Why The Autonomous Car Debate is Heating Up Quickly
The guide we recommend: Tonneau covers and their bests
Experimenting with self-driving cars has a history nearly as long as the automobile itself. In 1925, Francis P Houdina dof the Houdina Radio Control Company demonstrated the first recorded driverless car through heavy traffic in New York City, traveling up Broadway and down Fifth Avenue. It was a 1926 Chandler equipped with a transmitting antenna and operated from a trailing second car with a transmitter. The radio signals were able to control small electric motors which led every movement and function of the vehicle. In the decades that followed work in this space has progressed.
The first genuinely autonomous car that was self-sufficient appeared in the 1980s, when a Mercedes-Benz robotic van, designed by Ernst Dickmanns and his team at Bundeswehr University Munich in Germany, was able to achieve a speed of 39 miles per hour (63km per hour) on streets with no traffic.
Today, we are inching ever closer to seeing fully autonomous cars on urban roads all over the world. Driverless cars are now being tested on roads in Australia. Australia recently joined only a handful of countries piloting the mass-scale deployment of autonomous cars.
But experts are raising concerns about the readiness and security of driverless car technologies, especially given the recent fatalities involving self-driven cars that have dealt a major blow to public confidence concerning the technology. However, when we compare the statistics to car accidents resulting in deaths caused by human error in traditional cars, self-driven cars could prove to be the most significant automobile safety initiatives in history. In Australia alone, there has been over 190,000 fatalities sinc accurate car record keeping commenced in 1925.
As of the moment, there are as many people who are willing to embrace the technology soon as there are those who’d rather not see autonomous vehicles in our lifetime. According to a recent survey, over 55% of respondents were willing to embrace the technology if it offered irrefutable proof of the safety of autonomous cars. However, to obtain evidence, driverless cars must hit the road live under real-life traffic circumstances. And this is something that many people are deeply wary of as pedestrians and road users might be put at risk when these imperfect self-driven cars roam our streets.
Toll operator Eastlink recently surveyed of over 18,000 Australian motorists which showed the majority of drivers do not want autonomous cars on the road. Men were more likely to embrace the technology with a recorded 29% in favour, whereas only 17% of female respondents supportive.
According to Professor Sandeep Gopalan, a law expert at Deakin University, level 5 driverless cars (those with the highest level of sophistication) are still unfit to be tested in urban environments.
“Driverless cars are not as close to being on our roads as the initial marketing hype led us to believe,” Professor Gopalan said during an interview with The New Daily.
“All of the evidence including the Arizona crash report and other research shows that these cars still have a long way to go” he added.
Professor Damon Honnery, deputy head of the department of mechanical and aerospace engineering at Monash University, echoed similar sentiments in an interview with The New Daily.
“For minor degrees of autonomous vehicles, that’s not an issue, and it’s already happening at some universities. But if you’re talking about cars that can drive you to work, there are risks involved with this.”
“When it comes to mixed traffic environments, the vehicles have to respond very quickly to this. And if you’re dealing with complex road structures, this will definitely pose challenges.”
However, the push for the perfection of self-driving innovations is being powered by state sponsors across the globe. In Australia, Victoria, New South Wales, and Queensland have all been offered up as rare testing grounds for autonomous driving technology. However, authorities down under have been more conservative towards the test run of the technology, preferring to tread cautiously with trials not yet at more advanced stages as they are overseas.
But the National Transport Commission spokesman Ron Grasso disclosed that the Transport and Infrastructure Council is poised to see the introduction of autonomous cars in Australia come 2020.
“Our main aim is all about making sure that when these vehicles do come into the market, that they operate as safely as possible,” Mr. Grasso informed The New Daily.
Driverless cars can be placed into six different levels, from 0 to 5, depending on their level of sophistication. The levels also indicate the extent to which the car requires driver assistance. Level 1 leaves a whole lot to human control with very little autonomous vehicle technology. Level 2 and 3 are partially automated, while level 4 has high automation. Level 5 is full automation, where no human intervention is required.
The level of autonomy of a driverless vehicle also depends on the number of sensors it’s fitted with. All autonomous cars deploy sensors and cameras to mine data about a car’s surroundings. The onboard software that processes and acts on the collected data is also a critical determinant of the vehicle’s level of sophistication. These factors distinguish between various models of autonomous cars. Companies that have had self-driving software and sensors in the works for years are significantly ahead of their competitors. Chip-maker Nvidia is a name that is leading the charge in developing hardware and software that could bring about the future in driving. The company’s origins have been in developing chips for navigating 3D worlds in video games – The computing problems that Nvidia was concerned with resolving, turned out to be similar to the issues involved in artificial intelligence involved in self-driven cars.
Key Challenges Fueling the Scepticism Over the Safety of Self-Driven Cars
- Inherent redundancy: Aircraft are quite safe because their computing systems operate with multiple levels of redundancies, switching between different sensors, computers, engines and other components in the event of a failure of any particular part. This multi-layered redundancy, which has so far eluded self-driven car producers, is a lynchpin that can ensure that all the function of the autonomous vehicle work at all times without fail.
- Cybersecurity threats: The possibility of cybersecurity threats is always lurking around, with the frequency of cyber-attacks increasing by the day. In 2015, hackers were able to control a Jeep Cherokee remotely, forcing the car company to recall over 1.3 million cars. Cyber threats will remain the greatest cause for concern regarding the safety of autonomous vehicles in the near future.
- Lack of regulations on disclosures: Experts believe that there would be a more positive public perception of the self-driving technology if manufacturers testing autonomous vehicles are obliged to reveal a lot more information concerning tests.
“It’s not just about crash data. We need to know if the cars are picking up certain objects, not slowing in time or requiring the driver to intervene before a crash” Professor Sandeep Gopalan told The New Daily
“Regulators need to ask developers to reveal their results in certain environments before they are allowed to test on the roads” he added.
Autonomous driving technology may be on the verge of becoming more mainstream; however, there are a few obstacles to overcome before we are using this technology in our daily life.
Only a handful of Countries and Cities in the world have been gracious enough to accommodate autonomous vehicle test runs on their streets. Australia, being one of those countries, does make an ideal test ground. Australia is home to many long, straight and deserted highways, where it’s possible to travel hundreds of kilometres and only pass a handful of cars and perhaps some wildlife.
Studies show that human error accounts for over 90% of road accidents, and self-driven cars are designed to eliminate these human errors. Although they’re not yet perfect, they’re far less prone to making fatal driving mistakes than humans.
We all want the end result, it is just a matter of how we navigate to its realisation. Holding back on testing in real-life scenarios only obstructs the possibility of reducing major motor accidents, including those that result in death caused by human error.