Driverless car technology has come a long way in recent decades, but it’s far from perfect. In her entry, awarded runner-up in the 2019 UNSW Bragg Student Prize for Science Writing, under the theme “not-so-smart technology“, Year 8 Presbyterian Ladies College student Phoebe Adam drew inspiration from her own family tragedy to explore the history, current state of play and future of driverless cars on our roads.
“Phoebe’s piece successfully punctures some of the hype surrounding the promise of driverless cars,” said science writer Deborah Smith, one the 2019 judges.
Read Phoebe’s full essay below.
2019 UNSW Bragg Student Prize for Science Writing runner-up essay:
by Phoebe Adam (Year 8, Presbyterian Ladies College Croydon, NSW)
On a cold, dark, rainy evening, a teenage girl was walking home alone from a friend’s party. As she walked down the road she came to a steep turn and in the dark confusion, as she began to cross the road, she was suddenly struck by a speeding taxi who hadn’t seen her, resulting in a horrific crash. Though she survived this terrible accident, she was left with a serious head injury from which she never fully recovered. This event happened to my aunt Sarah 20 years ago, when she was 19 years old, only a few years older than I am now.
In the 12 months leading up to October 2018, 1226 people were killed in road-related deaths in Australia. Of these, 277 were under the age of 25 and 14% were pedestrians. The question that occurred to me was: in the years since her accident, have there been any developments in modern technology that could significantly improve road safety and prevent accidents like Sarah’s from happening to other pedestrians in the future?
The biggest factors in accidents caused by driver negligence are due to human error and include distracted driving, speeding, failing to see pedestrians, disobeying traffic signs, disregarding weather and driving under the influence of drugs or alcohol. If human error was removed when it comes to driving, would this prevent traffic accidents?
Some cars have removed the human factor, to an extent, and are known as autonomous vehicles. Although they seem like a recent idea, the notion of driverless cars dates back to the start of the space age in the 1960s, when AI researcher John McCarthy wrote an essay about ‘computer-controlled cars’. In reality, however, how far off is the development of such a vehicle?
In 2004, the US Defence Advanced Research Projects Agency established the Grand Challenge, a competition to develop an autonomous vehicle, in the hope of making the first fully autonomous vehicle a reality. The teams had to make a self-driving car that would complete a 240km track, travelling over hills and dirt tracks, and avoiding the odd cactus, in order to reach the finish line and win US$1 million. When the horn was blown, 15 vehicles sped off, but none ended up reaching the end of the race. In fact, some immediately crashed into trees and fell into gullies.
Although this might seem a failure, it was the start of a new collaborative process and in 2005 when a second race was held, five vehicles got to the finish line. Suddenly the possibility of a driverless car had become a reality.
Since then, many famous manufacturers and technology companies have promised us completely autonomous vehicles. However, in order to become truly ‘hands-free’, there are still many serious technical and safety challenges to overcome. Most cars are already partially autonomous with lane brake assistance and cruise control. There are five levels of automation, but currently available driverless cars, such as Tesla’s, are still limited to specific driving conditions.
Artificial Intelligence has been used to develop a multitude of car sensors that ‘see’ obstacles and other complications on the road in real-time. In theory, driverless cars seem like a great idea, making driving more efficient and safer for both passengers and pedestrians, more convenient for driving kids to school without parents, and also more helpful for disabled or elderly people to stay mobile.
In July 2016, Joshua Brown was driving his Tesla car in autopilot mode, which allows automatic control of the car during highway driving. Unfortunately, the software did not recognise a truck with a connected semi-trailer crossing the highway in front of the car. As a result, the car attempted to continue at full speed under the trailer, resulting in a fatal crash for the driver. There are a growing list of other such cases, a reminder that automated systems are far from perfect.
In truth, the idea of driverless cars has raised many ethical questions, such as: how are taxi drivers and truck drivers going to make a living? What about the risk of your driverless car software getting hijacked and someone else taking control of your car? With automatic brake sensors, there is a real risk that overconfident pedestrians might start playing ‘chicken’ on the road, triggering driverless cars to an emergency stop in the middle of the road to prevent collisions.
The recent development of driverless car technology is very promising but still has many challenges to overcome, as it is a very complex issue involving both human lives and the interaction between humans and machines. In the future, it is likely that driverless vehicles could be our answer to improving road safety and preventing pedestrian collisions such as the one that happened to my aunt Sarah all those years ago. But this has not happened yet, and I think personally, at this point in time, I would not trust a driverless car with my life.
For the 2019 UNSW Bragg Student Prize for Science Writing we asked Australian high school students to enter 800-word essays responding to the 2019 theme of ‘not-so-smart’ technology, identifying and discussing a problem in the world that has yet to be solved by contemporary science and technology.
Read the other entries:
- WINNER: “The science (or lack thereof) behind period and fertility trackers“ by Arwyn Stone (Year 9, Abbotsleigh, NSW), questioning the science behind increasingly popular fertility tracking apps
- RUNNER-UP: “Errare humanum est“ by William Flintoft (Year 10, Melbourne Grammar School, Vic) exploring our increasing reliance on automation and Artifical Intelligence
Author: STEM Contributor
This article was written by a STEM Contributor for Careers with STEM. To learn more, please visit our contact page.