The film Minority Report paints a surreal and mind-blowing picture of life in 2054. It features a wide array of awe-inspiring technology including: insect-sized robots, electronic paper, jetpacks, precognitive technology and driverless vehicles.
Self-driving vehicles are a staple of the science fiction genre. Films such as I, Robot and Total Recall, set in the distant future, also feature the autonomous technology. Even KITT from cult television series Knight Rider is one of these wonders, although with a bit more sass than you’d expect from a car.
The self-driving car has far surpassed its origins as a mainstay of science fiction to become reality for modern society. Swedish carmaker Volvo’s ‘Drive Me’ initiative aims to have driverless vehicles on Gothenberg roads by 2017, and Australia isn’t far behind.
This November, the South Australian Government hosts Australia’s first driverless (or automated) vehicle trial. Through its Australian Driverless Vehicle Initiative, ARRB is partnering with the South Australian Government, and local and international experts, including Volvo, Telstra and Bosch, to run the first driverless vehicle demonstration trials in the country and the Southern Hemisphere. “The South Australian trial is one of many we have planned for around the country,” says Gerard Waldron, ARRB Managing Director.
“In lots of ways, driverless vehicles is a concept that started to emerge years ago,” says Mr. Waldron. “I started talking about driverless vehicles a few years ago and people were saying: ‘That’s not going to happen.’
“My prediction is that we’re going to have people sharing the freeway with other drivers with their hands off the wheel in the next three years,” he asserts. “We certainly don’t want to see Australia miss out on the benefits and on the other hand, we don’t want to be on the bleeding edge… We decided it was time to press the matter.”
Mr. Waldron says that with companies such as Volvo demonstrating the capabilities of driverless vehicles internationally, Australia needs to keep up. This first trial is the next step in overcoming the initial challenges on the road to self-driving vehicles.
“What we’re trying to do is get everyone on the same page with the fundamentals on the timing and take-up of this technology,” he says. “We want [road agencies and government] to have an understanding of these opportunities and how pressing it is.”
ARRB’s Driverless Vehicle Initiative aims to harmonise Australian efforts to introduce the autonomous vehicle technology.
“The call to action here is that we want to get the conversation going and get anyone with a hand of the lever of policy to see the benefits,” he says. “[Policy or legislation-makers and road agencies] all have a place at the table and the idea is to not duplicate resources and not do things in competition.”
The trial vehicle is a Volvo XC90, which Mr. Waldron says has some of the technology for driverless capability already built into its system. It just needs to be tapped into.
Policy is one of the major hurdles to overcome in introducing self-driving cars. The driverless technology still needs to be approved for use on Australian roads, which is the motivation behind the trial. Mr. Waldron says regulations around it need to be clear, as there are many different factors to take into account. “It’s not going to be one approval,” he says.
Mr. Waldron explains that many policy makers in government will be apprehensive about such a dramatic technological change, especially one that requires a “leap of faith”.
Communicating the benefits of this technology to those individuals in government, and the nation as a whole, is a significant step Mr. Waldron is pushing for.
“This has such huge economic benefits for the country and for road safety. That “market pull through” is going to be really quite rapid,” he says.
The way roads are priced, vehicle registration fees and even car ownership may change immensely with the introduction of automated vehicles. Mr. Waldron suggests that once they become available, commuters may be able to use a service, like Uber, that can take them to work each day. As a result, the commuter won’t have a car sitting in a car park, unused for the majority of any working day.
“Once we give into that concept I don’t think people will own cars,” says Mr. Waldron. He likens the idea to a mobile phone plan with a provider – you top up each month or sign a contract to receive your service.
He suggests that some road users may have a predisposition to automated vehicles, particularly as it may reduce travel times, and allow them to use their travel time more productively.
“When you think about the current road model, it’s fairly dumb,” states Mr. Waldron. A main or arterial road will be fully utilised at peak commute times, but he estimates its full potential is not realised in roughly 22 hours of the day.
“One of the things that concerns me is that we’re building roads and major infrastructure projects, but we’re not allowing or providing for driverless vehicles in the construction or design.”
With automated vehicles comes connected vehicles – those that can communicate with one another and road signs adapted to function within an intelligent transport system (ITS). This can possibly lead to fewer static road signs, reduced maintenance costs and even less congestion on a road, particularly in how these assets are managed. “That all means it’s possible to get more out of the infrastructure,” he adds.
“Some of the things we’re doing now for human drivers will be quite acceptable for driverless vehicles,” surmises Mr. Waldron. Tactile edge markings (the gridded indents on roads), for instance, already tell a driver if they’re veering over the outer edge of the road. It is also something that is easily detectable and useful for driverless vehicles by way of an accelerometer.
“In urban environments we want driverless vehicles to be connected vehicles,” says Mr. Waldron.
Connected vehicles also have the potential to reduce road accidents. Two oncoming vehicles, for instance, can coordinate to prevent a crash before it happens.
Like Mr. Waldron, Kamal Weeratunga, Main Roads Western Australia’s Network Operations Planning Manager, is already mindful of the implications of driverless vehicles will have on how Australian roads are built and operating.
“The pace of development of automated vehicles is rapidly increasing year by year,” he says. The past three to five years has seen an uptake in the technology, he explains, particularly as many car manufacturers are looking into driverless vehicles technology very closely. “It’s led by the private sector and it’ll all go ahead whether we like it or not,” he adds.
Mr. Weeratunga spearheaded two reports on the automated vehicle technology – Automated Vehicles: Are We Ready? and Connected Vehicles: Are We Ready? The reports were released this past January and June respectively, and coincided with the road authority’s new strategic direction initiative, Keeping WA Moving, the vision of which is to provide a safe, reliable and sustainable road-based transport system.
“Within that strategy we’re looking at the new possibilities and opportunities that are transforming the transport landscape. The emergence of automated and connected vehicles is a key part of the changing external landscape,” he explains. “[The reports are] just the initial step to raise the awareness within the organisation on the oncoming changes in the future.”
Mr. Weeratunga is adamant that road agencies need to be thinking about the influence driverless vehicles will have on roads. “The impact on our own business is obviously huge,” he says. “Take the example of road planning and design. Normally we design roads for 40 years and bridges for up to 100 years. In this context we need to take into account that the automated vehicles will be operational on road networks within the next 20 years.
“To think that in about 20 years’ time there will be substantial numbers of automated vehicles. These will be within the design life of roads, which we are planning or building today,” says Mr. Weeratunga. “These roads will need to be able to support the new automated vehicles.
“A fundamental question road agencies are facing at the moment is: ‘What does this mean for planning and investment decisions we are making today?’
“From a road agency point of view, it’s obviously provided a huge challenge for us, but it’s also providing tremendous opportunities to solving some of the intractable problems of transport,” he says. “Automated vehicles are likely to go a long way to solve the problems concerning road safety, congestion and energy emissions.”
Mr. Weeratunga suggests automated vehicles can drive very close to each other in synchronisation at a uniform speed. “This means if you are looking at a spot on a traffic lane, more vehicles will travel through that spot in a minute or an hour. Some commentators say this could more than double the capacity of existing roads,” he says.
“Automated vehicles will travel with more precision and control,” he adds. “With no human control, you take human error out of the equation. Therefore we don’t need to have large safety margins in our road designs.
“What this means is road designs for automated vehicles are likely to be compact with narrow lanes. We may be able to accommodate more lanes with the same bitumen width or convert the surplus space for other uses such as for pedestrians, bicycles or streetscaping.”
With precision travel, the vehicles will travel in the same path within the lane over and over. This means the distribution of the load and deterioration of road pavement will be different compared to current pavements he explains, which will impact on how pavements are designed and built for automated vehicles.
Mr. Weeratunga says that one of the immediate problems they are grappling with is modelling a traffic stream with a mix of current generation vehicles and automated vehicles with different operational characteristics. “There are a lot of unknowns,” he says. “Current traffic models are not built to handle this scenario.”
Unlike Mr. Waldron, Mr. Weeratunga anticipates that automated vehicles are likely to be commercially available between 2020 and 2025. By 2035, he believes the traffic stream will have a significant proportion of driverless vehicles, followed by a saturation period.
“When you have automated vehicles operating alongside and with traditional vehicles, then, from a road authority point of view, that’s the most challenging period because we have to cater to vehicles with fundamentally different operating characteristics and a new complexity of traffic streams.”
He says there may need to be dedicated lanes or areas for driverless vehicles, but roads will still need to cater to both during the transition period.
To start thinking about these challenges, Main Roads WA is engaging with Austroads and ARRB in the Driverless Vehicles Initiative.
Main Roads may also investigate a driverless vehicle trial in the state. Mr. Weeratunga says WA has unique challenges such as the vast expanse of the state, long distances, communications coverage in remote areas and large configurations of freight vehicles associated with the mining industry. Main Roads WA sees it as an opportunity to trial the technology with freight transport.
“But we can’t charge ahead on our own. We have to synchronise our actions with Austroads and other agencies. On the other hand we can’t afford to lose any time either,” he says.
“We already have automated trucks operating within some mine sites in WA, albeit within a controlled environment and some real-time oversight and control from their control centres in Perth.
“With decreasing iron-ore prices and associated cost pressures, mining and freight may be one of the first industries to embrace driverless technology with its significant potential to bring down operating costs and also improve safety,” he says.
“Whichever way you look at it, the automated vehicles are going to fundamentally change the transport landscape,” he adds. “It’s a once-in-a-century transformation. We need to be prepared for that and be prepared soon.”
The human factor with the introduction of driverless vehicles is one change Mike Regan asserts we need to be mindful of and prepare for as well.
Prof. Regan, Chief Scientist – Human Factors at ARRB, presented his research into the human element of automated vehicles at the inaugural Australasian Road Safety Conference (ARSC2015) at the Gold Coast in October.
He says that there will certainly be a whole range of benefits with the introduction of driverless vehicles. “It is estimated that about 90 per cent of the crashes that drivers have are due to human error,” he says. “Many of these errors stem from inadequate road design. However, if you can automate those driving tasks most prone to human error, you can eliminate many of those crashes.”
Like Mr. Waldron and Mr. Weeratunga, he asserts that there may be fewer road accidents, increased mobility on the roads, less congestion and reduced car emissions with the introduction of driverless vehicles. However, he explains that there are a number of issues that come with the technology, particularly where the human element is involved.
“I think the main challenges from a human factors perspective relate more to what’s going to happen in the interim period,” he says. The interim period Prof. Regan refers to is the point in time when driverless cars and traditional cars share the road.
“Trust is an issue that’s always coming up from a human factor perspective,” he says. “Some people are open to it and will trust driverless vehicles, and other people won’t.
“When we transition to driverless vehicles, there are going to be times where either the car fails or where the system can’t operate autonomously for some reason – the driver has to take control.” A human passenger, for instance, could be busy reading, working or even sleeping when the need for them to retake control of the vehicles arises.
“What we need to understand is how prepared people will be, how to support safe manual takeover, and what to do with the car if safe takeover is not possible,” asserts Prof. Regan.
He raises the point that driving skills may even diminish as society builds a reliance on the technology.
“What’s going to happen when people are in automated vehicles and interact with people in manual vehicles?”
He suspects that those in manually driven vehicles may even abuse the driverless vehicles to an extent.
They are designed to drive at a uniform speed and keep a certain distance from the car ahead of it. Prof. Regan says a human driver may take advantage of the situation and merge into a lane, cutting off a driverless vehicle (which will automatically brake), potentially causing an accident. “[People] might start abusing automated vehicles in ways we don’t understand yet,” he adds.
“I think the message is that there are many human factor challenges, but we’ve got to focus on the solutions, not the problems,” he says. “Public acceptance of driverless vehicles is going to be critical.”