driving technology startup that was born out of Google's self-driving car pilot project.
So, how much will it cost you to use RoboCap? Some studies say the cost will be mere cents per mile, much less than the cost of owning a conventional car. Thus, large segments of society will completely abandon their traditional cars.
Other studies are less optimistic, indicating that the cost of using a robocab can be more than twice the cost of owning a conventional car. One thing is certain: if the price is not right, reducing emissions will not be easy.
Sharing
When it comes to self-driving cars and the environment, the adage “don't care” is especially true. It turns out that the most important aspect of reducing emissions involves not only choosing robocabs over personal cars, but also carpooling.
A University of California Davis study found that without the adoption of “rideshare” — the practice of having more people in one car — car use, and thus emissions, could rise dramatically by 2050.
Why don't we share then? From one angle, people, due to safety and privacy concerns, are not enthusiastic about cars with strangers. Sharing also creates inconvenience by lengthening trips.
For these reasons, ride-sharing has struggled to gain supporters, such as when Uber introduced in 2014 an option for rides with others (it called it UberPool).
There are lower fares for those who opt for shared rides, but the concept remains so unpopular with riders that Uber has reprogrammed its app to "nudge" riders towards sharing.
Ridesharing, for all its benefits, is a practice consumers often avoid (and understandably so). This idea might lead to reduced emissions, but that prospect is unlikely unless consumers can be convinced that its advantages outweigh its disadvantages.
For those developing autonomous vehicles, technological and economic barriers may be easier to overcome than those related to human behaviour.
While computers may make it unnecessary to have humans behind the wheel, how we choose to use computers will ultimately determine the impact of this technology.