Driverless cars are coming whether we want them or not, University of Newcastle Professor Kristen Pammer says.
However, Professor Pammer believes they won't be common on the road for decades - at least 50 years.
"We are at the start of a steep learning curve - significant issues need to be addressed," she said.
These include ethics, culpability and accurate navigation, along with infrastructure and computing capability.
Professor Pammer, head of the university's school of psychology, is leading a research project titled, "Connecting humans and self-drive cars: A Safe Vision for Smart Cities".
The federal-funded research will examine how people interact and deal with self-driving cars.
"The role of the human driver will change dramatically," she said.
Everything about road safety - from training novice drivers to designing roads - will need "complete reconsideration".
Trust in self-driving cars will be a big factor.
Autonomous vehicles will enable people to do other activities while driving, but they'll need to be ready and able to take control in an emergency.
"We need to understand how to facilitate communication between autonomous vehicles and human operators," she said.
Otherwise they may be "even more dangerous than human-only drivers".
It has been suggested that the autonomous vehicle industry should mirror the aviation industry, which "recognised a generation ago" that humans were "the weakest link in safety standards".
Great caution was needed when assuming that people could "successfully monitor complex automatic machines and take over if the machine breaks down".
Strict guidelines were established with scientific evidence to account for this problem. Yet automated vehicles were "being tested and operated on public roads with no consideration or understanding" of this.
"The public and policy makers need data and a firm evidence base from which to develop road-safety legislation and make decisions going forward," she said.
"This data should be firmly grounded in human research."
She said autonomous vehicles had the potential to increase road safety "by performing all required operational manoeuvres".
By removing the "human factor" from driving, which represents about 90 per cent of crashes worldwide, road safety would be dramatically improved.
However, public awareness and acceptance of the technology has a long way to go.
"We are just entering what we consider to be the transition period to full autonomy," she said.
"We will be moving through this stage for a very long time while we work through some of these issues."
Nevertheless, she said autonomous vehicles had "the potential to have an enormous impact in reducing death and serious injury on the road".
"It is often quipped that 'driverless cars don't speed, drive drunk or experience fatigue'.
"In Australia alone, there are around 1100 driving-related fatalities every year, with 90 per cent of incidents caused by driver error such as lapses of attention."
From 1991 to 2001, European Union countries reduced road fatalities by 30 per cent with the use of "intelligent safety systems that helped reduce driver errors".
Widespread adoption of robot vehicles is expected to bring benefits such as stabilised traffic flow and vehicle platooning. Platooning relates to a group of vehicles driving together to boost the capacity of roads on an automated highway system.
Platoons decrease the distance between vehicles with "electronic coupling", enabling simultaneous acceleration or braking.
University of Newcastle social psychologist Dr Cassandra Gauld, who is also working on the project, said self-driving vehicles required "ongoing public education".
"While the development of automated vehicle technology is progressing, public trust and acceptance of such technologies can be slow," Dr Gauld said.
"As most people don't yet have first-hand experience of an automated vehicle, we would expect that the public's opinion will improve as they have the chance to experience this advanced technology."
She was involved in a 2019 study that investigated public perception of the Newcastle driverless shuttle bus.
"This was before the bus was trialled, so the public had not yet had a chance to use it first-hand," she said.
"Of the 215 people surveyed, we found that those who thought they would be more inclined to use it, believed it would be a safer form of transport."
These same people believed the technology would "extend public transport to new areas" and be "a novelty to use".
The researchers welcome anyone interested in taking part in driving simulator studies.
"We are always on the lookout for young, experienced, anxious/nervous drivers and anyone who is interested in road safety and autonomous vehicle research," Professor Pammer said.
